diff --git a/.gitattributes b/.gitattributes index 52f73b290c1ec27223dcc680dd8cb29ae83e1232..31472ce4563b33cee3f7cd88d633dfaa5538c416 100644 --- a/.gitattributes +++ b/.gitattributes @@ -109,3 +109,16 @@ baseline/outputs/gears_from_scratch/figures/evaluation_metrics_use_existing_test baseline/outputs/gears_from_scratch/figures/training_history_use_existing_test.png filter=lfs diff=lfs merge=lfs -text baseline/outputs/gears_from_scratch/pred_gears_from_scratch_use_existing_test.h5ad filter=lfs diff=lfs merge=lfs -text baseline/outputs/gears_from_scratch/real_gears_from_scratch_use_existing_test.h5ad filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_A_gene_mean_calibration_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_B_MA_plot_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_predicted_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_truth_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_F_perturb_corr_truth_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_G_perturb_corr_clustered_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_H_cell_PCA_truth_pred_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_J_metric_panels_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/bio_K_topvar_gene_scatters_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/evaluation_metrics_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/figures/training_history_use_existing_test.png filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/pred_cellot_ot_baseline_use_existing_test.h5ad filter=lfs diff=lfs merge=lfs -text +baseline/outputs/cellot_ot_baseline/real_cellot_ot_baseline_use_existing_test.h5ad filter=lfs diff=lfs merge=lfs -text diff --git a/baseline/CellOT.ipynb b/baseline/CellOT.ipynb new file mode 100644 index 0000000000000000000000000000000000000000..7c9cc9e8985a9701708a54f84e7e43a63148f1f9 --- /dev/null +++ b/baseline/CellOT.ipynb @@ -0,0 +1,2394 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "790bd077", + "metadata": {}, + "source": [ + "# CellOT-Style OT Baseline\n", + "\n", + "This notebook implements a **CellOT-inspired optimal-transport baseline** for the project data format.\n", + "\n", + "Core principle borrowed from CellOT:\n", + "\n", + "- perturbation prediction is treated as a **map between unpaired control and perturbed single-cell distributions**\n", + "- optimal transport is used to align control and perturbed cells without requiring one-to-one paired observations\n", + "- a parameterized neural transport map is learned so it can be applied to **unseen control cells**\n", + "\n", + "Practical simplification in this notebook:\n", + "\n", + "- instead of reproducing the official CellOT dual-ICNN training pipeline exactly, we build a stable **direct transport-map baseline**\n", + "- mini-batch Sinkhorn OT provides barycentric targets\n", + "- a conditional neural map learns `control -> transported state` in latent space\n", + "\n", + "This makes the notebook well suited as a thesis OT baseline while staying close to the CellOT motivation.\n" + ] + }, + { + "cell_type": "markdown", + "id": "86e9cb04", + "metadata": {}, + "source": [ + "## References\n", + "\n", + "Primary sources used for the design:\n", + "\n", + "- CellOT paper: Bunne et al., *Learning single-cell perturbation responses using neural optimal transport*, Nature Methods 2023\n", + "- Official CellOT repository: `bunnech/cellot`\n", + "\n", + "Key idea from the paper:\n", + "\n", + "learn a parameterized transport map between **unpaired** control and perturbed single-cell distributions, so predictions can be made for **new unseen control cells**.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "f9f00bff", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.\u001b[0m\u001b[33m\n", + "\u001b[0mProject root: /root/final\n", + "Input h5ad: /root/final/data/processed/scPerturb/prediction_ready/replogle_k562_essential_single_gene_prediction_ready_full.h5ad\n", + "Output dir: /root/final/baseline/outputs/cellot_ot_baseline\n", + "Device: cuda\n" + ] + } + ], + "source": [ + "! pip install -q anndata numpy pandas scipy torch matplotlib\n", + "from pathlib import Path\n", + "from copy import deepcopy\n", + "import json\n", + "import random\n", + "\n", + "import anndata as ad\n", + "import matplotlib as mpl\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "import pandas as pd\n", + "import scipy.sparse as sp\n", + "from scipy import stats\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "\n", + "\n", + "def find_project_root(start: Path | None = None) -> Path:\n", + " start = (start or Path.cwd()).resolve()\n", + " for path in [start, *start.parents]:\n", + " if (path / \"src\" / \"data\").exists() and (path / \"baseline\").exists():\n", + " return path\n", + " return start\n", + "\n", + "\n", + "PROJECT_ROOT = find_project_root()\n", + "DATASET_NAME = \"ReplogleWeissman2022_K562_essential\"\n", + "METHOD_NAME = \"cellot_ot_baseline\"\n", + "RANDOM_SEED = 42\n", + "\n", + "SPLIT_STRATEGY = \"use_existing\" # choices: \"use_existing\", \"seen_cell_split\"\n", + "TRAIN_FRAC = 0.80\n", + "VAL_FRAC = 0.10\n", + "EVAL_SPLIT = \"test\"\n", + "CONTROL_SPLIT = \"train\"\n", + "TRAIN_SPLIT = \"train\"\n", + "VAL_SPLIT = \"val\"\n", + "\n", + "USE_FEATURE_GENES = True\n", + "FEATURE_GENE_COLUMNS = [\"is_feature_gene\", \"highly_variable\"]\n", + "\n", + "MAX_TRAIN_CONDITIONS = None\n", + "MAX_EVAL_GENES = None\n", + "MAX_EVAL_CELLS_PER_GENE = None\n", + "\n", + "LATENT_DIM = 64\n", + "HIDDEN_DIM = 256\n", + "CONDITION_DIM = 128\n", + "DROPOUT = 0.10\n", + "NUM_TRANSPORT_BLOCKS = 4\n", + "\n", + "BATCH_SIZE = 128\n", + "EPOCHS = 20\n", + "STEPS_PER_EPOCH = 200\n", + "VALIDATION_STEPS = 32\n", + "LEARNING_RATE = 1e-3\n", + "WEIGHT_DECAY = 1e-5\n", + "GRAD_CLIP_NORM = 1.0\n", + "\n", + "RECON_WEIGHT = 1.0\n", + "MAP_WEIGHT = 1.0\n", + "LATENT_MMD_WEIGHT = 0.1\n", + "EXPRESSION_MMD_WEIGHT = 0.1\n", + "MEAN_WEIGHT = 1.0\n", + "DISPLACEMENT_REG = 0.01\n", + "\n", + "SINKHORN_EPSILON = 0.5\n", + "SINKHORN_ITERS = 30\n", + "PRED_BATCH_SIZE = 256\n", + "\n", + "INPUT_CANDIDATES = [\n", + " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready_full.h5ad\",\n", + " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready.h5ad\",\n", + " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready_fast_debug.h5ad\",\n", + "]\n", + "INPUT_PATH = next((path for path in INPUT_CANDIDATES if path.exists()), None)\n", + "if INPUT_PATH is None:\n", + " raise FileNotFoundError(\"Could not find a prediction-ready Replogle h5ad. Checked:\\n\" + \"\\n\".join(map(str, INPUT_CANDIDATES)))\n", + "\n", + "OUTPUT_DIR = PROJECT_ROOT / \"baseline\" / \"outputs\" / METHOD_NAME\n", + "OUTPUT_DIR.mkdir(parents=True, exist_ok=True)\n", + "MODEL_PATH = OUTPUT_DIR / \"cellot_ot_baseline_model.pt\"\n", + "\n", + "DEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", + "\n", + "np.random.seed(RANDOM_SEED)\n", + "random.seed(RANDOM_SEED)\n", + "torch.manual_seed(RANDOM_SEED)\n", + "if torch.cuda.is_available():\n", + " torch.cuda.manual_seed_all(RANDOM_SEED)\n", + "if torch.backends.cudnn.is_available():\n", + " torch.backends.cudnn.deterministic = True\n", + " torch.backends.cudnn.benchmark = False\n", + "\n", + "print(f\"Project root: {PROJECT_ROOT}\")\n", + "print(f\"Input h5ad: {INPUT_PATH}\")\n", + "print(f\"Output dir: {OUTPUT_DIR}\")\n", + "print(f\"Device: {DEVICE}\")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "bdcfaee2", + "metadata": {}, + "outputs": [], + "source": [ + "SCIENCE_COLORS = {\n", + " \"blue\": \"#0072B2\",\n", + " \"orange\": \"#E69F00\",\n", + " \"green\": \"#009E73\",\n", + " \"magenta\": \"#CC79A7\",\n", + " \"dark\": \"#333333\",\n", + " \"gray\": \"#999999\",\n", + "}\n", + "\n", + "\n", + "def setup_science_style() -> None:\n", + " mpl.rcParams.update(\n", + " {\n", + " \"figure.dpi\": 110,\n", + " \"savefig.dpi\": 300,\n", + " \"font.family\": \"sans-serif\",\n", + " \"font.sans-serif\": [\"DejaVu Sans\", \"Helvetica\", \"Arial\"],\n", + " \"font.size\": 9,\n", + " \"axes.titlesize\": 10,\n", + " \"axes.labelsize\": 9,\n", + " \"axes.linewidth\": 0.6,\n", + " \"axes.edgecolor\": SCIENCE_COLORS[\"dark\"],\n", + " \"axes.labelcolor\": SCIENCE_COLORS[\"dark\"],\n", + " \"xtick.labelsize\": 8,\n", + " \"ytick.labelsize\": 8,\n", + " \"xtick.major.width\": 0.6,\n", + " \"ytick.major.width\": 0.6,\n", + " \"legend.frameon\": False,\n", + " \"legend.fontsize\": 8,\n", + " \"axes.spines.top\": False,\n", + " \"axes.spines.right\": False,\n", + " \"figure.facecolor\": \"white\",\n", + " \"axes.facecolor\": \"white\",\n", + " \"grid.alpha\": 0.3,\n", + " \"grid.linewidth\": 0.4,\n", + " \"lines.linewidth\": 1.25,\n", + " \"pdf.fonttype\": 42,\n", + " \"ps.fonttype\": 42,\n", + " }\n", + " )\n", + "\n", + "\n", + "FIGURE_DIR = OUTPUT_DIR / \"figures\"\n", + "FIGURE_DIR.mkdir(parents=True, exist_ok=True)\n", + "setup_science_style()\n", + "\n", + "\n", + "def plot_training_history(history: pd.DataFrame, tag: str) -> None:\n", + " setup_science_style()\n", + " h = history.copy()\n", + " if h.empty:\n", + " return\n", + " fig, axes = plt.subplots(1, 2, figsize=(7.0, 2.45), constrained_layout=True)\n", + " e = h[\"epoch\"].to_numpy()\n", + " ax = axes[0]\n", + " if \"train_total\" in h.columns:\n", + " ax.plot(\n", + " e,\n", + " h[\"train_total\"],\n", + " \"o-\",\n", + " ms=3,\n", + " lw=1.1,\n", + " color=SCIENCE_COLORS[\"blue\"],\n", + " label=\"Train (total)\",\n", + " )\n", + " if \"val_total\" in h.columns and h[\"val_total\"].notna().any():\n", + " ax.plot(\n", + " e,\n", + " h[\"val_total\"],\n", + " \"s-\",\n", + " ms=3,\n", + " lw=1.1,\n", + " color=SCIENCE_COLORS[\"orange\"],\n", + " label=\"Validation (total)\",\n", + " )\n", + " best_i = int(h[\"val_total\"].idxmin())\n", + " best_ep = float(h.loc[best_i, \"epoch\"])\n", + " best_v = float(h.loc[best_i, \"val_total\"])\n", + " ax.scatter(\n", + " [best_ep],\n", + " [best_v],\n", + " s=55,\n", + " zorder=5,\n", + " facecolors=\"none\",\n", + " edgecolors=SCIENCE_COLORS[\"dark\"],\n", + " linewidths=1.1,\n", + " label=\"Best val\",\n", + " )\n", + " ax.set_xlabel(\"Epoch\")\n", + " ax.set_ylabel(\"Loss\")\n", + " ax.legend(loc=\"upper right\", handlelength=2.0)\n", + "\n", + " ax2 = axes[1]\n", + " for col, lab, c in (\n", + " (\"train_map\", \"Train transport map\", SCIENCE_COLORS[\"green\"]),\n", + " (\"train_recon\", \"Train reconstruction\", SCIENCE_COLORS[\"magenta\"]),\n", + " (\"train_latent_mmd\", \"Train latent MMD\", \"#D55E00\"),\n", + " (\"train_expression_mmd\", \"Train expression MMD\", \"#6A3D9A\"),\n", + " ):\n", + " if col in h.columns:\n", + " ax2.plot(e, h[col], \"o-\", ms=2.5, lw=1.0, color=c, label=lab)\n", + " ax2.set_xlabel(\"Epoch\")\n", + " ax2.set_ylabel(\"Component loss\")\n", + " ax2.set_yscale(\"symlog\", linthresh=1e-4)\n", + " ax2.legend(loc=\"upper right\", handlelength=2.0)\n", + "\n", + " fig.suptitle(\n", + " \"Training dynamics — CellOT-style OT baseline\",\n", + " fontsize=10.5,\n", + " color=SCIENCE_COLORS[\"dark\"],\n", + " y=1.02,\n", + " )\n", + " for ext in (\"png\", \"pdf\"):\n", + " fig.savefig(FIGURE_DIR / f\"training_history_{tag}.{ext}\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " plt.show()\n", + "\n", + "\n", + "def plot_evaluation_metrics(\n", + " metrics: pd.DataFrame,\n", + " tag: str,\n", + " train_genes: list[str] | None = None,\n", + ") -> None:\n", + " setup_science_style()\n", + " if metrics.empty:\n", + " return\n", + " m = metrics.copy()\n", + " if train_genes is not None:\n", + " seen = set(train_genes)\n", + " m[\"cohort\"] = np.where(m[\"perturbation_gene\"].isin(seen), \"Seen at train\", \"Unseen at train\")\n", + " else:\n", + " m[\"cohort\"] = \"All\"\n", + "\n", + " fig = plt.figure(figsize=(7.2, 5.9), constrained_layout=True)\n", + " gs = fig.add_gridspec(2, 2, height_ratios=[1.05, 1.0], hspace=0.28, wspace=0.28)\n", + "\n", + " ax1 = fig.add_subplot(gs[0, 0])\n", + " if m[\"cohort\"].nunique() > 1:\n", + " palette = {\"Seen at train\": SCIENCE_COLORS[\"blue\"], \"Unseen at train\": SCIENCE_COLORS[\"orange\"]}\n", + " for lab, g in m.groupby(\"cohort\"):\n", + " s = 14 + np.clip(g[\"n_cells\"].to_numpy(dtype=float), 0, 120) * 0.12\n", + " ax1.scatter(\n", + " g[\"mse_mean\"],\n", + " g[\"pearson_mean\"],\n", + " s=s,\n", + " alpha=0.78,\n", + " edgecolors=\"none\",\n", + " c=palette.get(lab, SCIENCE_COLORS[\"gray\"]),\n", + " label=lab,\n", + " )\n", + " ax1.legend(loc=\"lower left\")\n", + " else:\n", + " ax1.scatter(\n", + " m[\"mse_mean\"],\n", + " m[\"pearson_mean\"],\n", + " s=22,\n", + " alpha=0.75,\n", + " edgecolors=\"none\",\n", + " c=SCIENCE_COLORS[\"blue\"],\n", + " )\n", + " ax1.set_xlabel(\"MSE (predicted vs true, gene means)\")\n", + " ax1.set_ylabel(\"Pearson r (gene means)\")\n", + " ax1.set_xscale(\"log\")\n", + " ax1.text(0.02, 0.98, \"a\", transform=ax1.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " ax2 = fig.add_subplot(gs[0, 1])\n", + " m_sorted = m.sort_values(\"pearson_mean\", ascending=True)\n", + " max_rows = 36\n", + " if len(m_sorted) > max_rows:\n", + " head_n = max_rows // 2\n", + " plot_df = pd.concat([m_sorted.head(head_n), m_sorted.tail(head_n)], axis=0)\n", + " else:\n", + " plot_df = m_sorted\n", + " y = np.arange(len(plot_df))\n", + " ax2.barh(y, plot_df[\"pearson_mean\"], height=0.68, color=SCIENCE_COLORS[\"blue\"], alpha=0.88)\n", + " ax2.set_yticks(y)\n", + " ax2.set_yticklabels(plot_df[\"perturbation_gene\"], fontsize=5.8)\n", + " ax2.set_xlabel(\"Pearson r (gene means)\")\n", + " ax2.set_xlim(0, 1.02)\n", + " ax2.text(0.02, 0.98, \"b\", transform=ax2.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " ax3 = fig.add_subplot(gs[1, 0])\n", + " d = m[\"delta_l2\"].dropna().to_numpy()\n", + " if d.size:\n", + " nbin = int(np.clip(len(d) // 2, 10, 32))\n", + " ax3.hist(\n", + " d,\n", + " bins=nbin,\n", + " color=SCIENCE_COLORS[\"green\"],\n", + " alpha=0.88,\n", + " edgecolor=\"white\",\n", + " linewidth=0.35,\n", + " )\n", + " ax3.set_xlabel(\"L2 distance (pred delta vs true delta)\")\n", + " ax3.set_ylabel(\"Perturbations\")\n", + " ax3.text(0.02, 0.98, \"c\", transform=ax3.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " ax4 = fig.add_subplot(gs[1, 1])\n", + " ax4.scatter(\n", + " m[\"pearson_mean\"],\n", + " m[\"pearson_delta\"],\n", + " s=20,\n", + " c=SCIENCE_COLORS[\"orange\"],\n", + " alpha=0.72,\n", + " edgecolors=\"none\",\n", + " )\n", + " pm = m[\"pearson_mean\"].to_numpy(dtype=float)\n", + " pd_ = m[\"pearson_delta\"].to_numpy(dtype=float)\n", + " lo = float(np.nanmin([np.nanmin(pm), np.nanmin(pd_), 0.0]))\n", + " hi = float(np.nanmax([np.nanmax(pm), np.nanmax(pd_), 1.0]))\n", + " pad = 0.03 * (hi - lo + 1e-6)\n", + " ax4.plot([lo, hi], [lo, hi], ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.9, label=\"Identity\")\n", + " ax4.set_xlim(lo - pad, hi + pad)\n", + " ax4.set_ylim(lo - pad, hi + pad)\n", + " ax4.set_xlabel(\"Pearson r (means)\")\n", + " ax4.set_ylabel(\"Pearson r (delta vs control)\")\n", + " ax4.legend(loc=\"lower right\")\n", + " ax4.text(0.02, 0.98, \"d\", transform=ax4.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " fig.suptitle(\n", + " \"Held-out evaluation — per-perturbation metrics\",\n", + " fontsize=10.5,\n", + " color=SCIENCE_COLORS[\"dark\"],\n", + " )\n", + " for ext in (\"png\", \"pdf\"):\n", + " fig.savefig(FIGURE_DIR / f\"evaluation_metrics_{tag}.{ext}\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " plt.show()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "922806fd", + "metadata": {}, + "outputs": [], + "source": [ + "def normalize_bool(series: pd.Series) -> pd.Series:\n", + " if pd.api.types.is_bool_dtype(series):\n", + " return series.fillna(False).astype(bool)\n", + " text = series.astype(str).str.lower().str.strip()\n", + " return text.isin([\"true\", \"1\", \"yes\", \"y\"])\n", + "\n", + "\n", + "def require_columns(obs: pd.DataFrame, columns: list[str]) -> None:\n", + " missing = [col for col in columns if col not in obs.columns]\n", + " if missing:\n", + " raise KeyError(f\"Missing required obs columns: {missing}\")\n", + "\n", + "\n", + "def take_dense_rows(X, idx: np.ndarray) -> np.ndarray:\n", + " out = X[idx]\n", + " if sp.issparse(out):\n", + " out = out.toarray()\n", + " return np.asarray(out, dtype=np.float32)\n", + "\n", + "\n", + "def mean_vector(X) -> np.ndarray:\n", + " if sp.issparse(X):\n", + " return np.asarray(X.mean(axis=0)).ravel().astype(np.float32)\n", + " return np.asarray(X, dtype=np.float32).mean(axis=0)\n", + "\n", + "\n", + "def safe_pearson(x: np.ndarray, y: np.ndarray) -> float:\n", + " x = np.asarray(x).ravel()\n", + " y = np.asarray(y).ravel()\n", + " if x.size < 2 or np.std(x) == 0 or np.std(y) == 0:\n", + " return np.nan\n", + " return float(np.corrcoef(x, y)[0, 1])\n", + "\n", + "\n", + "def split_array_indices(indices: np.ndarray, rng: np.random.Generator) -> tuple[np.ndarray, np.ndarray, np.ndarray]:\n", + " indices = np.asarray(indices)\n", + " if len(indices) == 0:\n", + " return indices, indices, indices\n", + " shuffled = indices.copy()\n", + " rng.shuffle(shuffled)\n", + " if len(shuffled) < 3:\n", + " return np.sort(shuffled), np.array([], dtype=indices.dtype), np.array([], dtype=indices.dtype)\n", + " n_train = max(1, int(round(len(shuffled) * TRAIN_FRAC)))\n", + " n_val = int(round(len(shuffled) * VAL_FRAC))\n", + " n_train = min(n_train, len(shuffled) - 2)\n", + " n_val = max(1, min(n_val, len(shuffled) - n_train - 1))\n", + " train = np.sort(shuffled[:n_train])\n", + " val = np.sort(shuffled[n_train : n_train + n_val])\n", + " test = np.sort(shuffled[n_train + n_val :])\n", + " if len(test) == 0:\n", + " test = val[-1:].copy()\n", + " val = val[:-1]\n", + " if len(val) == 0:\n", + " val = train[-1:].copy()\n", + " train = train[:-1]\n", + " return train, val, test\n", + "\n", + "\n", + "def assign_seen_cell_split(obs: pd.DataFrame, seed: int) -> pd.Categorical:\n", + " rng = np.random.default_rng(seed)\n", + " split = np.full(len(obs), \"train\", dtype=object)\n", + " is_control = normalize_bool(obs[\"is_control\"]).to_numpy(dtype=bool)\n", + " genes = obs[\"perturbation_gene\"].astype(str).to_numpy()\n", + "\n", + " control_idx = np.where(is_control)[0]\n", + " train_idx, val_idx, test_idx = split_array_indices(control_idx, rng)\n", + " split[train_idx] = \"train\"\n", + " split[val_idx] = \"val\"\n", + " split[test_idx] = \"test\"\n", + "\n", + " for gene in sorted(set(genes[~is_control])):\n", + " idx = np.where((~is_control) & (genes == gene))[0]\n", + " train_idx, val_idx, test_idx = split_array_indices(idx, rng)\n", + " split[train_idx] = \"train\"\n", + " split[val_idx] = \"val\"\n", + " split[test_idx] = \"test\"\n", + "\n", + " return pd.Categorical(split, categories=[\"train\", \"val\", \"test\"], ordered=True)\n", + "\n", + "\n", + "def choose_feature_mask(var: pd.DataFrame) -> np.ndarray:\n", + " if not USE_FEATURE_GENES:\n", + " return np.ones(var.shape[0], dtype=bool)\n", + " for column in FEATURE_GENE_COLUMNS:\n", + " if column in var.columns:\n", + " mask = var[column].astype(bool).to_numpy()\n", + " if mask.any():\n", + " print(f\"Using feature mask from var['{column}']: {int(mask.sum())} genes\")\n", + " return mask\n", + " print(\"Feature-gene columns not found or empty; using all genes.\")\n", + " return np.ones(var.shape[0], dtype=bool)\n", + "\n", + "\n", + "def median_heuristic_sigma(x: torch.Tensor, y: torch.Tensor, max_points: int = 256) -> float:\n", + " z = torch.cat([x[:max_points], y[:max_points]], dim=0)\n", + " if z.shape[0] < 2:\n", + " return 1.0\n", + " d2 = torch.cdist(z, z, p=2).pow(2)\n", + " mask = ~torch.eye(d2.shape[0], dtype=torch.bool, device=d2.device)\n", + " valid = d2[mask]\n", + " valid = valid[valid > 0]\n", + " if valid.numel() == 0:\n", + " return 1.0\n", + " return float(torch.sqrt(valid.median()).item())\n", + "\n", + "\n", + "def rbf_mmd_loss(x: torch.Tensor, y: torch.Tensor, sigma: float | None = None) -> torch.Tensor:\n", + " if x.shape[0] < 2 or y.shape[0] < 2:\n", + " return x.new_tensor(0.0)\n", + " sigma = sigma or median_heuristic_sigma(x, y)\n", + " gamma = 1.0 / (2.0 * sigma * sigma + 1e-8)\n", + " d_xx = torch.cdist(x, x, p=2).pow(2)\n", + " d_yy = torch.cdist(y, y, p=2).pow(2)\n", + " d_xy = torch.cdist(x, y, p=2).pow(2)\n", + " k_xx = torch.exp(-gamma * d_xx)\n", + " k_yy = torch.exp(-gamma * d_yy)\n", + " k_xy = torch.exp(-gamma * d_xy)\n", + " return k_xx.mean() + k_yy.mean() - 2.0 * k_xy.mean()\n", + "\n", + "\n", + "def sinkhorn_barycentric_targets(x0: torch.Tensor, x1: torch.Tensor, epsilon: float, n_iters: int) -> torch.Tensor:\n", + " cost = torch.cdist(x0, x1, p=2).pow(2)\n", + " kernel = torch.exp(-cost / max(epsilon, 1e-4)).clamp_min(1e-8)\n", + "\n", + " a = torch.full((x0.shape[0],), 1.0 / x0.shape[0], device=x0.device, dtype=x0.dtype)\n", + " b = torch.full((x1.shape[0],), 1.0 / x1.shape[0], device=x1.device, dtype=x1.dtype)\n", + " u = torch.ones_like(a)\n", + " v = torch.ones_like(b)\n", + "\n", + " for _ in range(n_iters):\n", + " u = a / (kernel @ v + 1e-8)\n", + " v = b / (kernel.t() @ u + 1e-8)\n", + "\n", + " plan = u[:, None] * kernel * v[None, :]\n", + " row_mass = plan.sum(dim=1, keepdim=True).clamp_min(1e-8)\n", + " return (plan / row_mass) @ x1\n", + "\n", + "\n", + "def build_aggregate_frame(metrics: pd.DataFrame) -> pd.DataFrame:\n", + " numeric = metrics.select_dtypes(include=[np.number])\n", + " if numeric.empty:\n", + " return pd.DataFrame()\n", + " return numeric.agg([\"mean\", \"median\"]).T\n" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "6f4c5289", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Using feature mask from var['is_feature_gene']: 4334 genes\n", + "adata: AnnData object with n_obs × n_vars = 310385 × 4334\n", + " obs: 'batch', 'gene', 'gene_id', 'transcript', 'gene_transcript', 'guide_id', 'percent_mito', 'UMI_count', 'z_gemgroup_UMI', 'core_scale_factor', 'core_adjusted_UMI_count', 'disease', 'cancer', 'cell_line', 'sex', 'age', 'perturbation', 'organism', 'perturbation_type', 'tissue_type', 'ncounts', 'ngenes', 'nperts', 'percent_ribo', 'perturbation_label', 'perturbation_gene', 'is_control', 'is_single_perturbation', 'split', 'condition', 'batch_model'\n", + " var: 'chr', 'start', 'end', 'class', 'strand', 'length', 'in_matrix', 'mean', 'std', 'cv', 'fano', 'ensembl_id', 'ncounts', 'ncells', 'highly_variable', 'highly_variable_rank', 'means', 'variances', 'variances_norm', 'is_target_gene', 'is_feature_gene'\n", + " uns: 'hvg', 'log1p', 'prediction_ready'\n", + " layers: 'counts'\n", + "split counts:\n", + "split\n", + "train 248343\n", + "test 33235\n", + "val 28807\n", + "Name: count, dtype: int64\n", + "train conditions: 1645\n", + "val conditions: 206\n", + "eval conditions: 205\n", + "eval genes missing from training: 205\n", + "first unseen eval genes: ['ADSL', 'ANAPC2', 'ANAPC5', 'ANKRD11', 'ARL2', 'ARPC3', 'ATP6V0D1', 'BCAS2', 'BRF1', 'C14orf178']\n" + ] + } + ], + "source": [ + "adata = ad.read_h5ad(INPUT_PATH)\n", + "require_columns(adata.obs, [\"perturbation_gene\", \"is_control\", \"split\"])\n", + "\n", + "adata.obs[\"is_control\"] = normalize_bool(adata.obs[\"is_control\"])\n", + "adata.obs[\"perturbation_gene\"] = adata.obs[\"perturbation_gene\"].astype(str).str.strip()\n", + "adata.obs[\"perturbation_gene\"] = adata.obs[\"perturbation_gene\"].mask(adata.obs[\"is_control\"], \"__ctrl__\")\n", + "adata.obs[\"condition\"] = adata.obs[\"perturbation_gene\"].astype(str)\n", + "adata.obs[\"batch_model\"] = adata.obs[\"batch\"].astype(str) if \"batch\" in adata.obs.columns else \"global\"\n", + "\n", + "feature_mask = choose_feature_mask(adata.var)\n", + "adata = adata[:, feature_mask].copy()\n", + "\n", + "if SPLIT_STRATEGY == \"seen_cell_split\":\n", + " adata.obs[\"source_split\"] = adata.obs[\"split\"].astype(str)\n", + " adata.obs[\"split\"] = assign_seen_cell_split(adata.obs, RANDOM_SEED)\n", + "elif SPLIT_STRATEGY == \"use_existing\":\n", + " adata.obs[\"split\"] = pd.Categorical(\n", + " adata.obs[\"split\"].astype(str),\n", + " categories=[\"train\", \"val\", \"test\"],\n", + " ordered=True,\n", + " )\n", + "else:\n", + " raise ValueError(f\"Unknown SPLIT_STRATEGY: {SPLIT_STRATEGY}\")\n", + "\n", + "is_control = adata.obs[\"is_control\"].to_numpy(dtype=bool)\n", + "split = adata.obs[\"split\"].astype(str).to_numpy()\n", + "genes = adata.obs[\"perturbation_gene\"].astype(str).to_numpy()\n", + "batches = adata.obs[\"batch_model\"].astype(str).to_numpy()\n", + "\n", + "control_train_idx = np.where(is_control & (split == CONTROL_SPLIT))[0]\n", + "if len(control_train_idx) == 0:\n", + " print(f\"No controls found in CONTROL_SPLIT={CONTROL_SPLIT}; falling back to all controls.\")\n", + " control_train_idx = np.where(is_control)[0]\n", + "if len(control_train_idx) == 0:\n", + " raise ValueError(\"No control cells found.\")\n", + "\n", + "train_conditions = sorted(set(genes[(~is_control) & (split == TRAIN_SPLIT)]))\n", + "val_conditions = sorted(set(genes[(~is_control) & (split == VAL_SPLIT)]))\n", + "eval_conditions = sorted(set(genes[(~is_control) & (split == EVAL_SPLIT)]))\n", + "\n", + "if MAX_TRAIN_CONDITIONS is not None:\n", + " train_conditions = train_conditions[:MAX_TRAIN_CONDITIONS]\n", + "if MAX_EVAL_GENES is not None:\n", + " eval_conditions = eval_conditions[:MAX_EVAL_GENES]\n", + "\n", + "indices_by_split_gene = {split_name: {} for split_name in [TRAIN_SPLIT, VAL_SPLIT, EVAL_SPLIT]}\n", + "for split_name in [TRAIN_SPLIT, VAL_SPLIT, EVAL_SPLIT]:\n", + " split_mask = (~is_control) & (split == split_name)\n", + " allowed = train_conditions if split_name == TRAIN_SPLIT else val_conditions if split_name == VAL_SPLIT else eval_conditions\n", + " for gene in allowed:\n", + " idx = np.where(split_mask & (genes == gene))[0]\n", + " if len(idx) > 0:\n", + " indices_by_split_gene[split_name][gene] = idx\n", + "\n", + "control_idx_by_batch = {}\n", + "for batch_name in sorted(set(batches[control_train_idx])):\n", + " idx = control_train_idx[batches[control_train_idx] == batch_name]\n", + " if len(idx) > 0:\n", + " control_idx_by_batch[batch_name] = idx\n", + "\n", + "gene_vocab = [\"__ctrl__\", \"__unk__\", *train_conditions]\n", + "gene_to_idx = {gene: idx for idx, gene in enumerate(gene_vocab)}\n", + "batch_vocab = [\"__unk__\", *sorted(set(batches))]\n", + "batch_to_idx = {batch: idx for idx, batch in enumerate(batch_vocab)}\n", + "batch_index_per_obs = np.array([batch_to_idx.get(batch, batch_to_idx[\"__unk__\"]) for batch in batches], dtype=np.int64)\n", + "\n", + "control_mean = mean_vector(adata.X[control_train_idx])\n", + "unseen_eval_genes = sorted(set(eval_conditions) - set(train_conditions))\n", + "\n", + "print(\"adata:\", adata)\n", + "print(\"split counts:\")\n", + "print(adata.obs[\"split\"].astype(str).value_counts())\n", + "print(\"train conditions:\", len(train_conditions))\n", + "print(\"val conditions:\", len(val_conditions))\n", + "print(\"eval conditions:\", len(eval_conditions))\n", + "print(\"eval genes missing from training:\", len(unseen_eval_genes))\n", + "if unseen_eval_genes[:10]:\n", + " print(\"first unseen eval genes:\", unseen_eval_genes[:10])\n" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "e8265cef", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "CellOTStyleBaseline(\n", + " (encoder): MLPEncoder(\n", + " (net): Sequential(\n", + " (0): Linear(in_features=4334, out_features=512, bias=True)\n", + " (1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", + " (2): GELU(approximate='none')\n", + " (3): Dropout(p=0.1, inplace=False)\n", + " (4): Linear(in_features=512, out_features=256, bias=True)\n", + " (5): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n", + " (6): GELU(approximate='none')\n", + " (7): Dropout(p=0.1, inplace=False)\n", + " (8): Linear(in_features=256, out_features=64, bias=True)\n", + " )\n", + " )\n", + " (decoder): MLPDecoder(\n", + " (net): Sequential(\n", + " (0): Linear(in_features=64, out_features=256, bias=True)\n", + " (1): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n", + " (2): GELU(approximate='none')\n", + " (3): Dropout(p=0.1, inplace=False)\n", + " (4): Linear(in_features=256, out_features=512, bias=True)\n", + " (5): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", + " (6): GELU(approximate='none')\n", + " (7): Dropout(p=0.1, inplace=False)\n", + " (8): Linear(in_features=512, out_features=4334, bias=True)\n", + " )\n", + " )\n", + " (condition_encoder): ConditionEncoder(\n", + " (gene_embedding): Embedding(1647, 64)\n", + " (batch_embedding): Embedding(49, 32)\n", + " (mlp): Sequential(\n", + " (0): Linear(in_features=96, out_features=128, bias=True)\n", + " (1): LayerNorm((128,), eps=1e-05, elementwise_affine=True)\n", + " (2): GELU(approximate='none')\n", + " (3): Dropout(p=0.1, inplace=False)\n", + " (4): Linear(in_features=128, out_features=128, bias=True)\n", + " )\n", + " )\n", + " (transport_map): ConditionalTransportMap(\n", + " (input_proj): Linear(in_features=192, out_features=256, bias=True)\n", + " (blocks): ModuleList(\n", + " (0-3): 4 x TransportResidualBlock(\n", + " (norm1): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n", + " (norm2): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n", + " (fc1): Linear(in_features=256, out_features=256, bias=True)\n", + " (fc2): Linear(in_features=256, out_features=256, bias=True)\n", + " (film1): Linear(in_features=128, out_features=512, bias=True)\n", + " (film2): Linear(in_features=128, out_features=512, bias=True)\n", + " (dropout): Dropout(p=0.1, inplace=False)\n", + " )\n", + " )\n", + " (out): Linear(in_features=256, out_features=64, bias=True)\n", + " )\n", + ")\n" + ] + } + ], + "source": [ + "class MLPEncoder(nn.Module):\n", + " def __init__(self, input_dim: int, latent_dim: int, hidden_dim: int, dropout: float):\n", + " super().__init__()\n", + " self.net = nn.Sequential(\n", + " nn.Linear(input_dim, hidden_dim * 2),\n", + " nn.LayerNorm(hidden_dim * 2),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(hidden_dim * 2, hidden_dim),\n", + " nn.LayerNorm(hidden_dim),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(hidden_dim, latent_dim),\n", + " )\n", + "\n", + " def forward(self, x: torch.Tensor) -> torch.Tensor:\n", + " return self.net(x)\n", + "\n", + "\n", + "class MLPDecoder(nn.Module):\n", + " def __init__(self, latent_dim: int, output_dim: int, hidden_dim: int, dropout: float):\n", + " super().__init__()\n", + " self.net = nn.Sequential(\n", + " nn.Linear(latent_dim, hidden_dim),\n", + " nn.LayerNorm(hidden_dim),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(hidden_dim, hidden_dim * 2),\n", + " nn.LayerNorm(hidden_dim * 2),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(hidden_dim * 2, output_dim),\n", + " )\n", + "\n", + " def forward(self, z: torch.Tensor) -> torch.Tensor:\n", + " return F.softplus(self.net(z))\n", + "\n", + "\n", + "class ConditionEncoder(nn.Module):\n", + " def __init__(self, n_genes: int, n_batches: int, condition_dim: int, dropout: float):\n", + " super().__init__()\n", + " gene_dim = max(32, condition_dim // 2)\n", + " batch_dim = max(16, condition_dim // 4)\n", + " self.gene_embedding = nn.Embedding(n_genes, gene_dim)\n", + " self.batch_embedding = nn.Embedding(n_batches, batch_dim)\n", + " self.mlp = nn.Sequential(\n", + " nn.Linear(gene_dim + batch_dim, condition_dim),\n", + " nn.LayerNorm(condition_dim),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(condition_dim, condition_dim),\n", + " )\n", + "\n", + " def forward(self, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", + " gene_emb = self.gene_embedding(gene_idx)\n", + " batch_emb = self.batch_embedding(batch_idx)\n", + " return self.mlp(torch.cat([gene_emb, batch_emb], dim=-1))\n", + "\n", + "\n", + "class TransportResidualBlock(nn.Module):\n", + " def __init__(self, hidden_dim: int, condition_dim: int, dropout: float):\n", + " super().__init__()\n", + " self.norm1 = nn.LayerNorm(hidden_dim)\n", + " self.norm2 = nn.LayerNorm(hidden_dim)\n", + " self.fc1 = nn.Linear(hidden_dim, hidden_dim)\n", + " self.fc2 = nn.Linear(hidden_dim, hidden_dim)\n", + " self.film1 = nn.Linear(condition_dim, hidden_dim * 2)\n", + " self.film2 = nn.Linear(condition_dim, hidden_dim * 2)\n", + " self.dropout = nn.Dropout(dropout)\n", + "\n", + " def forward(self, x: torch.Tensor, cond: torch.Tensor) -> torch.Tensor:\n", + " gamma1, beta1 = self.film1(cond).chunk(2, dim=-1)\n", + " h = self.norm1(x)\n", + " h = h * (1 + gamma1) + beta1\n", + " h = F.gelu(self.fc1(h))\n", + " h = self.dropout(h)\n", + "\n", + " gamma2, beta2 = self.film2(cond).chunk(2, dim=-1)\n", + " h = self.norm2(h)\n", + " h = h * (1 + gamma2) + beta2\n", + " h = self.fc2(h)\n", + " h = self.dropout(h)\n", + " return x + h\n", + "\n", + "\n", + "class ConditionalTransportMap(nn.Module):\n", + " def __init__(self, latent_dim: int, condition_dim: int, hidden_dim: int, num_blocks: int, dropout: float):\n", + " super().__init__()\n", + " self.input_proj = nn.Linear(latent_dim + condition_dim, hidden_dim)\n", + " self.blocks = nn.ModuleList(\n", + " [TransportResidualBlock(hidden_dim, condition_dim, dropout) for _ in range(num_blocks)]\n", + " )\n", + " self.out = nn.Linear(hidden_dim, latent_dim)\n", + "\n", + " def forward(self, z_ctrl: torch.Tensor, cond: torch.Tensor) -> torch.Tensor:\n", + " h = self.input_proj(torch.cat([z_ctrl, cond], dim=-1))\n", + " for block in self.blocks:\n", + " h = block(h, cond)\n", + " delta = self.out(h)\n", + " return z_ctrl + delta\n", + "\n", + "\n", + "class CellOTStyleBaseline(nn.Module):\n", + " def __init__(self, input_dim: int, latent_dim: int, hidden_dim: int, condition_dim: int, n_genes: int, n_batches: int, num_blocks: int, dropout: float):\n", + " super().__init__()\n", + " self.encoder = MLPEncoder(input_dim=input_dim, latent_dim=latent_dim, hidden_dim=hidden_dim, dropout=dropout)\n", + " self.decoder = MLPDecoder(latent_dim=latent_dim, output_dim=input_dim, hidden_dim=hidden_dim, dropout=dropout)\n", + " self.condition_encoder = ConditionEncoder(n_genes=n_genes, n_batches=n_batches, condition_dim=condition_dim, dropout=dropout)\n", + " self.transport_map = ConditionalTransportMap(\n", + " latent_dim=latent_dim,\n", + " condition_dim=condition_dim,\n", + " hidden_dim=hidden_dim,\n", + " num_blocks=num_blocks,\n", + " dropout=dropout,\n", + " )\n", + "\n", + " def encode_x(self, x: torch.Tensor) -> torch.Tensor:\n", + " return self.encoder(x)\n", + "\n", + " def decode_z(self, z: torch.Tensor) -> torch.Tensor:\n", + " return self.decoder(z)\n", + "\n", + " def reconstruct(self, x: torch.Tensor) -> torch.Tensor:\n", + " return self.decode_z(self.encode_x(x))\n", + "\n", + " def transport(self, z_ctrl: torch.Tensor, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", + " cond = self.condition_encoder(gene_idx, batch_idx)\n", + " return self.transport_map(z_ctrl, cond)\n", + "\n", + "\n", + "model = CellOTStyleBaseline(\n", + " input_dim=adata.n_vars,\n", + " latent_dim=LATENT_DIM,\n", + " hidden_dim=HIDDEN_DIM,\n", + " condition_dim=CONDITION_DIM,\n", + " n_genes=len(gene_vocab),\n", + " n_batches=len(batch_vocab),\n", + " num_blocks=NUM_TRANSPORT_BLOCKS,\n", + " dropout=DROPOUT,\n", + ").to(DEVICE)\n", + "\n", + "optimizer = torch.optim.AdamW(model.parameters(), lr=LEARNING_RATE, weight_decay=WEIGHT_DECAY)\n", + "print(model)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "273f8ffa", + "metadata": {}, + "outputs": [], + "source": [ + "def sample_controls_like_targets(target_idx: np.ndarray, rng: np.random.Generator) -> np.ndarray:\n", + " sampled = []\n", + " for idx in target_idx:\n", + " batch_name = batches[idx]\n", + " pool = control_idx_by_batch.get(batch_name, control_train_idx)\n", + " sampled.append(rng.choice(pool))\n", + " return np.asarray(sampled, dtype=np.int64)\n", + "\n", + "\n", + "def make_condition_batch(condition: str, split_name: str, rng: np.random.Generator) -> dict:\n", + " target_pool = indices_by_split_gene[split_name][condition]\n", + " size = min(BATCH_SIZE, len(target_pool))\n", + " replace = len(target_pool) < size\n", + " target_idx = rng.choice(target_pool, size=size, replace=replace)\n", + " source_idx = sample_controls_like_targets(target_idx, rng)\n", + "\n", + " if MAX_EVAL_CELLS_PER_GENE is not None and len(target_idx) > MAX_EVAL_CELLS_PER_GENE:\n", + " target_idx = target_idx[:MAX_EVAL_CELLS_PER_GENE]\n", + " source_idx = source_idx[:MAX_EVAL_CELLS_PER_GENE]\n", + "\n", + " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", + " target_x = torch.from_numpy(take_dense_rows(adata.X, target_idx)).to(DEVICE)\n", + " gene_idx = torch.full(\n", + " (len(target_idx),),\n", + " fill_value=gene_to_idx.get(condition, gene_to_idx[\"__unk__\"]),\n", + " device=DEVICE,\n", + " dtype=torch.long,\n", + " )\n", + " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", + "\n", + " return {\n", + " \"condition\": condition,\n", + " \"source_idx\": source_idx,\n", + " \"target_idx\": target_idx,\n", + " \"source_x\": source_x,\n", + " \"target_x\": target_x,\n", + " \"gene_idx\": gene_idx,\n", + " \"batch_idx\": batch_idx,\n", + " }\n", + "\n", + "\n", + "def compute_loss_terms(batch: dict) -> dict[str, torch.Tensor]:\n", + " z_ctrl = model.encode_x(batch[\"source_x\"])\n", + " z_target = model.encode_x(batch[\"target_x\"])\n", + " z_target_bar = sinkhorn_barycentric_targets(\n", + " x0=z_ctrl,\n", + " x1=z_target,\n", + " epsilon=SINKHORN_EPSILON,\n", + " n_iters=SINKHORN_ITERS,\n", + " )\n", + "\n", + " z_pred = model.transport(z_ctrl=z_ctrl, gene_idx=batch[\"gene_idx\"], batch_idx=batch[\"batch_idx\"])\n", + " pred_x = model.decode_z(z_pred)\n", + "\n", + " source_recon = model.reconstruct(batch[\"source_x\"])\n", + " target_recon = model.reconstruct(batch[\"target_x\"])\n", + "\n", + " recon_loss = F.mse_loss(source_recon, batch[\"source_x\"]) + F.mse_loss(target_recon, batch[\"target_x\"])\n", + " map_loss = F.mse_loss(z_pred, z_target_bar)\n", + " latent_mmd = rbf_mmd_loss(z_pred, z_target)\n", + " expression_mmd = rbf_mmd_loss(pred_x, batch[\"target_x\"])\n", + " mean_loss = F.mse_loss(pred_x.mean(dim=0), batch[\"target_x\"].mean(dim=0))\n", + " displacement_reg = torch.mean((z_pred - z_ctrl) ** 2)\n", + "\n", + " total_loss = (\n", + " RECON_WEIGHT * recon_loss\n", + " + MAP_WEIGHT * map_loss\n", + " + LATENT_MMD_WEIGHT * latent_mmd\n", + " + EXPRESSION_MMD_WEIGHT * expression_mmd\n", + " + MEAN_WEIGHT * mean_loss\n", + " + DISPLACEMENT_REG * displacement_reg\n", + " )\n", + "\n", + " return {\n", + " \"total\": total_loss,\n", + " \"recon\": recon_loss.detach(),\n", + " \"map\": map_loss.detach(),\n", + " \"latent_mmd\": latent_mmd.detach(),\n", + " \"expression_mmd\": expression_mmd.detach(),\n", + " \"mean\": mean_loss.detach(),\n", + " \"disp_reg\": displacement_reg.detach(),\n", + " }\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def evaluate_split_loss(split_name: str, conditions: list[str], seed: int, steps: int) -> dict[str, float]:\n", + " if not conditions:\n", + " return {}\n", + "\n", + " model.eval()\n", + " rng = np.random.default_rng(seed)\n", + " subset = conditions.copy()\n", + " rng.shuffle(subset)\n", + " subset = subset[: min(len(subset), steps)]\n", + "\n", + " rows = []\n", + " for condition in subset:\n", + " batch = make_condition_batch(condition=condition, split_name=split_name, rng=rng)\n", + " losses = compute_loss_terms(batch)\n", + " rows.append({key: float(value.item()) for key, value in losses.items()})\n", + "\n", + " return pd.DataFrame(rows).mean().to_dict()\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def predict_gene_expression(condition: str, split_name: str, rng: np.random.Generator) -> tuple[np.ndarray, np.ndarray, pd.DataFrame]:\n", + " target_idx_all = indices_by_split_gene[split_name][condition]\n", + " if MAX_EVAL_CELLS_PER_GENE is not None and len(target_idx_all) > MAX_EVAL_CELLS_PER_GENE:\n", + " target_idx_all = target_idx_all[:MAX_EVAL_CELLS_PER_GENE]\n", + "\n", + " pred_blocks = []\n", + " real_blocks = []\n", + " obs_blocks = []\n", + " gene_id = gene_to_idx.get(condition, gene_to_idx[\"__unk__\"])\n", + " condition_source = \"trained_gene_embedding\" if condition in gene_to_idx else \"unk_gene_embedding\"\n", + "\n", + " for start in range(0, len(target_idx_all), PRED_BATCH_SIZE):\n", + " target_idx = target_idx_all[start : start + PRED_BATCH_SIZE]\n", + " source_idx = sample_controls_like_targets(target_idx, rng)\n", + "\n", + " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", + " z_ctrl = model.encode_x(source_x)\n", + " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", + " gene_idx = torch.full((len(target_idx),), gene_id, device=DEVICE, dtype=torch.long)\n", + "\n", + " z_pred = model.transport(z_ctrl=z_ctrl, gene_idx=gene_idx, batch_idx=batch_idx)\n", + " pred_x = model.decode_z(z_pred).cpu().numpy().astype(np.float32)\n", + " real_x = take_dense_rows(adata.X, target_idx)\n", + "\n", + " pred_blocks.append(pred_x)\n", + " real_blocks.append(real_x)\n", + "\n", + " obs_chunk = adata.obs.iloc[target_idx][[\"perturbation_gene\", \"condition\", \"split\"]].copy()\n", + " obs_chunk[\"target_cell_id\"] = adata.obs_names[target_idx].astype(str)\n", + " obs_chunk[\"sampled_control_cell_id\"] = adata.obs_names[source_idx].astype(str)\n", + " obs_chunk[\"condition_source\"] = condition_source\n", + " obs_blocks.append(obs_chunk)\n", + "\n", + " pred_gene = np.vstack(pred_blocks).astype(np.float32)\n", + " real_gene = np.vstack(real_blocks).astype(np.float32)\n", + " obs_gene = pd.concat(obs_blocks, axis=0)\n", + " return pred_gene, real_gene, obs_gene\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def evaluate_gene_metrics(split_name: str, conditions: list[str], seed: int) -> tuple[pd.DataFrame, ad.AnnData, ad.AnnData]:\n", + " rng = np.random.default_rng(seed)\n", + "\n", + " pred_blocks = []\n", + " real_blocks = []\n", + " obs_blocks = []\n", + " metrics_rows = []\n", + "\n", + " for condition in conditions:\n", + " pred_gene, real_gene, obs_gene = predict_gene_expression(condition=condition, split_name=split_name, rng=rng)\n", + " pred_blocks.append(pred_gene)\n", + " real_blocks.append(real_gene)\n", + " obs_blocks.append(obs_gene)\n", + "\n", + " pred_mean = pred_gene.mean(axis=0)\n", + " real_mean = real_gene.mean(axis=0)\n", + " pred_delta = pred_mean - control_mean\n", + " real_delta = real_mean - control_mean\n", + "\n", + " mse_cell = float(np.mean((pred_gene - real_gene) ** 2))\n", + " spearman_mean = float(stats.spearmanr(pred_mean, real_mean, nan_policy=\"omit\").statistic)\n", + " if not np.isfinite(spearman_mean):\n", + " spearman_mean = float(\"nan\")\n", + "\n", + " metrics_rows.append(\n", + " {\n", + " \"perturbation_gene\": condition,\n", + " \"n_cells\": int(real_gene.shape[0]),\n", + " \"mse_mean\": float(np.mean((pred_mean - real_mean) ** 2)),\n", + " \"mse_cell\": mse_cell,\n", + " \"mae_mean\": float(np.mean(np.abs(pred_mean - real_mean))),\n", + " \"pearson_mean\": safe_pearson(pred_mean, real_mean),\n", + " \"spearman_mean\": spearman_mean,\n", + " \"pearson_delta\": safe_pearson(pred_delta, real_delta),\n", + " \"delta_l2\": float(np.linalg.norm(pred_delta - real_delta)),\n", + " }\n", + " )\n", + "\n", + " pred = ad.AnnData(\n", + " X=np.vstack(pred_blocks).astype(np.float32),\n", + " obs=pd.concat(obs_blocks, axis=0),\n", + " var=adata.var.copy(),\n", + " )\n", + " real = ad.AnnData(\n", + " X=np.vstack(real_blocks).astype(np.float32),\n", + " obs=pd.concat(obs_blocks, axis=0),\n", + " var=adata.var.copy(),\n", + " )\n", + " metrics = pd.DataFrame(metrics_rows).sort_values(\"perturbation_gene\").reset_index(drop=True)\n", + " return metrics, pred, real\n" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "b4f974cf", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{'epoch': 1, 'train_total': 0.4352, 'train_map': 0.0751, 'train_recon': 0.2994, 'val_total': 0.3468}\n", + "{'epoch': 2, 'train_total': 0.3052, 'train_map': 0.0051, 'train_recon': 0.2728, 'val_total': 0.3564}\n", + "{'epoch': 3, 'train_total': 0.3028, 'train_map': 0.0029, 'train_recon': 0.2739, 'val_total': 0.3425}\n", + "{'epoch': 4, 'train_total': 0.2992, 'train_map': 0.0023, 'train_recon': 0.2712, 'val_total': 0.3496}\n", + "{'epoch': 5, 'train_total': 0.3061, 'train_map': 0.0019, 'train_recon': 0.2771, 'val_total': 0.3417}\n", + "{'epoch': 6, 'train_total': 0.3027, 'train_map': 0.0017, 'train_recon': 0.2741, 'val_total': 0.3446}\n", + "{'epoch': 7, 'train_total': 0.3003, 'train_map': 0.0013, 'train_recon': 0.2728, 'val_total': 0.3428}\n", + "{'epoch': 8, 'train_total': 0.3001, 'train_map': 0.001, 'train_recon': 0.2734, 'val_total': 0.3222}\n", + "{'epoch': 9, 'train_total': 0.2983, 'train_map': 0.0008, 'train_recon': 0.272, 'val_total': 0.3329}\n", + "{'epoch': 10, 'train_total': 0.2985, 'train_map': 0.0008, 'train_recon': 0.2721, 'val_total': 0.333}\n", + "{'epoch': 11, 'train_total': 0.3015, 'train_map': 0.0007, 'train_recon': 0.275, 'val_total': 0.3225}\n", + "{'epoch': 12, 'train_total': 0.3013, 'train_map': 0.0006, 'train_recon': 0.2752, 'val_total': 0.3278}\n", + "{'epoch': 13, 'train_total': 0.2992, 'train_map': 0.0006, 'train_recon': 0.2734, 'val_total': 0.3154}\n", + "{'epoch': 14, 'train_total': 0.2961, 'train_map': 0.0006, 'train_recon': 0.2712, 'val_total': 0.3043}\n", + "{'epoch': 15, 'train_total': 0.299, 'train_map': 0.0007, 'train_recon': 0.273, 'val_total': 0.3167}\n", + "{'epoch': 16, 'train_total': 0.2969, 'train_map': 0.0006, 'train_recon': 0.2712, 'val_total': 0.3215}\n", + "{'epoch': 17, 'train_total': 0.3004, 'train_map': 0.0007, 'train_recon': 0.2738, 'val_total': 0.3284}\n", + "{'epoch': 18, 'train_total': 0.2992, 'train_map': 0.0006, 'train_recon': 0.2736, 'val_total': 0.3302}\n", + "{'epoch': 19, 'train_total': 0.2988, 'train_map': 0.0008, 'train_recon': 0.2723, 'val_total': 0.311}\n", + "{'epoch': 20, 'train_total': 0.2964, 'train_map': 0.0006, 'train_recon': 0.2718, 'val_total': 0.3052}\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
epochtrain_totaltrain_recontrain_maptrain_latent_mmdtrain_expression_mmdtrain_meantrain_disp_regval_totalval_reconval_mapval_latent_mmdval_expression_mmdval_meanval_disp_reg
010.4351980.2994260.0751210.2357430.1894580.0174380.0693440.3468350.2711210.0033790.4873400.1796950.0055970.003379
120.3052270.2727610.0050560.0480900.1750270.0050630.0035580.3563670.2760070.0010950.5588130.1797290.0054000.001094
230.3028410.2738650.0028550.0287160.1787260.0053600.0016530.3424720.2772120.0004640.4051010.1793680.0063440.000464
340.2992050.2711800.0022630.0329100.1769100.0047670.0012440.3495630.2673920.0006110.5800610.1825480.0052920.000608
450.3061260.2770710.0018800.0333350.1791200.0059200.0009360.3417020.2726430.0002580.4440430.1834050.0060530.000259
560.3026830.2740750.0016920.0347310.1788810.0055480.0006650.3445770.2739150.0002480.4761490.1746330.0053330.000248
670.3003150.2727560.0012670.0353980.1772850.0050200.0003900.3428030.2697420.0001550.5038080.1731500.0052080.000155
780.3001230.2734410.0009890.0283170.1776680.0050930.0002080.3222010.2654330.0000560.3632640.1637550.0040090.000055
890.2983070.2719960.0008300.0269400.1778040.0050050.0001170.3329480.2717400.0000390.3961360.1591060.0056450.000038
9100.2984750.2721160.0007900.0259510.1788610.0050870.0000760.3330090.2769220.0000300.3231990.1634930.0073880.000027
10110.3015070.2749500.0007270.0239820.1790260.0055290.0000510.3224820.2723510.0000200.2882180.1573520.0055540.000018
11120.3013000.2752070.0006410.0218170.1782890.0054410.0000380.3277880.2694780.0000220.3900240.1528920.0039960.000021
12130.2991670.2733770.0006180.0226760.1771540.0051880.0000340.3154170.2673560.0000150.2809330.1554110.0044110.000013
13140.2961290.2711900.0005960.0211980.1768700.0045370.0000270.3043450.2613530.0000120.2363380.1539180.0039540.000010
14150.2990290.2729680.0006580.0238600.1778840.0052290.0000370.3167220.2686800.0000280.2781380.1561230.0045880.000024
15160.2969250.2712170.0005890.0213280.1789790.0050870.0000320.3214860.2742460.0000230.2579550.1541870.0060030.000018
16170.3004040.2737760.0006500.0251070.1781160.0056550.0000530.3284460.2752670.0000310.3134520.1584670.0059560.000027
17180.2992470.2736140.0005800.0211050.1781840.0051240.0000270.3302240.2746100.0000200.3510050.1555170.0049420.000019
18190.2987900.2723080.0007860.0291970.1785070.0049250.0001020.3110260.2670340.0000550.2465810.1521600.0040620.000048
19200.2964450.2717780.0006300.0157460.1773850.0047230.0000340.3052410.2729720.0000090.1321900.1496270.0040780.000006
\n", + "
" + ], + "text/plain": [ + " epoch train_total train_recon train_map train_latent_mmd \\\n", + "0 1 0.435198 0.299426 0.075121 0.235743 \n", + "1 2 0.305227 0.272761 0.005056 0.048090 \n", + "2 3 0.302841 0.273865 0.002855 0.028716 \n", + "3 4 0.299205 0.271180 0.002263 0.032910 \n", + "4 5 0.306126 0.277071 0.001880 0.033335 \n", + "5 6 0.302683 0.274075 0.001692 0.034731 \n", + "6 7 0.300315 0.272756 0.001267 0.035398 \n", + "7 8 0.300123 0.273441 0.000989 0.028317 \n", + "8 9 0.298307 0.271996 0.000830 0.026940 \n", + "9 10 0.298475 0.272116 0.000790 0.025951 \n", + "10 11 0.301507 0.274950 0.000727 0.023982 \n", + "11 12 0.301300 0.275207 0.000641 0.021817 \n", + "12 13 0.299167 0.273377 0.000618 0.022676 \n", + "13 14 0.296129 0.271190 0.000596 0.021198 \n", + "14 15 0.299029 0.272968 0.000658 0.023860 \n", + "15 16 0.296925 0.271217 0.000589 0.021328 \n", + "16 17 0.300404 0.273776 0.000650 0.025107 \n", + "17 18 0.299247 0.273614 0.000580 0.021105 \n", + "18 19 0.298790 0.272308 0.000786 0.029197 \n", + "19 20 0.296445 0.271778 0.000630 0.015746 \n", + "\n", + " train_expression_mmd train_mean train_disp_reg val_total val_recon \\\n", + "0 0.189458 0.017438 0.069344 0.346835 0.271121 \n", + "1 0.175027 0.005063 0.003558 0.356367 0.276007 \n", + "2 0.178726 0.005360 0.001653 0.342472 0.277212 \n", + "3 0.176910 0.004767 0.001244 0.349563 0.267392 \n", + "4 0.179120 0.005920 0.000936 0.341702 0.272643 \n", + "5 0.178881 0.005548 0.000665 0.344577 0.273915 \n", + "6 0.177285 0.005020 0.000390 0.342803 0.269742 \n", + "7 0.177668 0.005093 0.000208 0.322201 0.265433 \n", + "8 0.177804 0.005005 0.000117 0.332948 0.271740 \n", + "9 0.178861 0.005087 0.000076 0.333009 0.276922 \n", + "10 0.179026 0.005529 0.000051 0.322482 0.272351 \n", + "11 0.178289 0.005441 0.000038 0.327788 0.269478 \n", + "12 0.177154 0.005188 0.000034 0.315417 0.267356 \n", + "13 0.176870 0.004537 0.000027 0.304345 0.261353 \n", + "14 0.177884 0.005229 0.000037 0.316722 0.268680 \n", + "15 0.178979 0.005087 0.000032 0.321486 0.274246 \n", + "16 0.178116 0.005655 0.000053 0.328446 0.275267 \n", + "17 0.178184 0.005124 0.000027 0.330224 0.274610 \n", + "18 0.178507 0.004925 0.000102 0.311026 0.267034 \n", + "19 0.177385 0.004723 0.000034 0.305241 0.272972 \n", + "\n", + " val_map val_latent_mmd val_expression_mmd val_mean val_disp_reg \n", + "0 0.003379 0.487340 0.179695 0.005597 0.003379 \n", + "1 0.001095 0.558813 0.179729 0.005400 0.001094 \n", + "2 0.000464 0.405101 0.179368 0.006344 0.000464 \n", + "3 0.000611 0.580061 0.182548 0.005292 0.000608 \n", + "4 0.000258 0.444043 0.183405 0.006053 0.000259 \n", + "5 0.000248 0.476149 0.174633 0.005333 0.000248 \n", + "6 0.000155 0.503808 0.173150 0.005208 0.000155 \n", + "7 0.000056 0.363264 0.163755 0.004009 0.000055 \n", + "8 0.000039 0.396136 0.159106 0.005645 0.000038 \n", + "9 0.000030 0.323199 0.163493 0.007388 0.000027 \n", + "10 0.000020 0.288218 0.157352 0.005554 0.000018 \n", + "11 0.000022 0.390024 0.152892 0.003996 0.000021 \n", + "12 0.000015 0.280933 0.155411 0.004411 0.000013 \n", + "13 0.000012 0.236338 0.153918 0.003954 0.000010 \n", + "14 0.000028 0.278138 0.156123 0.004588 0.000024 \n", + "15 0.000023 0.257955 0.154187 0.006003 0.000018 \n", + "16 0.000031 0.313452 0.158467 0.005956 0.000027 \n", + "17 0.000020 0.351005 0.155517 0.004942 0.000019 \n", + "18 0.000055 0.246581 0.152160 0.004062 0.000048 \n", + "19 0.000009 0.132190 0.149627 0.004078 0.000006 " + ] + }, + "execution_count": 16, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "history_rows = []\n", + "best_state = None\n", + "best_val_total = float(\"inf\")\n", + "rng = np.random.default_rng(RANDOM_SEED)\n", + "\n", + "train_sampling_conditions = [g for g in train_conditions if g in indices_by_split_gene[TRAIN_SPLIT]]\n", + "if not train_sampling_conditions:\n", + " raise ValueError(\"No train perturbation conditions are available.\")\n", + "\n", + "for epoch in range(1, EPOCHS + 1):\n", + " model.train()\n", + " epoch_rows = []\n", + "\n", + " for _ in range(STEPS_PER_EPOCH):\n", + " condition = rng.choice(train_sampling_conditions)\n", + " batch = make_condition_batch(condition=condition, split_name=TRAIN_SPLIT, rng=rng)\n", + "\n", + " optimizer.zero_grad(set_to_none=True)\n", + " losses = compute_loss_terms(batch)\n", + " losses[\"total\"].backward()\n", + " nn.utils.clip_grad_norm_(model.parameters(), max_norm=GRAD_CLIP_NORM)\n", + " optimizer.step()\n", + "\n", + " epoch_rows.append({key: float(value.item()) for key, value in losses.items()})\n", + "\n", + " train_summary = pd.DataFrame(epoch_rows).mean().to_dict()\n", + " val_summary = evaluate_split_loss(\n", + " split_name=VAL_SPLIT,\n", + " conditions=val_conditions,\n", + " seed=RANDOM_SEED + epoch,\n", + " steps=VALIDATION_STEPS,\n", + " ) if val_conditions else {}\n", + "\n", + " row = {\"epoch\": epoch, **{f\"train_{k}\": v for k, v in train_summary.items()}}\n", + " row.update({f\"val_{k}\": v for k, v in val_summary.items()})\n", + " history_rows.append(row)\n", + "\n", + " current_val_total = val_summary.get(\"total\", train_summary[\"total\"])\n", + " if current_val_total < best_val_total:\n", + " best_val_total = current_val_total\n", + " best_state = deepcopy(model.state_dict())\n", + "\n", + " print(\n", + " {\n", + " \"epoch\": epoch,\n", + " \"train_total\": round(train_summary[\"total\"], 4),\n", + " \"train_map\": round(train_summary[\"map\"], 4),\n", + " \"train_recon\": round(train_summary[\"recon\"], 4),\n", + " \"val_total\": round(current_val_total, 4),\n", + " }\n", + " )\n", + "\n", + "if best_state is not None:\n", + " model.load_state_dict(best_state)\n", + "\n", + "history = pd.DataFrame(history_rows)\n", + "history\n" + ] + }, + { + "cell_type": "markdown", + "id": "18e5d97a", + "metadata": {}, + "source": [ + "## Figures — training dynamics\n", + "\n", + "Loss curves (total / validation) and CellOT baseline loss components (transport map, reconstruction, MMD terms).\n" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "8d72e5a1", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "_fig_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "plot_training_history(history, tag=_fig_tag)\n" + ] + }, + { + "cell_type": "code", + "execution_count": 18, + "id": "90c83b99", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Aggregate summary:\n", + "method_name cellot_ot_baseline\n", + "input_path /root/final/data/processed/scPerturb/predictio...\n", + "eval_split test\n", + "train_split train\n", + "val_split val\n", + "n_control_cells_train 10816\n", + "n_eval_cells 33235\n", + "n_train_genes 1645\n", + "n_eval_genes 205\n", + "n_eval_genes_missing_from_train 205\n", + "mse_mean_avg 0.005048\n", + "mse_cell_avg 0.144353\n", + "mae_mean_avg 0.04422\n", + "pearson_mean_avg 0.993442\n", + "spearman_mean_avg 0.973503\n", + "pearson_delta_avg 0.148961\n", + "delta_l2_avg 4.179005\n", + "dtype: object\n", + "\n", + "Saved:\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/cellot_ot_baseline_model.pt\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/pred_cellot_ot_baseline_use_existing_test.h5ad\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/real_cellot_ot_baseline_use_existing_test.h5ad\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/metrics_by_gene_cellot_ot_baseline_use_existing_test.csv\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/aggregate_cellot_ot_baseline_use_existing_test.csv\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/summary_cellot_ot_baseline_use_existing_test.csv\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/training_history_cellot_ot_baseline_use_existing_test.csv\n", + "- /root/final/baseline/outputs/cellot_ot_baseline/config_cellot_ot_baseline_use_existing_test.json\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
perturbation_genen_cellsmse_meanmse_cellmae_meanpearson_meanspearman_meanpearson_deltadelta_l2
0ADSL1330.0017660.1274420.0306980.9976690.9865180.0787232.766647
1ANAPC22370.0014300.1392030.0266480.9981430.9913270.1872012.489359
2ANAPC5400.0042960.1328130.0492470.9942880.9662000.0065844.314793
3ANKRD112600.0024250.1145910.0316800.9972800.990275-0.0832913.242010
4ARL2740.0043000.1649560.0474810.9947820.9710090.2608074.316945
\n", + "
" + ], + "text/plain": [ + " perturbation_gene n_cells mse_mean mse_cell mae_mean pearson_mean \\\n", + "0 ADSL 133 0.001766 0.127442 0.030698 0.997669 \n", + "1 ANAPC2 237 0.001430 0.139203 0.026648 0.998143 \n", + "2 ANAPC5 40 0.004296 0.132813 0.049247 0.994288 \n", + "3 ANKRD11 260 0.002425 0.114591 0.031680 0.997280 \n", + "4 ARL2 74 0.004300 0.164956 0.047481 0.994782 \n", + "\n", + " spearman_mean pearson_delta delta_l2 \n", + "0 0.986518 0.078723 2.766647 \n", + "1 0.991327 0.187201 2.489359 \n", + "2 0.966200 0.006584 4.314793 \n", + "3 0.990275 -0.083291 3.242010 \n", + "4 0.971009 0.260807 4.316945 " + ] + }, + "execution_count": 18, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "eval_conditions_to_run = [condition for condition in eval_conditions if condition in indices_by_split_gene[EVAL_SPLIT]]\n", + "if not eval_conditions_to_run:\n", + " raise ValueError(f\"No conditions found for EVAL_SPLIT={EVAL_SPLIT}.\")\n", + "\n", + "metrics, pred, real = evaluate_gene_metrics(\n", + " split_name=EVAL_SPLIT,\n", + " conditions=eval_conditions_to_run,\n", + " seed=RANDOM_SEED + 999,\n", + ")\n", + "aggregate = build_aggregate_frame(metrics)\n", + "aggregate_summary = pd.Series(\n", + " {\n", + " \"method_name\": METHOD_NAME,\n", + " \"input_path\": str(INPUT_PATH),\n", + " \"eval_split\": EVAL_SPLIT,\n", + " \"train_split\": TRAIN_SPLIT,\n", + " \"val_split\": VAL_SPLIT,\n", + " \"n_control_cells_train\": int(len(control_train_idx)),\n", + " \"n_eval_cells\": int(pred.n_obs),\n", + " \"n_train_genes\": int(len(train_conditions)),\n", + " \"n_eval_genes\": int(len(eval_conditions_to_run)),\n", + " \"n_eval_genes_missing_from_train\": int(len(set(eval_conditions_to_run) - set(train_conditions))),\n", + " \"mse_mean_avg\": float(metrics[\"mse_mean\"].mean()),\n", + " \"mse_cell_avg\": float(metrics[\"mse_cell\"].mean()) if \"mse_cell\" in metrics.columns else float(\"nan\"),\n", + " \"mae_mean_avg\": float(metrics[\"mae_mean\"].mean()),\n", + " \"pearson_mean_avg\": float(metrics[\"pearson_mean\"].mean(skipna=True)),\n", + " \"spearman_mean_avg\": float(metrics[\"spearman_mean\"].mean(skipna=True)) if \"spearman_mean\" in metrics.columns else float(\"nan\"),\n", + " \"pearson_delta_avg\": float(metrics[\"pearson_delta\"].mean(skipna=True)),\n", + " \"delta_l2_avg\": float(metrics[\"delta_l2\"].mean()),\n", + " }\n", + ")\n", + "\n", + "tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "pred_path = OUTPUT_DIR / f\"pred_{METHOD_NAME}_{tag}.h5ad\"\n", + "real_path = OUTPUT_DIR / f\"real_{METHOD_NAME}_{tag}.h5ad\"\n", + "metrics_path = OUTPUT_DIR / f\"metrics_by_gene_{METHOD_NAME}_{tag}.csv\"\n", + "aggregate_path = OUTPUT_DIR / f\"aggregate_{METHOD_NAME}_{tag}.csv\"\n", + "summary_path = OUTPUT_DIR / f\"summary_{METHOD_NAME}_{tag}.csv\"\n", + "history_path = OUTPUT_DIR / f\"training_history_{METHOD_NAME}_{tag}.csv\"\n", + "config_path = OUTPUT_DIR / f\"config_{METHOD_NAME}_{tag}.json\"\n", + "\n", + "config = {\n", + " \"name\": METHOD_NAME,\n", + " \"input_path\": str(INPUT_PATH),\n", + " \"split_strategy\": SPLIT_STRATEGY,\n", + " \"eval_split\": EVAL_SPLIT,\n", + " \"random_seed\": int(RANDOM_SEED),\n", + " \"latent_dim\": int(LATENT_DIM),\n", + " \"hidden_dim\": int(HIDDEN_DIM),\n", + " \"condition_dim\": int(CONDITION_DIM),\n", + " \"dropout\": float(DROPOUT),\n", + " \"num_transport_blocks\": int(NUM_TRANSPORT_BLOCKS),\n", + " \"batch_size\": int(BATCH_SIZE),\n", + " \"epochs\": int(EPOCHS),\n", + " \"steps_per_epoch\": int(STEPS_PER_EPOCH),\n", + " \"validation_steps\": int(VALIDATION_STEPS),\n", + " \"learning_rate\": float(LEARNING_RATE),\n", + " \"weight_decay\": float(WEIGHT_DECAY),\n", + " \"recon_weight\": float(RECON_WEIGHT),\n", + " \"map_weight\": float(MAP_WEIGHT),\n", + " \"latent_mmd_weight\": float(LATENT_MMD_WEIGHT),\n", + " \"expression_mmd_weight\": float(EXPRESSION_MMD_WEIGHT),\n", + " \"mean_weight\": float(MEAN_WEIGHT),\n", + " \"displacement_reg\": float(DISPLACEMENT_REG),\n", + " \"sinkhorn_epsilon\": float(SINKHORN_EPSILON),\n", + " \"sinkhorn_iters\": int(SINKHORN_ITERS),\n", + " \"train_frac\": float(TRAIN_FRAC),\n", + " \"val_frac\": float(VAL_FRAC),\n", + " \"n_train_conditions\": int(len(train_conditions)),\n", + " \"n_eval_conditions\": int(len(eval_conditions_to_run)),\n", + " \"n_unseen_eval_conditions\": int(len(set(eval_conditions_to_run) - set(train_conditions))),\n", + " \"model_path\": str(MODEL_PATH),\n", + " \"note\": \"CellOT-inspired direct neural OT baseline with Sinkhorn barycentric supervision; not an exact reproduction of official dual-ICNN CellOT training\",\n", + "}\n", + "\n", + "pred.uns[\"baseline\"] = config\n", + "real.uns[\"baseline\"] = config\n", + "\n", + "torch.save(model.state_dict(), MODEL_PATH)\n", + "pred.write_h5ad(pred_path, compression=\"gzip\")\n", + "real.write_h5ad(real_path, compression=\"gzip\")\n", + "metrics.to_csv(metrics_path, index=False)\n", + "aggregate.to_csv(aggregate_path)\n", + "aggregate_summary.to_frame(name=\"value\").to_csv(summary_path)\n", + "history.to_csv(history_path, index=False)\n", + "with config_path.open(\"w\") as f:\n", + " json.dump(config, f, indent=2)\n", + "\n", + "print(\"Aggregate summary:\")\n", + "print(aggregate_summary)\n", + "print(\"\\nSaved:\")\n", + "for path in [MODEL_PATH, pred_path, real_path, metrics_path, aggregate_path, summary_path, history_path, config_path]:\n", + " print(\"-\", path)\n", + "\n", + "metrics.head()\n" + ] + }, + { + "cell_type": "markdown", + "id": "eb993237", + "metadata": {}, + "source": [ + "## Figures — held-out evaluation\n", + "\n", + "Per-perturbation metrics (MSE vs Pearson, ranked Pearson, delta L2, mean vs delta correlation). Seen vs unseen training genes are colored when both cohorts exist.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 19, + "id": "724eaa4c", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "_fig_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "plot_evaluation_metrics(metrics, tag=_fig_tag, train_genes=train_conditions)\n" + ] + }, + { + "cell_type": "markdown", + "id": "bdbcab5f", + "metadata": {}, + "source": [ + "## Extended biological analysis figures\n", + "\n", + "Gene-level calibration, MA plot, perturbation heatmaps / similarity, cell-space PCA, error distributions, and high-variance gene scatters (same layout as `flow_matching.ipynb`). Requires `pred`, `real`, `adata`, `metrics`, `control_mean`, and `eval_conditions_to_run`.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 20, + "id": "84b4b7a2", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.\u001b[0m\u001b[33m\n", + "\u001b[0m" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Extended biological figures saved under: /root/final/baseline/outputs/cellot_ot_baseline/figures\n" + ] + } + ], + "source": [ + "! pip install -q scikit-learn\n", + "\n", + "from sklearn.decomposition import PCA\n", + "from scipy.cluster import hierarchy\n", + "from scipy.spatial.distance import pdist\n", + "\n", + "\n", + "def _as_dense_rows(X, max_rows: int | None = None, rng: np.random.Generator | None = None) -> np.ndarray:\n", + " if sp.issparse(X):\n", + " arr = X.toarray()\n", + " else:\n", + " arr = np.asarray(X, dtype=np.float64)\n", + " if max_rows is not None and arr.shape[0] > max_rows:\n", + " rng = rng or np.random.default_rng(0)\n", + " idx = rng.choice(arr.shape[0], size=max_rows, replace=False)\n", + " arr = arr[idx]\n", + " return arr\n", + "\n", + "\n", + "def _group_mean_matrix(\n", + " X,\n", + " obs_series: pd.Series,\n", + " conditions: list[str],\n", + " max_conditions: int = 48,\n", + ") -> tuple[np.ndarray, list[str]]:\n", + " rows = []\n", + " labels: list[str] = []\n", + " for c in conditions[:max_conditions]:\n", + " m = obs_series.astype(str).values == c\n", + " if m.sum() < 1:\n", + " continue\n", + " block = X[m]\n", + " if sp.issparse(block):\n", + " v = np.asarray(block.mean(axis=0)).ravel()\n", + " else:\n", + " v = np.asarray(block, dtype=np.float64).mean(axis=0)\n", + " rows.append(v)\n", + " labels.append(c)\n", + " if not rows:\n", + " return np.zeros((0, 0)), []\n", + " return np.stack(rows, axis=0), labels\n", + "\n", + "\n", + "_bio_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "_rng_bio = np.random.default_rng(RANDOM_SEED + 8800)\n", + "setup_science_style()\n", + "\n", + "gene_names = np.array(adata.var_names.astype(str))\n", + "ctrl = np.asarray(control_mean, dtype=np.float64).ravel()\n", + "\n", + "real_mean_gene = np.asarray(real.X.mean(axis=0)).ravel()\n", + "pred_mean_gene = np.asarray(pred.X.mean(axis=0)).ravel()\n", + "\n", + "# --- Fig A: global gene-mean calibration (log1p) ---\n", + "fig, ax = plt.subplots(figsize=(3.2, 3.2), constrained_layout=True)\n", + "xr = np.log1p(real_mean_gene)\n", + "xp = np.log1p(pred_mean_gene)\n", + "ax.scatter(xr, xp, s=4, alpha=0.35, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\", rasterized=True)\n", + "lim = [float(np.min([xr.min(), xp.min()])), float(np.max([xr.max(), xp.max()]))]\n", + "ax.plot(lim, lim, ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.9)\n", + "ax.set_xlabel(\"log1p(mean expression), held-out truth\")\n", + "ax.set_ylabel(\"log1p(mean expression), predicted\")\n", + "ax.set_title(\"Gene-wise mean calibration\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_A_gene_mean_calibration_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_A_gene_mean_calibration_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig B: MA-style (mean abundance vs log2 ratio) ---\n", + "eps = 1e-6\n", + "M = 0.5 * (real_mean_gene + pred_mean_gene)\n", + "A = np.log2((pred_mean_gene + eps) / (real_mean_gene + eps))\n", + "ok = np.isfinite(A) & np.isfinite(M)\n", + "fig, ax = plt.subplots(figsize=(3.4, 3.0), constrained_layout=True)\n", + "ax.scatter(M[ok], A[ok], s=3, alpha=0.3, c=SCIENCE_COLORS[\"orange\"], edgecolors=\"none\", rasterized=True)\n", + "ax.axhline(0.0, color=SCIENCE_COLORS[\"gray\"], lw=0.8)\n", + "ax.set_xlabel(\"Mean abundance (avg pred & truth)\")\n", + "ax.set_ylabel(\"log2(pred / truth)\")\n", + "ax.set_title(\"MA-style bias vs abundance\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_B_MA_plot_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_B_MA_plot_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig C: per-gene mean error distribution ---\n", + "ge = pred_mean_gene - real_mean_gene\n", + "fig, ax = plt.subplots(figsize=(3.2, 2.6), constrained_layout=True)\n", + "ax.hist(ge, bins=80, color=SCIENCE_COLORS[\"green\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", + "ax.axvline(0.0, color=SCIENCE_COLORS[\"dark\"], lw=0.9)\n", + "ax.set_xlabel(\"Pred − truth (gene mean)\")\n", + "ax.set_ylabel(\"Genes\")\n", + "ax.set_title(\"Global gene-level error\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_C_gene_error_hist_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_C_gene_error_hist_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig D–E: perturbation × gene heatmaps (delta vs control, z-scored genes) ---\n", + "R_mat, perts = _group_mean_matrix(real.X, real.obs[\"perturbation_gene\"], eval_conditions_to_run, max_conditions=40)\n", + "P_mat, perts_p = _group_mean_matrix(pred.X, pred.obs[\"perturbation_gene\"], eval_conditions_to_run, max_conditions=40)\n", + "if R_mat.size and P_mat.size and R_mat.shape == P_mat.shape:\n", + " d_real = R_mat - ctrl\n", + " d_pred = P_mat - ctrl\n", + " var_g = np.var(d_real, axis=0)\n", + " top_k = min(60, d_real.shape[1])\n", + " gi = np.argsort(var_g)[-top_k:]\n", + "\n", + " def _z_rows(D):\n", + " sub = D[:, gi]\n", + " return (sub - sub.mean(axis=1, keepdims=True)) / (sub.std(axis=1, keepdims=True) + 1e-8)\n", + "\n", + " Zr, Zp = _z_rows(d_real), _z_rows(d_pred)\n", + " for name, Zm in ((\"truth\", Zr), (\"predicted\", Zp)):\n", + " fig, ax = plt.subplots(figsize=(5.5, 4.2), constrained_layout=True)\n", + " im = ax.imshow(Zm, aspect=\"auto\", cmap=\"RdBu_r\", vmin=-2.5, vmax=2.5, interpolation=\"nearest\")\n", + " ax.set_yticks(np.arange(len(perts)))\n", + " ax.set_yticklabels(perts, fontsize=5)\n", + " ax.set_xlabel(\"Genes (high-variance subset, z-scored per perturbation)\")\n", + " ax.set_ylabel(\"Perturbation\")\n", + " ax.set_title(f\"Perturbation effects ({name})\")\n", + " plt.colorbar(im, ax=ax, fraction=0.035, pad=0.02, label=\"Row z-score\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_DE_heatmap_delta_{name}_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_DE_heatmap_delta_{name}_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig F: perturbation–perturbation correlation of delta profiles (truth) ---\n", + "if R_mat.shape[0] >= 3:\n", + " d_real = R_mat - ctrl\n", + " C = np.corrcoef(d_real)\n", + " fig, ax = plt.subplots(figsize=(4.8, 4.2), constrained_layout=True)\n", + " im = ax.imshow(C, cmap=\"RdBu_r\", vmin=-1, vmax=1, aspect=\"auto\")\n", + " ax.set_xticks(np.arange(len(perts)))\n", + " ax.set_xticklabels(perts, rotation=90, fontsize=4.5)\n", + " ax.set_yticks(np.arange(len(perts)))\n", + " ax.set_yticklabels(perts, fontsize=4.5)\n", + " ax.set_title(\"Perturbation similarity (Pearson of Δ vs control)\")\n", + " plt.colorbar(im, ax=ax, fraction=0.046, pad=0.04)\n", + " fig.savefig(FIGURE_DIR / f\"bio_F_perturb_corr_truth_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_F_perturb_corr_truth_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig G: hierarchical clustering on subset of perturbations (truth delta) ---\n", + "if R_mat.shape[0] >= 4:\n", + " d_real = R_mat - ctrl\n", + " sub = min(24, d_real.shape[0])\n", + " cond_dist = pdist(d_real[:sub], metric=\"correlation\")\n", + " y = hierarchy.linkage(cond_dist, method=\"average\")\n", + " order = hierarchy.leaves_list(y)\n", + " labels_sub = [perts[i] for i in order]\n", + " Csub = np.corrcoef(d_real[:sub][order])\n", + " fig, ax = plt.subplots(figsize=(4.5, 4.0), constrained_layout=True)\n", + " im = ax.imshow(Csub, cmap=\"RdBu_r\", vmin=-1, vmax=1)\n", + " ax.set_xticks(np.arange(sub))\n", + " ax.set_xticklabels(labels_sub, rotation=90, fontsize=5)\n", + " ax.set_yticks(np.arange(sub))\n", + " ax.set_yticklabels(labels_sub, fontsize=5)\n", + " ax.set_title(\"Clustered perturbation similarity (subset)\")\n", + " plt.colorbar(im, ax=ax, fraction=0.046, pad=0.04)\n", + " fig.savefig(FIGURE_DIR / f\"bio_G_perturb_corr_clustered_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_G_perturb_corr_clustered_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig H: cell-level PCA (pred vs real, subsampled) ---\n", + "n_sub = min(6000, real.n_obs)\n", + "idx = _rng_bio.choice(real.n_obs, size=n_sub, replace=False) if real.n_obs > n_sub else np.arange(real.n_obs)\n", + "Xr = _as_dense_rows(real.X[idx], max_rows=None)\n", + "Xp = _as_dense_rows(pred.X[idx], max_rows=None)\n", + "lab = real.obs[\"perturbation_gene\"].astype(str).iloc[idx].to_numpy()\n", + "uniq = np.unique(lab)\n", + "if len(uniq) > 24:\n", + " keep = set(_rng_bio.choice(uniq, size=24, replace=False))\n", + " mask = np.array([g in keep for g in lab])\n", + " idx = idx[mask]\n", + " Xr, Xp, lab = Xr[mask], Xp[mask], lab[mask]\n", + "Z = np.vstack([Xr, Xp])\n", + "pca = PCA(n_components=2, random_state=RANDOM_SEED)\n", + "Z2 = pca.fit_transform(Z)\n", + "nr = int(Xr.shape[0])\n", + "Z_truth = Z2[:nr]\n", + "Z_pred = Z2[nr:]\n", + "fig, axes = plt.subplots(1, 2, figsize=(6.8, 3.0), constrained_layout=True)\n", + "for ax, name, Zpart in zip(axes, [\"truth\", \"pred\"], [Z_truth, Z_pred]):\n", + " for j, g in enumerate(np.unique(lab)):\n", + " m = lab == g\n", + " if not m.any():\n", + " continue\n", + " ax.scatter(Zpart[m, 0], Zpart[m, 1], s=3, alpha=0.45, label=g if j < 12 else None)\n", + " ax.set_title(f\"PCA — {name}\")\n", + " ax.set_xlabel(\"PC1\")\n", + " ax.set_ylabel(\"PC2\")\n", + "axes[0].legend(bbox_to_anchor=(1.02, 1), loc=\"upper left\", fontsize=5, markerscale=2)\n", + "fig.suptitle(\"Expression PCA (subsampled cells)\", y=1.02)\n", + "fig.savefig(FIGURE_DIR / f\"bio_H_cell_PCA_truth_pred_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_H_cell_PCA_truth_pred_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig I: per-cell L2 error distribution ---\n", + "diff = _as_dense_rows(pred.X[idx], max_rows=None) - _as_dense_rows(real.X[idx], max_rows=None)\n", + "l2 = np.linalg.norm(diff, axis=1)\n", + "fig, ax = plt.subplots(figsize=(3.2, 2.6), constrained_layout=True)\n", + "ax.hist(l2, bins=60, color=SCIENCE_COLORS[\"magenta\"], alpha=0.88, edgecolor=\"white\", linewidth=0.3)\n", + "ax.set_xlabel(\"L2(pred − truth) per cell\")\n", + "ax.set_ylabel(\"Cells\")\n", + "ax.set_title(\"Cell-level reconstruction error\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_I_cell_L2_error_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_I_cell_L2_error_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig J: metrics overview ---\n", + "if len(metrics) > 0:\n", + " fig, axes = plt.subplots(2, 2, figsize=(6.0, 4.8), constrained_layout=True)\n", + " axes[0, 0].scatter(metrics[\"mse_mean\"], metrics[\"pearson_mean\"], s=12, alpha=0.65, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\")\n", + " axes[0, 0].set_xlabel(\"MSE (means)\")\n", + " axes[0, 0].set_ylabel(\"Pearson r\")\n", + " axes[0, 1].scatter(metrics[\"n_cells\"], metrics[\"delta_l2\"], s=12, alpha=0.65, c=SCIENCE_COLORS[\"green\"], edgecolors=\"none\")\n", + " axes[0, 1].set_xlabel(\"Cells per perturbation\")\n", + " axes[0, 1].set_ylabel(\"Delta L2\")\n", + " axes[1, 0].hist(metrics[\"pearson_delta\"].dropna(), bins=40, color=SCIENCE_COLORS[\"orange\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", + " axes[1, 0].set_xlabel(\"Pearson r (delta)\")\n", + " axes[1, 0].set_ylabel(\"Perturbations\")\n", + " axes[1, 1].hist(metrics[\"mae_mean\"], bins=40, color=SCIENCE_COLORS[\"blue\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", + " axes[1, 1].set_xlabel(\"MAE (means)\")\n", + " axes[1, 1].set_ylabel(\"Perturbations\")\n", + " fig.suptitle(\"Perturbation-level metric panels\", y=1.02)\n", + " fig.savefig(FIGURE_DIR / f\"bio_J_metric_panels_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_J_metric_panels_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig K: high cell-wise variance genes — pred vs truth scatter ---\n", + "if sp.issparse(real.X):\n", + " mu = np.asarray(real.X.mean(axis=0)).ravel()\n", + " mu2 = np.asarray(real.X.power(2).mean(axis=0)).ravel()\n", + " var_per_gene = mu2 - mu**2\n", + "else:\n", + " var_per_gene = np.var(np.asarray(real.X, dtype=np.float64), axis=0)\n", + "top_idx = np.argsort(var_per_gene)[-12:][::-1]\n", + "fig, axes = plt.subplots(3, 4, figsize=(7.2, 5.4), constrained_layout=True)\n", + "axes = axes.ravel()\n", + "for ax, gi in zip(axes, top_idx):\n", + " if sp.issparse(real.X):\n", + " vr = real.X[:, gi].toarray().ravel()\n", + " else:\n", + " vr = np.asarray(real.X[:, gi], dtype=np.float64).ravel()\n", + " if sp.issparse(pred.X):\n", + " vp = pred.X[:, gi].toarray().ravel()\n", + " else:\n", + " vp = np.asarray(pred.X[:, gi], dtype=np.float64).ravel()\n", + " if vr.size > 8000:\n", + " si = _rng_bio.choice(vr.size, size=8000, replace=False)\n", + " vr, vp = vr[si], vp[si]\n", + " ax.scatter(vr, vp, s=2, alpha=0.25, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\", rasterized=True)\n", + " lim = [min(vr.min(), vp.min()), max(vr.max(), vp.max())]\n", + " ax.plot(lim, lim, ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.7)\n", + " ax.set_title(gene_names[gi], fontsize=7)\n", + " ax.set_xlabel(\"Truth\", fontsize=6)\n", + " ax.set_ylabel(\"Pred\", fontsize=6)\n", + "fig.suptitle(\"High-variance genes: cell-level pred vs truth\", y=1.01)\n", + "fig.savefig(FIGURE_DIR / f\"bio_K_topvar_gene_scatters_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_K_topvar_gene_scatters_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "print(\"Extended biological figures saved under:\", FIGURE_DIR)\n" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.13" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} diff --git a/baseline/outputs/cellot_ot_baseline/aggregate_cellot_ot_baseline_use_existing_test.csv b/baseline/outputs/cellot_ot_baseline/aggregate_cellot_ot_baseline_use_existing_test.csv new file mode 100644 index 0000000000000000000000000000000000000000..30cb7e18188e0b91821fb07cae663c8bf1fff693 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/aggregate_cellot_ot_baseline_use_existing_test.csv @@ -0,0 +1,9 @@ +,mean,median +n_cells,162.1219512195122,141.0 +mse_mean,0.005047890109225835,0.0027004487346857786 +mse_cell,0.14435312380877938,0.13683290779590607 +mae_mean,0.04422009014865247,0.03574778512120247 +pearson_mean,0.9934415852860539,0.9964321144783835 +spearman_mean,0.9735030627041961,0.9831866522675426 +pearson_delta,0.14896064115191923,0.12608604794350597 +delta_l2,4.179004827359828,3.4210736751556396 diff --git a/baseline/outputs/cellot_ot_baseline/cellot_ot_baseline_model.pt b/baseline/outputs/cellot_ot_baseline/cellot_ot_baseline_model.pt new file mode 100644 index 0000000000000000000000000000000000000000..241ba6068f55d43a61ce0ca19011df58282e2c5a --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/cellot_ot_baseline_model.pt @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8529c08923ee446addedf075efd4bf4fbd82344762f28cde8042e27254779920 +size 24041641 diff --git a/baseline/outputs/cellot_ot_baseline/config_cellot_ot_baseline_use_existing_test.json b/baseline/outputs/cellot_ot_baseline/config_cellot_ot_baseline_use_existing_test.json new file mode 100644 index 0000000000000000000000000000000000000000..53517aca70b1943d2f30d959123b63a66f6c58c5 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/config_cellot_ot_baseline_use_existing_test.json @@ -0,0 +1,33 @@ +{ + "name": "cellot_ot_baseline", + "input_path": "/root/final/data/processed/scPerturb/prediction_ready/replogle_k562_essential_single_gene_prediction_ready_full.h5ad", + "split_strategy": "use_existing", + "eval_split": "test", + "random_seed": 42, + "latent_dim": 64, + "hidden_dim": 256, + "condition_dim": 128, + "dropout": 0.1, + "num_transport_blocks": 4, + "batch_size": 128, + "epochs": 20, + "steps_per_epoch": 200, + "validation_steps": 32, + "learning_rate": 0.001, + "weight_decay": 1e-05, + "recon_weight": 1.0, + "map_weight": 1.0, + "latent_mmd_weight": 0.1, + "expression_mmd_weight": 0.1, + "mean_weight": 1.0, + "displacement_reg": 0.01, + "sinkhorn_epsilon": 0.5, + "sinkhorn_iters": 30, + "train_frac": 0.8, + "val_frac": 0.1, + "n_train_conditions": 1645, + "n_eval_conditions": 205, + "n_unseen_eval_conditions": 205, + "model_path": "/root/final/baseline/outputs/cellot_ot_baseline/cellot_ot_baseline_model.pt", + "note": "CellOT-inspired direct neural OT baseline with Sinkhorn barycentric supervision; not an exact reproduction of official dual-ICNN CellOT training" +} \ No newline at end of file diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_A_gene_mean_calibration_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_A_gene_mean_calibration_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..54e431551273c126d5a95293628c1c973c82c4fa Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_A_gene_mean_calibration_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_A_gene_mean_calibration_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_A_gene_mean_calibration_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..3b6b327f1de6936092dfd334123e696dd6e507d6 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_A_gene_mean_calibration_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:7c70b9bcd55c6441032bcc727b4d74afa91c119a2be27b68c70f1575d5ed6cd4 +size 115861 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_B_MA_plot_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_B_MA_plot_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..0a6554420415b1e00b3417730203da498dfdc03e Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_B_MA_plot_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_B_MA_plot_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_B_MA_plot_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..bdf2f592dee99b6fb43c527dda656c795b1871ec --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_B_MA_plot_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:9c408b2a3b0500f35e02850566979288284e85c01a7242c6d05f11197f244da6 +size 167121 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_C_gene_error_hist_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_C_gene_error_hist_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..a1d1e7c1be0ba2e05480c9fbf7d5153aa5373ee6 Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_C_gene_error_hist_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_C_gene_error_hist_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_C_gene_error_hist_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..08c8a2d481e695142d3de8716704d6f831856c95 Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_C_gene_error_hist_use_existing_test.png differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_predicted_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_predicted_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..93d1959858ecf5de6648b60a806601aa154c8076 Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_predicted_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_predicted_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_predicted_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..61862188ab532bfa50ed50c67b0ff1531449c213 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_predicted_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8f483433e215883eeab39d676f351c3aa8498b5f067f03e748996bb1554bd0a4 +size 140962 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_truth_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_truth_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..3fbcac4929f94646c2590f7a04faefb5fc967b1e Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_truth_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_truth_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_truth_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..5580c3bc96cf8281a3c2788a5f5de0a25de0c62f --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_DE_heatmap_delta_truth_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:63e4a3269822f7d795e499a8d98de0bc1860245e102d8791cabd8c051694d8dd +size 153164 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_F_perturb_corr_truth_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_F_perturb_corr_truth_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..b05d56fba08bd4b01fba9db1b5dd6bf5509d6bcd Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_F_perturb_corr_truth_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_F_perturb_corr_truth_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_F_perturb_corr_truth_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..eee154ec58723e222f31c6aeb63b75f8bd153b0e --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_F_perturb_corr_truth_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:beeedaedad00cfc6273b59f2b962a0b6b04494a6b706437dd3dce1d98cfdf961 +size 161519 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_G_perturb_corr_clustered_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_G_perturb_corr_clustered_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..9cdf13bb430cb1ca3018ec9bb2a5cb868759a559 Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_G_perturb_corr_clustered_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_G_perturb_corr_clustered_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_G_perturb_corr_clustered_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..a79e1e6f3346614ce6d9a0c30400cfd0edfa50e7 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_G_perturb_corr_clustered_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:bb4fa47ca780812d27c365178701f03bccc78371bea490c8e68134aac9969d91 +size 131557 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_H_cell_PCA_truth_pred_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_H_cell_PCA_truth_pred_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..7eecc935fc047ed33ea7a09f721d4c4caeb91b0d Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_H_cell_PCA_truth_pred_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_H_cell_PCA_truth_pred_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_H_cell_PCA_truth_pred_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..85d98bb2871bac0da282d0dbfbd82de60cf336d2 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_H_cell_PCA_truth_pred_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:23dd4107152d768240a84f2849313d19e9c55e2ad50fd9b0e6d6d2d59ab8a4f4 +size 364729 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_I_cell_L2_error_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_I_cell_L2_error_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..9bb37d2cb15761e8a6e3b612c8f277beb511b752 Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_I_cell_L2_error_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_I_cell_L2_error_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_I_cell_L2_error_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..28322caeccdfbdd6f13bd0a81809c72b5058b574 Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_I_cell_L2_error_use_existing_test.png differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_J_metric_panels_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_J_metric_panels_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..a4e00c998419f24dad9cd83e3dd3eed48a9728ad Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_J_metric_panels_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_J_metric_panels_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_J_metric_panels_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..e00a957d832f5bb67b2e59e23d111dac78808334 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_J_metric_panels_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:c16b65d9d471cf3233b7b142306e6d18b1a27051dc8dbb26e75893640700259f +size 188310 diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_K_topvar_gene_scatters_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/bio_K_topvar_gene_scatters_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..8d0d8a403ba0262d2527c18714257da12ebcea78 Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/bio_K_topvar_gene_scatters_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/bio_K_topvar_gene_scatters_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/bio_K_topvar_gene_scatters_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..585bcf81cc17e9addf925eb6670d2053e714eedd --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/bio_K_topvar_gene_scatters_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:8ca39c708f7f84170ccb4e1df01767b2e538673e1fc1634e3c06aa9141d40ad2 +size 217856 diff --git a/baseline/outputs/cellot_ot_baseline/figures/evaluation_metrics_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/evaluation_metrics_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..f80f41b967a442c1a041e399b8c5471197d8b86e Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/evaluation_metrics_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/evaluation_metrics_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/evaluation_metrics_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..396e4c47e1b33e666f6294aac48b42f899007084 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/evaluation_metrics_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:f80af7db0e2054b08c98b119e0d85808e2a6d03ab5f320a3f920fabdef9fcdac +size 285963 diff --git a/baseline/outputs/cellot_ot_baseline/figures/training_history_use_existing_test.pdf b/baseline/outputs/cellot_ot_baseline/figures/training_history_use_existing_test.pdf new file mode 100644 index 0000000000000000000000000000000000000000..620ae90915f886dd3570fc2ff4036dea2bd6597d Binary files /dev/null and b/baseline/outputs/cellot_ot_baseline/figures/training_history_use_existing_test.pdf differ diff --git a/baseline/outputs/cellot_ot_baseline/figures/training_history_use_existing_test.png b/baseline/outputs/cellot_ot_baseline/figures/training_history_use_existing_test.png new file mode 100644 index 0000000000000000000000000000000000000000..98397e0b0c9d845c1d4146db518f051fbe51690b --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/figures/training_history_use_existing_test.png @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:0850c23cfcd8372e6263a2f1fa0a86fb5ae8bbd7718da8e5dd4b6233ebd64fbb +size 167639 diff --git a/baseline/outputs/cellot_ot_baseline/metrics_by_gene_cellot_ot_baseline_use_existing_test.csv b/baseline/outputs/cellot_ot_baseline/metrics_by_gene_cellot_ot_baseline_use_existing_test.csv new file mode 100644 index 0000000000000000000000000000000000000000..b7a6fcdd90189514e577476643b499a3d138af9f --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/metrics_by_gene_cellot_ot_baseline_use_existing_test.csv @@ -0,0 +1,206 @@ +perturbation_gene,n_cells,mse_mean,mse_cell,mae_mean,pearson_mean,spearman_mean,pearson_delta,delta_l2 +ADSL,133,0.0017661136807873845,0.12744224071502686,0.030697986483573914,0.997669197966977,0.9865178236142582,0.07872325692524042,2.7666473388671875 +ANAPC2,237,0.0014298363821581006,0.13920311629772186,0.02664765529334545,0.9981431016235106,0.991327497127016,0.18720066482168,2.4893593788146973 +ANAPC5,40,0.004295671358704567,0.13281328976154327,0.04924652352929115,0.9942879316879918,0.9662004951482234,0.006583867197388726,4.314793109893799 +ANKRD11,260,0.0024251569993793964,0.11459104716777802,0.03167976066470146,0.9972802260142666,0.9902751539122008,-0.08329096237938384,3.2420103549957275 +ARL2,74,0.004299957305192947,0.16495579481124878,0.047481197863817215,0.99478238388988,0.9710085305866951,0.26080691090713076,4.3169450759887695 +ARPC3,141,0.0014289122773334384,0.1306191384792328,0.027897432446479797,0.9981575765242124,0.9887370318146942,0.0321144727944707,2.4885549545288086 +ATP6V0D1,101,0.0019260409753769636,0.1341322511434555,0.032743729650974274,0.9974887539048924,0.9847058927150046,0.03534939702778406,2.88919734954834 +BCAS2,227,0.006920636165887117,0.14888356626033783,0.055636629462242126,0.9906697211817724,0.9661499227393787,0.18942832648416497,5.476681232452393 +BRF1,166,0.0014903609408065677,0.1350143551826477,0.02836649678647518,0.9980722370067827,0.988748727321471,0.14426499928963737,2.5415005683898926 +C14orf178,143,0.008575747720897198,0.17134767770767212,0.06467457860708237,0.9889894897336414,0.9605766483476067,0.2779514504017412,6.096498012542725 +C19orf53,169,0.0016915039159357548,0.12949423491954803,0.029037799686193466,0.9978802989648229,0.98855218760497,-0.06397088773356434,2.70757794380188 +C1QBP,210,0.002544180490076542,0.14237827062606812,0.03196968510746956,0.9966353281372977,0.9874384464501853,0.32813814665029295,3.3206140995025635 +C1orf109,240,0.0011465903371572495,0.1313076764345169,0.024479122832417488,0.9985713235829958,0.9922138332656585,-0.02700172274246734,2.2291977405548096 +CBLN1,214,0.0012490585213527083,0.13123108446598053,0.025913914665579796,0.9983831864116096,0.9907052053347482,0.1440124131944915,2.3266756534576416 +CCDC115,211,0.0011965250596404076,0.12892168760299683,0.025369135662913322,0.9985166197511018,0.9919819792852197,-0.0006775050169476909,2.2772216796875 +CCDC144NL,87,0.0018458969425410032,0.12408580631017685,0.032335687428712845,0.9976276257537765,0.984609557281176,0.06112303833980765,2.8284475803375244 +CCDC59,125,0.0030449458863586187,0.14173531532287598,0.03888615965843201,0.9959185540210771,0.98484403154698,0.10542056112365672,3.632739305496216 +CCNK,327,0.0016306305769830942,0.1348920613527298,0.026348678395152092,0.9977772663292102,0.991169043418431,0.09390259613825225,2.658411741256714 +CCP110,47,0.0031258640810847282,0.12414247542619705,0.042774032801389694,0.9959762014343746,0.9741507944384046,-0.018015824807491622,3.680692195892334 +CCT2,199,0.006967850960791111,0.15181170403957367,0.0541812926530838,0.9905146520969133,0.9659149784346119,0.2514067861962013,5.495331287384033 +CDC42,70,0.0027139612939208746,0.12505072355270386,0.03755507618188858,0.9964321144783835,0.9788855046357957,0.01314456370372874,3.429621934890747 +CDC7,95,0.0026538220699876547,0.1312597244977951,0.03636082261800766,0.9963707078714948,0.9812762711409124,0.09537766503738059,3.3914105892181396 +CEP192,202,0.0014597282279282808,0.14103686809539795,0.02709212154150009,0.9980443231610533,0.9900836556204078,0.2827796961715207,2.5152461528778076 +CHCHD2,90,0.002766229910776019,0.13426372408866882,0.03302137926220894,0.9963646151090937,0.984619747480988,0.08733648451738993,3.4624907970428467 +CHD4,44,0.006420290097594261,0.13370640575885773,0.05476275831460953,0.9913631109173782,0.9570802822228256,0.09463732069775557,5.274991512298584 +CHMP3,130,0.024655617773532867,0.1844552606344223,0.09974869340658188,0.9661620794421983,0.9027634252545715,0.5599566057582943,10.337187767028809 +CHMP4B,71,0.002232175786048174,0.1382192075252533,0.03589479252696037,0.9969939017678453,0.9791689137558742,0.18624827491194992,3.1103456020355225 +CHMP6,624,0.0037591843865811825,0.1461392492055893,0.03962044045329094,0.995003786714274,0.9793464374991617,0.544636137502501,4.036372661590576 +COPA,243,0.002029828727245331,0.14688663184642792,0.03192514553666115,0.9972889782931588,0.9872382943336936,0.4366904113646547,2.966020345687866 +COPZ1,207,0.002460768446326256,0.15398624539375305,0.03450104221701622,0.9966998101454565,0.9853311549179168,0.4510858613768957,3.2657265663146973 +COX11,109,0.0024743203539401293,0.13284458220005035,0.03548481687903404,0.9968284676786623,0.9823685959657309,0.1064605795750759,3.2747068405151367 +COX7B,120,0.002134365029633045,0.1293608695268631,0.031229211017489433,0.9972225776286817,0.9863838663897924,0.1435569164463171,3.0414366722106934 +CTPS1,14,0.013480032794177532,0.14611385762691498,0.0874410942196846,0.9825184272061912,0.9090276964356722,0.11871378857612308,7.643458843231201 +DCTN1,197,0.0011717821471393108,0.13824331760406494,0.025030942633748055,0.9984346702228624,0.9906226794286216,0.13852250238460356,2.253553628921509 +DDX20,114,0.0035337088629603386,0.1393408179283142,0.04060862958431244,0.9954120088271297,0.9794479165106892,0.1259596797663313,3.913450241088867 +DDX24,286,0.005610427353531122,0.1422639787197113,0.050002146512269974,0.9926493939991011,0.9769911575381753,0.2701590408456322,4.931084156036377 +DDX49,45,0.0053138467483222485,0.1412876844406128,0.05313730612397194,0.9927410743704342,0.9669404630465763,0.02622487407744086,4.798980236053467 +DDX5,151,0.0027004487346857786,0.13214854896068573,0.03525342419743538,0.9967048619375283,0.9853456365425216,-0.07623358271263606,3.4210736751556396 +DDX59,154,0.0020989764016121626,0.13404583930969238,0.032173700630664825,0.9972589106003733,0.985041053427697,0.013898071020855406,3.016117572784424 +DHX15,359,0.015164013020694256,0.1683514416217804,0.08242012560367584,0.9800254378226545,0.9341498075140401,0.40784292404169004,8.106837272644043 +DNMT1,170,0.0024855018127709627,0.12853647768497467,0.033864300698041916,0.9967919930829598,0.9837597712193331,-0.11616361043274678,3.282097816467285 +DPH6,101,0.0018826111918315291,0.1257137954235077,0.03266622871160507,0.9976606482910401,0.9859340446441341,0.005712410005376584,2.8564376831054688 +EBNA1BP2,82,0.006114142946898937,0.1364148110151291,0.054445795714855194,0.9919735046915537,0.9698719513380586,0.12608604794350597,5.147688388824463 +EIF3G,67,0.007292256224900484,0.15298324823379517,0.05928192287683487,0.9900183222635421,0.9616289186319925,0.24133477737722925,5.621799945831299 +EIF3I,61,0.006233828607946634,0.13744239509105682,0.05483521148562431,0.9916898436270251,0.9662755386029921,0.201349057791813,5.1978278160095215 +ELP3,119,0.0024210396222770214,0.12640349566936493,0.03585690259933472,0.9970166424194226,0.9854325496507234,0.02988866859638767,3.2392570972442627 +EXOSC2,226,0.014247632585465908,0.17851920425891876,0.06913814693689346,0.9806800677872978,0.9381941514084096,0.19848630335348075,7.858068466186523 +EXOSC9,40,0.01588987372815609,0.19638590514659882,0.08436562865972519,0.9782907962916211,0.9115001507014916,0.2093972728385314,8.29859733581543 +FAF2,165,0.0015688055427744985,0.13051456212997437,0.027857907116413116,0.9980924810031497,0.9893891746794409,-0.0680083491122089,2.6075282096862793 +FAM207A,103,0.0017971262568607926,0.13203619420528412,0.03150390833616257,0.9976180697295023,0.9855966808407889,0.0995496936919802,2.79083251953125 +FARSB,97,0.010146849788725376,0.15910497307777405,0.06753779202699661,0.9868522594774115,0.9593475010599333,0.26029131450817783,6.631474494934082 +FCF1,231,0.008798979222774506,0.1551915407180786,0.06333712488412857,0.9889089213505151,0.9736863635353125,0.1640425154774112,6.1753363609313965 +FNBP4,160,0.0018815292278304696,0.1313140094280243,0.03045640140771866,0.997715174361387,0.9880952984886195,0.06274881814952356,2.855616807937622 +FOLR3,304,0.001431640237569809,0.13272389769554138,0.02570345066487789,0.9980698895352877,0.989850396235915,0.22363819415049793,2.490929365158081 +GABPA,110,0.0018741213716566563,0.14342667162418365,0.03262556344270706,0.9975089457634917,0.9835868068108852,0.16153832286770065,2.849989652633667 +GATA1,108,0.05218387022614479,0.32615262269973755,0.14879055321216583,0.9327013553832422,0.8290302058977023,0.4524889246259503,15.038779258728027 +GGPS1,167,0.0016154370969161391,0.12187512218952179,0.029473328962922096,0.9980561793802981,0.9892020894594556,-0.1937123991844902,2.6459977626800537 +GPR61,112,0.0016598730580881238,0.12881238758563995,0.030745763331651688,0.9978760389295155,0.9865733575671061,-0.02946138123351643,2.682142734527588 +GSDMA,110,0.001934427535161376,0.1235177293419838,0.03252508491277695,0.9976165302463996,0.9858383290516639,-0.11231907936620045,2.8954808712005615 +GTF2F2,59,0.0070544658228755,0.15471288561820984,0.05942796915769577,0.9905278033050201,0.9572070241322846,0.27698053513321763,5.529381275177002 +GTF2H1,434,0.005459043197333813,0.16733860969543457,0.04978045076131821,0.9930587590559847,0.9758381489918039,0.39452559650738744,4.864102363586426 +HAUS3,164,0.0013184043345972896,0.13196636736392975,0.026668092235922813,0.9982530869630813,0.9896694511304819,0.045188161187821096,2.390390157699585 +HAUS7,239,0.0010190762113779783,0.13025915622711182,0.023447517305612564,0.9987089970511837,0.9926446617750319,0.0364633701369992,2.1015892028808594 +HGS,99,0.0070278337225317955,0.1607460230588913,0.05864536389708519,0.9907258184063906,0.9616538773452378,0.40619770515353665,5.5189337730407715 +HIST1H2BM,156,0.0012796660885214806,0.13520513474941254,0.026668673381209373,0.9983150599723014,0.9892288594940556,0.08047225055739328,2.3550102710723877 +HSD17B10,26,0.0072830053977668285,0.13736404478549957,0.06158268451690674,0.9907948405131336,0.9511932262355229,0.001739236514754079,5.6182332038879395 +HSPA8,255,0.006017681211233139,0.16174380481243134,0.05067957937717438,0.9926199874772748,0.9789354381627665,0.22277392064031606,5.106919765472412 +HUS1,201,0.0014657338615506887,0.12314032763242722,0.027521485462784767,0.9982487816726522,0.9911280283464611,-0.1270441473916655,2.5204145908355713 +INTS7,79,0.01253111008554697,0.17946301400661469,0.07625474780797958,0.982951587913768,0.9359199973696616,0.30726704970873014,7.369520664215088 +INTS8,323,0.020728949457406998,0.2105289101600647,0.0952417403459549,0.9717466196683738,0.9111008385750851,0.3653525470112931,9.478358268737793 +ISG20L2,73,0.00653561856597662,0.13943590223789215,0.05634218081831932,0.9914233402500904,0.9662078487197164,0.09415416909407141,5.3221588134765625 +KATNB1,178,0.0011153966188430786,0.140666663646698,0.02513321489095688,0.998528701233729,0.99064762597913,0.19653055319492793,2.198665142059326 +KCMF1,154,0.0015989101957529783,0.13299626111984253,0.02920498698949814,0.9979200056357902,0.988240614361191,0.18142307131382057,2.632427930831909 +KIF18B,122,0.002271506004035473,0.1283552646636963,0.034140314906835556,0.9970663388972202,0.9854950321651951,0.025572348136294474,3.137627601623535 +KIN,160,0.003019525669515133,0.14185041189193726,0.03922407329082489,0.9959607312461445,0.9831866522675426,0.21729497338931558,3.617543935775757 +LETM1,124,0.0029627536423504353,0.1419290155172348,0.037490032613277435,0.9960944525002206,0.9824095394359874,0.16909504898921532,3.5833747386932373 +LIMS1,287,0.0011049542808905244,0.12537571787834167,0.024139055982232094,0.9987087531390788,0.9935700849178861,-0.015812127296581477,2.1883492469787598 +LIN52,13,0.0112039465457201,0.1359524130821228,0.08135876804590225,0.9849142071836379,0.9138757008913276,0.09926125112054912,6.968349933624268 +LRPPRC,110,0.006397755816578865,0.15155375003814697,0.0418226420879364,0.9924250599260653,0.9836770087259631,0.07026374651901868,5.265726566314697 +LSM11,130,0.004379140678793192,0.14585652947425842,0.041790809482336044,0.9941932271506233,0.9770009769750021,0.20324201730119662,4.356512069702148 +LSM2,357,0.011429285630583763,0.15536615252494812,0.07170885801315308,0.9854294907685852,0.9512987547517268,0.24540413047660906,7.038076877593994 +LURAP1,71,0.0021448968909680843,0.13231830298900604,0.03547605127096176,0.9971546588094137,0.9807191357676517,0.08371624561497806,3.04893159866333 +MCM7,160,0.0018211633432656527,0.12848690152168274,0.029184289276599884,0.9975456284970877,0.9873712676972705,0.09428382456982845,2.809434175491333 +MDN1,84,0.0037406855262815952,0.12896455824375153,0.043805673718452454,0.995158700783148,0.9768591366361918,0.04892961913707988,4.026429176330566 +MED12,573,0.02325749211013317,0.2437211573123932,0.10226873308420181,0.9708153728307105,0.9081736461115851,0.47872021598861897,10.03981876373291 +MED22,289,0.013430985622107983,0.2145242840051651,0.07561123371124268,0.9822323893246631,0.9437882376937288,0.4646162274966355,7.629540920257568 +MED28,73,0.013839105144143105,0.20468610525131226,0.08160752803087234,0.9820735548848679,0.930594062630581,0.4173989456796667,7.744590759277344 +MED31,38,0.005462013650685549,0.14809493720531464,0.05499592423439026,0.9924736899111316,0.9570964999569236,0.1685123367500524,4.8654255867004395 +MED6,246,0.012198859825730324,0.20895402133464813,0.0744771882891655,0.9851118107233775,0.9478841573656932,0.45900540807505996,7.27116584777832 +MIOS,163,0.0020752844866365194,0.13216784596443176,0.032703500241041183,0.9973450889222535,0.9863268153365345,0.17680569348009903,2.999047040939331 +MIS18A,129,0.001516127260401845,0.13227993249893188,0.02923048473894596,0.9980400767375601,0.9867053899667413,0.030617828262013147,2.56337571144104 +MKRN1,160,0.0014780416386201978,0.13176579773426056,0.028198111802339554,0.9980988742320019,0.9895199303238147,-0.02680443935926676,2.5309746265411377 +MLST8,78,0.002332650823518634,0.13364548981189728,0.03574778512120247,0.9968923356113235,0.9808560182976195,0.07435773433467245,3.179576873779297 +MMGT1,183,0.0014875673223286867,0.12662473320960999,0.028237370774149895,0.9981399509767545,0.9897224736571539,-0.14214445529706016,2.5391173362731934 +MRPL14,211,0.0019686701707541943,0.12983658909797668,0.028742795810103416,0.9975623788685866,0.9895905036846291,0.12574998479742527,2.9209957122802734 +MRPL41,69,0.0037825254257768393,0.13218320906162262,0.04267828166484833,0.9950858866825439,0.9765512707577689,0.13805820442862946,4.048884391784668 +MRPL45,179,0.0014237832510843873,0.1346355676651001,0.027339892461895943,0.9982297806997791,0.9898326919903256,0.06897960916235485,2.4840848445892334 +MRPL53,153,0.001929658930748701,0.12674832344055176,0.031211426481604576,0.9975800266217992,0.9880958528818027,0.20784921297837136,2.8919098377227783 +MRPS22,165,0.001832684502005577,0.13822755217552185,0.03037695400416851,0.9977665075926584,0.9883537116388456,0.06679683504755267,2.8183069229125977 +MTRNR2L1,43,0.003422209993004799,0.1398918479681015,0.04505808278918266,0.995451818794875,0.9688561806943902,0.12924756094744858,3.851215124130249 +NAA10,137,0.0020023337565362453,0.1413273811340332,0.031735993921756744,0.9972515869278922,0.986460887942385,0.13391695655709526,2.945864200592041 +NAA15,128,0.0024648255202919245,0.14382526278495789,0.03529847040772438,0.9966043367713988,0.9826463054196294,0.22219184356685193,3.2684178352355957 +NCBP2,992,0.0087853679433465,0.18081676959991455,0.05641685426235199,0.9880322119604659,0.9640909682932797,0.1980912254582107,6.170557498931885 +NDUFAF3,81,0.002293795347213745,0.13212114572525024,0.03559286519885063,0.9970392563097508,0.9808464520663861,0.09845497848846214,3.152984142303467 +NELFA,89,0.006122341845184565,0.15094198286533356,0.05300942808389664,0.9915777021857709,0.9689293818082718,0.17071152920024346,5.151138782501221 +NHP2,66,0.0034143123775720596,0.1408107429742813,0.04265405982732773,0.9954395826625833,0.9773281711053591,0.1351032513495795,3.846768856048584 +NOL7,73,0.0030264153610914946,0.13683290779590607,0.03968532010912895,0.9959033726041101,0.9798963991469491,0.0236043028523168,3.621668815612793 +NUBP2,153,0.0013812220422551036,0.13017581403255463,0.027257315814495087,0.9982036589266797,0.9890474467437608,0.13941190478866983,2.4466745853424072 +NUP133,79,0.0045686149969697,0.13628624379634857,0.04856446012854576,0.994260037187692,0.9729593529367575,0.09382516493607289,4.449761390686035 +NUP214,132,0.002446993486955762,0.1278180181980133,0.03321298956871033,0.9968078417049756,0.9857238992620727,-0.08538970352879571,3.25657320022583 +NVL,110,0.004177423194050789,0.12486598640680313,0.04459813982248306,0.994678499663463,0.9788053846874398,0.019856667971131194,4.25499153137207 +NXF1,251,0.0010383540065959096,0.1321997046470642,0.023761151358485222,0.9986617569552105,0.9924858845473081,0.052457662520924445,2.1213736534118652 +OPA1,60,0.0038194519001990557,0.1377415508031845,0.0453575998544693,0.9950291908833884,0.9731350147576793,0.15636082420605812,4.068599700927734 +OR2T5,106,0.001654065796174109,0.13296838104724884,0.030859408900141716,0.9978343479659141,0.9855168066892114,0.1293859262970965,2.6774466037750244 +PARN,123,0.0018791541224345565,0.13458675146102905,0.03169511631131172,0.9975004395078639,0.9867473053932881,0.024202993055427398,2.853813886642456 +PCBP2,172,0.001562452525831759,0.12801706790924072,0.028148137032985687,0.9980222986043379,0.9899538629415914,0.01994199208727953,2.602243185043335 +PDCD5,194,0.0015667597763240337,0.12765629589557648,0.026850132271647453,0.9980212544369218,0.9902583502415486,0.019846214109406405,2.605827569961548 +PELO,286,0.0023849555291235447,0.1424388736486435,0.03368685394525528,0.9967539497799894,0.9867725022023248,0.34808838063099434,3.21502685546875 +PGS1,104,0.0019077920587733388,0.13492491841316223,0.032638099044561386,0.997522757521149,0.9848324863205521,0.06549822028843873,2.8754775524139404 +PHF5A,231,0.010281607508659363,0.15591086447238922,0.06927333027124405,0.9866147037579117,0.9522464773655553,0.30059921310938775,6.675364017486572 +POLR2K,22,0.02603873983025551,0.23177094757556915,0.11541703343391418,0.9656795460887846,0.8706580944292375,0.3864217046030058,10.623177528381348 +PPP1R15B,135,0.001839761040173471,0.1356036514043808,0.031215859577059746,0.9975051038055607,0.9865142048020675,0.23192452638628894,2.8237431049346924 +PRPF31,27,0.019291777163743973,0.1689402163028717,0.10043372958898544,0.9763176066345948,0.8992069993244809,0.15506581617260007,9.143880844116211 +PRRC2A,89,0.0033127523493021727,0.12779495120048523,0.041353922337293625,0.9961176161462945,0.9791891447585938,-0.0682640915657155,3.7891249656677246 +PSMB6,200,0.004634104669094086,0.15939094126224518,0.04490577057003975,0.9937008066146414,0.9771784288202734,0.6408709435741006,4.481541156768799 +PSMC2,432,0.00631877314299345,0.17658910155296326,0.052645210176706314,0.9917372413668073,0.9732473355070016,0.6857953097998986,5.233121871948242 +PSMD1,281,0.001350793638266623,0.13986839354038239,0.025014517828822136,0.9981504741344854,0.9910303534727258,0.5293963670731957,2.41957426071167 +PSMD11,80,0.005767566617578268,0.16794085502624512,0.053009048104286194,0.9922119319245823,0.966226444809192,0.57713836697284,4.999663352966309 +PSME2,151,0.0016225927975028753,0.12639005482196808,0.028770150616765022,0.9979582021888626,0.9879946997844814,-0.03713180989797055,2.6518516540527344 +PSMG2,164,0.001268457737751305,0.13067801296710968,0.026389244943857193,0.9983628696947543,0.9894284958749959,0.06284521434915945,2.3446738719940186 +PTK2,144,0.0013979325303807855,0.1342674195766449,0.02796258218586445,0.9981354490803904,0.9882746681946389,0.13296286249819558,2.461430549621582 +PTPN23,140,0.0016263396246358752,0.12772224843502045,0.02947498857975006,0.9979683053358094,0.987921828404609,-0.07286812520376185,2.654911756515503 +PXN,161,0.0018042813753709197,0.14064937829971313,0.029001250863075256,0.9976491107973804,0.9873832142369916,0.040095850707731785,2.7963826656341553 +RAC1,194,0.0015600362094119191,0.12925119698047638,0.02695908397436142,0.9980695478378522,0.9909020999314505,0.009923561667918128,2.6002302169799805 +RAD51,241,0.002899210900068283,0.11780177056789398,0.03258920833468437,0.9961217719126643,0.9864258780349695,0.1640966583493441,3.5447397232055664 +RANGAP1,130,0.0069275894202291965,0.13726721704006195,0.05602840334177017,0.9908255930079733,0.9670503771480554,0.006032230297286999,5.479431629180908 +RBM14,228,0.006603272166103125,0.13622117042541504,0.04969737306237221,0.9910466248972701,0.960735970423712,0.15467307760663618,5.349633693695068 +RBM22,5,0.037024933844804764,0.14280113577842712,0.1463848203420639,0.9560598336936679,0.8013546484887168,0.012938919495846031,12.667520523071289 +RFC5,111,0.0023116639349609613,0.12344861775636673,0.03431343659758568,0.9968979710876514,0.9834249753722759,0.2245105642762494,3.165241003036499 +RIOK2,174,0.007709289435297251,0.15417954325675964,0.058500390499830246,0.9901262742626489,0.9761596500876121,0.164700285984095,5.7803168296813965 +RNASEH2C,102,0.002832066034898162,0.14974160492420197,0.03824911266565323,0.9963368580963758,0.9798214962580426,0.13376520244359313,3.5034518241882324 +RNF113A,176,0.00995408184826374,0.16743157804012299,0.07104036957025528,0.9886657347985297,0.9690649290990196,0.12667034173827552,6.568180561065674 +RNF168,61,0.0029230457730591297,0.12076324969530106,0.040641747415065765,0.996361915112183,0.9793591474883433,-0.08915891605068366,3.5592808723449707 +RNF20,124,0.008355429396033287,0.1475989669561386,0.06188984587788582,0.9887314887963815,0.9577563651710496,0.336287091093025,6.01767635345459 +RPA1,469,0.0019569138530641794,0.15066957473754883,0.031118987128138542,0.9975050249263712,0.9896794881382579,0.35476720115502597,2.9122610092163086 +RPF1,125,0.0040880246087908745,0.13729265332221985,0.04471045359969139,0.994631207524218,0.97662336427728,0.18040033341610284,4.2092156410217285 +RPL31,294,0.01428364496678114,0.1595403403043747,0.07992318272590637,0.9818479687392726,0.9539080956705115,0.3993436361275307,7.867993354797363 +RPL8,384,0.008475079201161861,0.154808908700943,0.05937367305159569,0.9889147585737343,0.9698961846693187,0.3678795177866763,6.060609817504883 +RRN3,39,0.01120865810662508,0.16867002844810486,0.07388749718666077,0.9850532425515423,0.9444128487048624,0.23775498097921957,6.969815254211426 +RTCB,171,0.0019704452715814114,0.1331815868616104,0.03129201382398605,0.9975560655192249,0.988289006171327,0.009281373233852124,2.9223122596740723 +RTTN,125,0.0016843959456309676,0.136875718832016,0.03049822896718979,0.9977880597664353,0.9866997963218148,0.07021048104964911,2.70188307762146 +SARS2,163,0.0018101187888532877,0.14023424685001373,0.03081277199089527,0.9976585582941428,0.9871244423001547,0.09291815589543642,2.8009023666381836 +SCAP,140,0.001760732615366578,0.13490962982177734,0.030042041093111038,0.9977478450785505,0.987306715200578,0.097787905109908,2.7624292373657227 +SDE2,151,0.0023562503047287464,0.13512802124023438,0.03451750800013542,0.9969163540138095,0.9865307550821634,0.09888466927514165,3.19562029838562 +SDHA,62,0.0031959936022758484,0.12282105535268784,0.04191073402762413,0.9957578933589502,0.976044797761076,-0.015855167551596733,3.7217516899108887 +SEC16A,153,0.0013772830134257674,0.13371068239212036,0.02763332612812519,0.9981851897351672,0.988851560366166,0.036811712977625546,2.443183422088623 +SLC35G6,163,0.0012062143068760633,0.13284659385681152,0.025979066267609596,0.9984491348235749,0.9905447844586645,0.06811740521021521,2.286423683166504 +SLU7,225,0.022236013785004616,0.19720585644245148,0.10459069907665253,0.9727552644736019,0.9320834438632172,0.2025242926232378,9.816866874694824 +SMC6,125,0.0016486281529068947,0.13268402218818665,0.029385093599557877,0.9977767955040892,0.9866501169221452,0.21252985782727687,2.673042058944702 +SNRNP27,47,0.007150729186832905,0.15210281312465668,0.06091313436627388,0.9906024650461688,0.9588526227732574,0.050360543892788635,5.566979885101318 +SNRNP35,168,0.003784075379371643,0.15123148262500763,0.042132798582315445,0.995003216986505,0.9789876658052548,0.21846102590024738,4.049714088439941 +SNRNP70,129,0.012199117802083492,0.16685190796852112,0.07513562589883804,0.9831873615861,0.9419289250478802,0.3061690886386513,7.271243095397949 +SNX15,637,0.0008149196510203183,0.1412104070186615,0.019828803837299347,0.9989183870431232,0.995142679370556,0.3533260128877782,1.8793247938156128 +SPC25,199,0.011676941066980362,0.20849041640758514,0.07345981895923615,0.9864176452838779,0.9504211090689509,0.551470707218813,7.11392068862915 +SPTSSA,124,0.0017428650753572583,0.12386389076709747,0.030916284769773483,0.997805128843114,0.9863996962977833,-0.05407775807481128,2.7483773231506348 +SRSF10,133,0.003173605538904667,0.1377021074295044,0.03803239390254021,0.9957362283340409,0.9800218237149305,0.1259576086794763,3.708693265914917 +SRSF3,29,0.020446058362722397,0.21154262125492096,0.10063017159700394,0.9722777492548769,0.9009426886084584,0.3488167157939628,9.413458824157715 +SSX4,101,0.0019717488903552294,0.1296040415763855,0.032851144671440125,0.9974640457754418,0.9841896257823867,-0.08330386131152803,2.923279047012329 +TAF2,144,0.011281286366283894,0.18624615669250488,0.06750062108039856,0.9854036710776236,0.9585911387636892,0.27499095916011174,6.992359161376953 +TARDBP,169,0.0018323005642741919,0.12911051511764526,0.029414217919111252,0.9976037813438317,0.9885237114592813,0.09216458524674347,2.81801176071167 +TBCC,141,0.001471452647820115,0.13123519718647003,0.0284162275493145,0.9980898638195269,0.9885981034773027,-0.016218834059237457,2.52532696723938 +TBP,151,0.002799972891807556,0.14767254889011383,0.0351555272936821,0.9961414284279948,0.9845614030518917,0.12467416553728769,3.4835445880889893 +TDGF1,106,0.0017357633914798498,0.13196498155593872,0.03158402815461159,0.997763570492751,0.9854675905496696,-0.00521426134548089,2.742772102355957 +TFRC,57,0.003211050760000944,0.13768835365772247,0.041828397661447525,0.9955768801262784,0.9729178900479851,0.25725753882209024,3.73050856590271 +THAP1,110,0.0037865641061216593,0.1480860561132431,0.04290902614593506,0.9949714518232965,0.9776185375200765,0.500746029050156,4.0510454177856445 +THOC3,147,0.004204204771667719,0.13812261819839478,0.04277757555246353,0.9943094002844515,0.9793123359062373,0.17666008643575568,4.268609523773193 +TMEM258,193,0.0013784673064947128,0.1380365937948227,0.025880642235279083,0.9981753588604888,0.9903267204381597,0.21642534963063953,2.4442334175109863 +TOMM22,104,0.004548344295471907,0.1486646682024002,0.043796442449092865,0.9942241194919255,0.9788592992386995,0.13925090902430642,4.439878940582275 +TRA2B,222,0.003653536085039377,0.13921526074409485,0.03781604766845703,0.9953353768734879,0.982750889754,0.05378530530411838,3.9792492389678955 +TRAPPC3,163,0.0013986736303195357,0.12788167595863342,0.02651451900601387,0.9981829619879548,0.9894213335741939,0.09906928933040296,2.462082624435425 +TRAPPC5,132,0.0017959041288122535,0.1368853896856308,0.030695190653204918,0.9975532925389806,0.9857944615302151,0.1961215519476069,2.7898833751678467 +TREX2,275,0.002283626003190875,0.15755710005760193,0.03262047842144966,0.9970594627889677,0.9876235645771854,0.4072880001903961,3.145987033843994 +TRNT1,203,0.0013238298706710339,0.13097615540027618,0.026243053376674652,0.9983203738792598,0.9900805328287066,0.0744457428508775,2.39530348777771 +TSEN34,275,0.0009165623923763633,0.13129617273807526,0.02233295887708664,0.9988247887698931,0.9930822657916296,0.09097249792709353,1.9930833578109741 +TSEN54,52,0.0034545050002634525,0.12842659652233124,0.04348631948232651,0.9954765313517446,0.9726887768192798,0.02597673072615936,3.8693442344665527 +TUBA1B,74,0.0037638212088495493,0.1536332666873932,0.040493592619895935,0.9949448663614533,0.9753987620643951,0.27062026619400176,4.038861274719238 +TUBGCP2,238,0.001870886073447764,0.1263977289199829,0.02990315668284893,0.9975329524587184,0.9895283861833636,0.044822288984763585,2.8475289344787598 +TUBGCP6,153,0.0012963308254256845,0.13066677749156952,0.02689777873456478,0.9983379445659041,0.9894202747591707,0.04166606294420464,2.3702948093414307 +TWF1,140,0.001519830897450447,0.13452495634555817,0.029049912467598915,0.9980405023276298,0.987944882522166,0.05369752398730856,2.566504955291748 +TXNL4B,135,0.0015039246063679457,0.12989668548107147,0.028645699843764305,0.9980365663977191,0.9878139935128464,0.022678123509378326,2.553039312362671 +UQCRC2,68,0.0034541788045316935,0.12812715768814087,0.042733997106552124,0.9955060735461921,0.9753024539006379,0.1580715267478751,3.869161605834961 +USP7,276,0.0010239967377856374,0.12839998304843903,0.023172009736299515,0.9987462713305558,0.9933366859773427,0.06967136664513172,2.106656551361084 +VPS37C,177,0.001109667238779366,0.12784874439239502,0.02506229467689991,0.9985924669558757,0.9912420258313934,0.0771355825525407,2.1930110454559326 +WDR1,102,0.0018403077265247703,0.12772779166698456,0.032278165221214294,0.997689942317896,0.9855481390442953,-0.0012083730384112403,2.824162244796753 +WDR3,365,0.006463979836553335,0.15723824501037598,0.052433207631111145,0.9916809184848103,0.9805479679048609,0.15758740288416145,5.292909622192383 +WDR36,285,0.006621744018048048,0.14963370561599731,0.05423545837402344,0.991990460485541,0.9833864881622298,0.17946483577111294,5.357110977172852 +WDR46,71,0.005404026247560978,0.15215718746185303,0.051509760320186615,0.9927167676263143,0.9720183873650198,0.16480931901719753,4.839529991149902 +WDR74,178,0.00184480135794729,0.13144448399543762,0.031160088256001472,0.9976623392241168,0.9882503698239038,0.003372687833590047,2.827608346939087 +WDR77,42,0.004887836519628763,0.12264140695333481,0.0518919937312603,0.9938509738662225,0.9640771799974374,-0.012168121247511373,4.602595329284668 +WDR82,99,0.01257177721709013,0.18571601808071136,0.07667265087366104,0.9832263148457695,0.9377775904530234,0.39147150769481404,7.381468772888184 +YEATS4,231,0.003566995495930314,0.13135793805122375,0.0408487543463707,0.995723706794117,0.9806972173371289,0.009715476293962003,3.9318392276763916 +YTHDC1,237,0.0030887946486473083,0.15069982409477234,0.03665463998913765,0.995812071163119,0.9821069265480797,0.3025559232733639,3.6588027477264404 +ZC3H8,21,0.009119912981987,0.13415563106536865,0.07208750396966934,0.9880001232626717,0.9353861038436775,0.066414343877782,6.286946773529053 +ZMAT5,100,0.00355200981721282,0.1472642570734024,0.041981685906648636,0.9951521175325037,0.9759353831947067,0.17932647467389687,3.9235711097717285 +ZNF236,85,0.007222693879157305,0.1495947241783142,0.056809842586517334,0.9904442562253192,0.971905060724555,0.13504633129219457,5.5949225425720215 diff --git a/baseline/outputs/cellot_ot_baseline/pred_cellot_ot_baseline_use_existing_test.h5ad b/baseline/outputs/cellot_ot_baseline/pred_cellot_ot_baseline_use_existing_test.h5ad new file mode 100644 index 0000000000000000000000000000000000000000..f3f394710e7cb0f60edd4e5acd6878e05bafe569 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/pred_cellot_ot_baseline_use_existing_test.h5ad @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:93b9a0d8cae139120df5063449c42e19dd1e8b7a31cce18c41c925d329db194f +size 477037547 diff --git a/baseline/outputs/cellot_ot_baseline/real_cellot_ot_baseline_use_existing_test.h5ad b/baseline/outputs/cellot_ot_baseline/real_cellot_ot_baseline_use_existing_test.h5ad new file mode 100644 index 0000000000000000000000000000000000000000..6cd49466fb4f50938b5a1cb404d2a5ab902c646d --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/real_cellot_ot_baseline_use_existing_test.h5ad @@ -0,0 +1,3 @@ +version https://git-lfs.github.com/spec/v1 +oid sha256:14978a475e94f1e7d9397691d45d043e8ea51d415af1766cf19d7a1eaa3ffb26 +size 149497912 diff --git a/baseline/outputs/cellot_ot_baseline/summary_cellot_ot_baseline_use_existing_test.csv b/baseline/outputs/cellot_ot_baseline/summary_cellot_ot_baseline_use_existing_test.csv new file mode 100644 index 0000000000000000000000000000000000000000..6827a3623d5e623d24d9d97c6f70de72b457bdd6 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/summary_cellot_ot_baseline_use_existing_test.csv @@ -0,0 +1,18 @@ +,value +method_name,cellot_ot_baseline +input_path,/root/final/data/processed/scPerturb/prediction_ready/replogle_k562_essential_single_gene_prediction_ready_full.h5ad +eval_split,test +train_split,train +val_split,val +n_control_cells_train,10816 +n_eval_cells,33235 +n_train_genes,1645 +n_eval_genes,205 +n_eval_genes_missing_from_train,205 +mse_mean_avg,0.005047890109225835 +mse_cell_avg,0.14435312380877938 +mae_mean_avg,0.04422009014865247 +pearson_mean_avg,0.9934415852860539 +spearman_mean_avg,0.9735030627041961 +pearson_delta_avg,0.14896064115191923 +delta_l2_avg,4.179004827359828 diff --git a/baseline/outputs/cellot_ot_baseline/training_history_cellot_ot_baseline_use_existing_test.csv b/baseline/outputs/cellot_ot_baseline/training_history_cellot_ot_baseline_use_existing_test.csv new file mode 100644 index 0000000000000000000000000000000000000000..cd260445199793b7fbb692df0d234a09a9aab3f9 --- /dev/null +++ b/baseline/outputs/cellot_ot_baseline/training_history_cellot_ot_baseline_use_existing_test.csv @@ -0,0 +1,21 @@ +epoch,train_total,train_recon,train_map,train_latent_mmd,train_expression_mmd,train_mean,train_disp_reg,val_total,val_recon,val_map,val_latent_mmd,val_expression_mmd,val_mean,val_disp_reg +1,0.4351984862983227,0.2994258677214384,0.07512082519009709,0.2357425656914711,0.18945791125297545,0.017438311617006548,0.06934359385399148,0.34683538787066936,0.27112142415717244,0.0033794566697906703,0.4873400442302227,0.17969520762562752,0.005597195598966209,0.0033789140798035078 +2,0.30522705495357516,0.27276057451963426,0.005056204128777608,0.04808979719877243,0.1750274407863617,0.0050629764335462825,0.003557629472343251,0.3563673049211502,0.27600713493302464,0.0010946315742330626,0.5588131323456764,0.1797294169664383,0.005400345675298013,0.0010938364648609422 +3,0.30284063145518303,0.27386495023965834,0.002855300881783478,0.028716273307800293,0.1787255197763443,0.0053596693923464046,0.0016532479634042828,0.3424721360206604,0.2772123012691736,0.00046438568369921995,0.4051009379327297,0.17936796322464943,0.0063439139412366785,0.00046443672363238875 +4,0.2992048604786396,0.2711798367649317,0.002263234831625596,0.03290950208902359,0.17690979361534118,0.004767422667937353,0.0012437378437607548,0.34956309106200933,0.26739230938255787,0.000611350436884095,0.580060500651598,0.1825483813881874,0.00529246400037664,0.0006081835763325216 +5,0.306126167178154,0.27707135044038295,0.0018803436192683875,0.03333548039197922,0.17912028670310975,0.005919533837586641,0.0009363106897217221,0.3417016016319394,0.2726433230563998,0.0002582029851510015,0.4440433718264103,0.18340514600276947,0.00605263545003254,0.0002587190606391232 +6,0.3026826225221157,0.27407530263066293,0.0016917025978909804,0.03473086088895798,0.1788805067539215,0.005547835445031524,0.0006646595937490929,0.34457709547132254,0.27391466218978167,0.0002484152987562993,0.4761493019759655,0.17463326081633568,0.005333292989234906,0.0002475381588737946 +7,0.3003147205710411,0.27275603339076043,0.0012666964653180912,0.03539767861366272,0.1772846585512161,0.0050198589783394705,0.0003896997933043167,0.34280285611748695,0.2697424329817295,0.0001553056617922266,0.5038080029189587,0.17315049096941948,0.005207715330470819,0.00015502651694987435 +8,0.30012349903583524,0.2734407402575016,0.000989391700422857,0.02831693917512894,0.1776679915189743,0.005092795387608931,0.00020768183305335696,0.3222009623423219,0.26543337432667613,5.567748371504422e-05,0.3632640242576599,0.16375548020005226,0.004009411466540769,5.490223702508956e-05 +9,0.2983067686855793,0.27199625059962274,0.0008302987441129516,0.026939606368541716,0.17780435144901274,0.005004654062795453,0.00011710623562976253,0.3329483298584819,0.2717398926615715,3.91961282275588e-05,0.39613643288612366,0.15910564363002777,0.005644648877932923,3.8481070532725425e-05 +10,0.29847508028149605,0.2721160004287958,0.0007897229507216252,0.025950647592544555,0.17886114954948426,0.005087419442716055,7.576533465908142e-05,0.3330086823552847,0.2769216811284423,2.9613484343826713e-05,0.32319945469498634,0.163493063300848,0.007387865349301137,2.6746850664949307e-05 +11,0.3015074972808361,0.2749501469731331,0.0007265926613763441,0.023982208371162415,0.1790255057811737,0.005529474340146407,5.1257514005556e-05,0.32248205225914717,0.27235058043152094,2.025229605351342e-05,0.2882181704044342,0.15735230967402458,0.005553992850764189,1.8268724829795246e-05 +12,0.30129998683929443,0.27520734176039696,0.000640567380542052,0.021816883385181427,0.17828874111175538,0.005441135295550339,3.81817454672273e-05,0.3277882970869541,0.2694784188643098,2.249234660212096e-05,0.39002396538853645,0.15289177000522614,0.003995599130576011,2.1234872122022352e-05 +13,0.29916656881570813,0.2733767249435186,0.0006182765130506595,0.02267637461423874,0.17715445816516875,0.005188139023957774,3.435694873587636e-05,0.3154166229069233,0.2673559067770839,1.4946740435561878e-05,0.28093312680721283,0.15541113540530205,0.00441121740732342,1.3170451722999132e-05 +14,0.29612924516201017,0.2711896861344576,0.000595758765048231,0.021198163628578185,0.17686958968639374,0.004536755884764716,2.6835256976482922e-05,0.3043453246355057,0.2613531374372542,1.2317419532337226e-05,0.23633838072419167,0.1539182886481285,0.003954094605433056,1.037251541902151e-05 +15,0.29902917131781576,0.2729677454382181,0.0006575842888923944,0.023859563171863555,0.17788366317749024,0.005229149949154816,3.692381847486104e-05,0.3167215194553137,0.2686799461953342,2.7653049016862497e-05,0.27813754603266716,0.156122837215662,0.004587642484693788,2.3581467473832163e-05 +16,0.29692463830113414,0.2712174865603447,0.0005888743303876253,0.021327650547027587,0.17897921085357665,0.005087267523631454,3.236947470213636e-05,0.32148604560643435,0.2742463946342468,2.2532801324359752e-05,0.25795451924204826,0.15418720245361328,0.006002775062370347,1.7686154592411185e-05 +17,0.3004035332798958,0.27377601139247415,0.0006500627526111202,0.025107052624225617,0.1781159782409668,0.005654626676114276,5.306729851326964e-05,0.3284460175782442,0.27526704873889685,3.101439861552535e-05,0.31345183216035366,0.15846654027700424,0.005955847074801568,2.70954769518994e-05 +18,0.29924720391631127,0.2736141280829906,0.0005802306529221823,0.021104623675346375,0.17818372786045075,0.005123738358961418,2.7144545260853192e-05,0.33022357895970345,0.27460985723882914,1.955303321210522e-05,0.35100532323122025,0.15551679581403732,0.004941769482684322,1.8929488533103722e-05 +19,0.29879038825631143,0.2723075287789106,0.0007860649344365811,0.029196833968162538,0.17850688695907593,0.004925404325476847,0.00010195142454904272,0.31102578341960907,0.2670343955978751,5.470041821808991e-05,0.24658112227916718,0.15215956047177315,0.004062140829773853,4.808303629033617e-05 +20,0.29644450947642326,0.2717776880413294,0.0006301876441648347,0.01574607938528061,0.17738543450832367,0.004723146437900141,3.3567319865142056e-05,0.30524126812815666,0.2729722894728184,8.69060653485576e-06,0.13219034671783447,0.1496274620294571,0.004078448655491229,6.209175523963495e-06 diff --git a/baseline/outputs/conditional_flow_matching_method/aggregate_conditional_flow_matching_method_seen_cell_split_test.csv b/baseline/outputs/conditional_flow_matching_method/aggregate_conditional_flow_matching_method_seen_cell_split_test.csv index ccffdeb5c242dcf6d7a87d67e46784d520169857..030915270f083e08371990fc69ceef6e82b7493a 100644 --- a/baseline/outputs/conditional_flow_matching_method/aggregate_conditional_flow_matching_method_seen_cell_split_test.csv +++ b/baseline/outputs/conditional_flow_matching_method/aggregate_conditional_flow_matching_method_seen_cell_split_test.csv @@ -1,7 +1,7 @@ ,mean,median n_cells,14.574416342412452,12.0 -mse_mean,0.019163491268137628,0.013615784235298634 -mae_mean,0.09751620552328628,0.08805787190794945 -pearson_mean,0.9756440617310962,0.9822932157833737 -pearson_delta,0.08957929501688626,0.07334081619542766 -delta_l2,8.494131993922743,7.681849241256714 +mse_mean,0.01911219263722004,0.01367475325241685 +mae_mean,0.09772463235369945,0.08835862949490547 +pearson_mean,0.9756348678854861,0.9821302642656909 +pearson_delta,0.11158561568526223,0.09204006229546872 +delta_l2,8.491365785728632,7.698466062545776 diff --git a/baseline/outputs/conditional_flow_matching_method/conditional_flow_matching_model.pt b/baseline/outputs/conditional_flow_matching_method/conditional_flow_matching_model.pt index ef08aac76369921ad27c991c663bc24692316756..daa5e9dadcdf1466220ce7ef8ac8733955196a64 100644 --- a/baseline/outputs/conditional_flow_matching_method/conditional_flow_matching_model.pt +++ b/baseline/outputs/conditional_flow_matching_method/conditional_flow_matching_model.pt @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:2cbc0b045c38038d9f8d5d98981ffa3fa19b228a642095c43e80078f418680a9 +oid sha256:c668243198f702bae1fee13a874912506f2e8db35f6fa442c2edf607860b53d6 size 76474311 diff --git a/baseline/outputs/conditional_flow_matching_method/config_conditional_flow_matching_method_seen_cell_split_test.json b/baseline/outputs/conditional_flow_matching_method/config_conditional_flow_matching_method_seen_cell_split_test.json index a81c2c74bbb2eb71078064e1938a936bf2ad1e7c..5aa7a128a47b2cd858fded0dec6a2debbe610951 100644 --- a/baseline/outputs/conditional_flow_matching_method/config_conditional_flow_matching_method_seen_cell_split_test.json +++ b/baseline/outputs/conditional_flow_matching_method/config_conditional_flow_matching_method_seen_cell_split_test.json @@ -34,6 +34,10 @@ "delta_cos_weight": 0.0, "delta_mmd_weight": 0.0, "delta_bulk_weight": 0.0, + "ot_decode_mmd_weight": 0.05, + "ot_decode_mean_weight": 0.25, + "displacement_reg_weight": 0.01, + "velocity_l2_weight": 0.0, "noise_std": 0.05, "sinkhorn_epsilon": 0.5, "sinkhorn_iters": 30, diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.pdf index 45ed0c856344b4395836daa673992db9f9f96a80..361b9361fd08b869a89aa2c94e8b320fa8050aeb 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.png index 8eb3def9ae200e7071f792ac7ad6142f650d436c..f0228ab77ab79ed656c12362945d3e811e451375 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/bio_A_gene_mean_calibration_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:8f1cf15ddfa6c1be488eb4e610e7df8d9e90bd928494a00f3e69f80c9d2cd39c -size 108315 +oid sha256:f1d87ff578601e3f852cdc59cebd5e0174afd90733b16933ad7a092b1e6b8c26 +size 108546 diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.pdf index 67336e9ce52d06493ef6c3a68234b44a98d82a5e..a964b6be7293fd345ad5e2b608616f7be4c3911d 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.png index 3d28969810f8ca81414a6ce42fb30dad3621d661..7376e4b7a8b06f941a80c09d53e6a7a1c2587f21 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/bio_B_MA_plot_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:cdfe5f73aa1b569a60cc9e123884f660ae6535adaf35792f40fea9a04c99873f -size 164140 +oid sha256:6cc419faae77d1790bb3a644847ff4ecd1759a0c3f4c3d1c37548db7d0fb17fe +size 160673 diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.pdf index 28427371cb6d63ec95eccd24fb27162ca0422d6b..a3da2f9f719218f2b32502e88eebac73e05733d9 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.png index 6d6a8d608e9b00af39cda4641176622e8a655864..68e537bb6a16a616aa1f5de0bf7f8cf93de17bc4 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.png and b/baseline/outputs/conditional_flow_matching_method/figures/bio_C_gene_error_hist_seen_cell_split_test.png differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.pdf index 6f7a157dcaab967f7018e8c615915c98a34cf9be..3627f4b4707ff664e95ac159012de841a66b2519 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.png index 6b656a21e4524deba08bd88b2c17fcde353b2e61..819314ae70a2cdbd33a8bf92665207782dc342ce 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_predicted_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:dbdb4135389f036fc8c600a9a054800f7556117d656e90f6e9eda413ffaab213 -size 136428 +oid sha256:5427c1e169e175f0d38c5e09f0eeb354a48570c35eaa111dcbfce0158c2d1108 +size 136320 diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_truth_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_truth_seen_cell_split_test.pdf index b8e163ba1a38ddb5bcfe05d7319419f9be751b21..6650c16854b54f760ccef4ec3957024fb53da032 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_truth_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_DE_heatmap_delta_truth_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_F_perturb_corr_truth_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_F_perturb_corr_truth_seen_cell_split_test.pdf index c24a312caa873365da7dc2945b2c12d94dcc42c2..576d2ca4e1670d7dcc29c03f8ec90e0bd8eae559 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_F_perturb_corr_truth_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_F_perturb_corr_truth_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_G_perturb_corr_clustered_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_G_perturb_corr_clustered_seen_cell_split_test.pdf index efd77603ffb08a367805cf18f261d1ae39e1c9ae..bfb38a4399c1f4a8026f5bf77b3de8b3e6c0d3db 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_G_perturb_corr_clustered_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_G_perturb_corr_clustered_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.pdf index 5f86cdb118413a3664cbdcc0da9a596316c8b3c2..04ddfe4cf1d6f986102e7b2c092920d899b02954 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.png index 8de48c8ef30747b9b82a2c95c1f9d124685e2492..a12f82b3a95c6766f10780902edd4cea8eeab641 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/bio_H_cell_PCA_truth_pred_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:735a3fe9e1f36a86050e0341b4e057305a6d5c8c83adccaab9c50ee4a17faf9f -size 139846 +oid sha256:5444735d0184b710547df1e91ea4efca2966663947baa5079c974a673b07fa4d +size 139435 diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.pdf index ae533a8437f0261946aa5f796c8d3971475fbb0a..b0e5ebfd9164d310bf2370c8235c4727875d3625 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.png index b97c07952de6585849441dba1f0580b707096964..4ebe78def1ab513cbf8e239d911f77d843a983b8 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.png and b/baseline/outputs/conditional_flow_matching_method/figures/bio_I_cell_L2_error_seen_cell_split_test.png differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.pdf index 7ea0470f6badb17662eeabc773e6d1a724f7cbfd..071802adc55ed175c5e77e78471dcafbd09fa9f2 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.png index e9793b4cef39b91dd6663753e62477665d3cd610..943b32e7405523983f2d391481d2b6777a3a8f88 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/bio_J_metric_panels_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:943c27c53a32e6f9ff9849df85d01bf75636243244fb092713dc0558e815c97e -size 194065 +oid sha256:5a884dbaf237b21ba61f9d07d360e6a1f1dde844ea56394d8a9cd739d5d495ff +size 194671 diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.pdf index c86ca368792bc248a3d46d2932c1bb2dfdcb417c..e51bcc9862cec9dcb7d45d61c8cf098c34b0664d 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.png index fb3ed54d1b61a9287cd875c4086b2082d5a6e46c..67cdc3ebdbb711f4c53d03f07e262b96247cff55 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/bio_K_topvar_gene_scatters_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:2d6e5bb4e67c08fbbdab05a67e3a0b0292bfd2657212d861d9805ab898925cd1 -size 159463 +oid sha256:2c33d41b83d6cb2f411f083362da72c37c346bfd06fefe4122bfa9004b24820f +size 159467 diff --git a/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.pdf index b0e3f152cf1e2fbba1af95f32226d5a9e96f80c3..62a384cb23ffefd0ef14980be5b2d1872130c56b 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.png index 9d3b4de953226643d6a573381eea0073869454a1..80e726110c5e0b8a0f7200dfc6a883ce76e008ee 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/evaluation_metrics_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:b34049a7114c1b2385de176741ba1f2b8066ab8abdac1917b7cd9788104ad5e7 -size 291589 +oid sha256:b7991198bb55bcdaca1399b5a2f9b54f5f5164b03cd95a6cfa2c531521e93b23 +size 299043 diff --git a/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.pdf b/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.pdf index f39c30d2aaa74c54335ed77975c307d0696a80c0..57cd8551abba76db69bbaaed06e879366cff7bdb 100644 Binary files a/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.pdf and b/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.pdf differ diff --git a/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.png b/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.png index b4039ced1b9cdd1ccefeca1a416fbecf761c12d0..44dc76649b9ba6a80d8d8eb8c05940b57831bb32 100644 --- a/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.png +++ b/baseline/outputs/conditional_flow_matching_method/figures/training_history_seen_cell_split_test.png @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:6d6ce4de4c5599ae9c612c8617e318d2019ac9d7475bed5b615d0437a9f1d883 -size 120459 +oid sha256:96bbe47ebe5bd57f86b175b0d911f4e0e5cc59c99c4476e1528ff16326f6498f +size 155144 diff --git a/baseline/outputs/conditional_flow_matching_method/metrics_by_gene_conditional_flow_matching_method_seen_cell_split_test.csv b/baseline/outputs/conditional_flow_matching_method/metrics_by_gene_conditional_flow_matching_method_seen_cell_split_test.csv index 1f50102e5cf4b374d7168fa01d9845758e49a643..e7dbc15f8d5b1b69e66ae6f0eda23fd1264fb141 100644 --- a/baseline/outputs/conditional_flow_matching_method/metrics_by_gene_conditional_flow_matching_method_seen_cell_split_test.csv +++ b/baseline/outputs/conditional_flow_matching_method/metrics_by_gene_conditional_flow_matching_method_seen_cell_split_test.csv @@ -1,2057 +1,2057 @@ perturbation_gene,n_cells,mse_mean,mae_mean,pearson_mean,pearson_delta,delta_l2 -AAAS,21,0.006696952041238546,0.06215137243270874,0.9913107329515842,-0.04660446320851371,5.387447357177734 -AAMP,10,0.019449541345238686,0.10457880049943924,0.9742157438393497,0.24909615306923805,9.181193351745605 -AARS,4,0.04485318064689636,0.16069871187210083,0.9435049313695678,0.2996378067074979,13.942513465881348 -AARS2,18,0.007287762593477964,0.0660644993185997,0.9902741459128103,0.036956359119331705,5.620068073272705 -AASDHPPT,5,0.030114671215415,0.13346679508686066,0.9641452256525193,0.018527004436662053,11.424403190612793 -AATF,30,0.01074981689453125,0.07496591657400131,0.9860244587538445,0.1412728348640422,6.825665473937988 -ABCB10,5,0.02811603434383869,0.1276351362466812,0.9637491587709762,0.13782269109083287,11.038789749145508 -ABCB7,20,0.0069134170189499855,0.06387723237276077,0.9907193170083942,0.06503195216537558,5.4738240242004395 -ABCE1,25,0.015911825001239777,0.09032203257083893,0.9787201378214035,0.4426860243074601,8.304327964782715 -ABCF1,33,0.0051780035719275475,0.05393582582473755,0.993079188961998,0.19310747231362205,4.737242698669434 -ABCG1,15,0.009320114739239216,0.07442016154527664,0.9879356634644005,-0.043350753645654744,6.355578899383545 -ABHD11,5,0.02597569115459919,0.12574267387390137,0.966962921345184,0.03404527891549259,10.610307693481445 -ABHD17A,9,0.01615077070891857,0.09731706976890564,0.9790891219469555,0.08209576911092653,8.366447448730469 -ABT1,10,0.01712382771074772,0.09886696934700012,0.9774450145339995,0.018871540649215198,8.614792823791504 -ACD,7,0.016884220764040947,0.10007361322641373,0.9792490194826328,-0.09716514866639937,8.554309844970703 -ACIN1,19,0.00747084803879261,0.06554136425256729,0.9902504662290323,-0.12207662083414225,5.690224647521973 -ACOT12,18,0.007710541598498821,0.06753220409154892,0.9898006807578048,0.06247092095913108,5.780786514282227 -ACTB,17,0.008010935969650745,0.06680751591920853,0.9893555671832113,0.03961238687198208,5.892316818237305 -ACTL6A,5,0.030975092202425003,0.13328135013580322,0.9603591786110144,0.02340777667684463,11.58646011352539 -ACTR10,11,0.012739605270326138,0.08714152127504349,0.9829066912823194,0.13282439915845495,7.430575370788574 -ACTR1A,15,0.009048545733094215,0.0734034851193428,0.9883981074255914,0.0023365589859628633,6.26230001449585 -ACTR1B,20,0.00733625702559948,0.06585538387298584,0.9901765355057215,-0.07418454883315105,5.638735294342041 -ACTR2,23,0.0061353216879069805,0.0582694411277771,0.9919608808431692,0.03850411239790475,5.1565961837768555 -ACTR3,12,0.011500660330057144,0.08192110061645508,0.9849544976649945,0.02841755734582976,7.060018539428711 -ACTR6,11,0.014145107008516788,0.09211817383766174,0.9831090673399062,-0.033984276406730046,7.829744338989258 -ACTR8,25,0.008586031384766102,0.0690857544541359,0.9885474044246302,0.11866845511430119,6.100152492523193 -ADAT2,9,0.01514959055930376,0.09529407322406769,0.9806046910236246,0.049412867250954846,8.102982521057129 -ADAT3,13,0.0104887206107378,0.07830286771059036,0.9859053978990902,0.030004038447188526,6.742263317108154 -ADRM1,11,0.01541415136307478,0.09579683840274811,0.9801858442624772,-0.04592670555823596,8.173428535461426 -ADSL,14,0.009553765878081322,0.07432369887828827,0.987610085972608,0.0008510785159658715,6.434751033782959 -AFG3L2,1,0.12381644546985626,0.2804323136806488,0.8713172629566291,0.0711974578450189,23.165069580078125 -AHCTF1,29,0.00526813231408596,0.05482714623212814,0.9929393317982305,0.0734185091249861,4.778293132781982 -AHCY,14,0.011654109694063663,0.08194632828235626,0.9849918619963535,-0.1479408251890873,7.106962203979492 -AIFM1,5,0.02577965334057808,0.12482752650976181,0.9673595107400184,-0.013078334428223587,10.570194244384766 -AKIRIN2,17,0.01231791265308857,0.08359944820404053,0.9836841710500104,0.13763652788955358,7.30656099319458 -ALDOA,8,0.021402178332209587,0.11111019551753998,0.9716815529081182,0.14411481118241495,9.6310453414917 -ALG1,7,0.020058033987879753,0.10958924144506454,0.9744528196378733,-0.014378131392994835,9.323707580566406 -ALG11,9,0.01451684720814228,0.09186592698097229,0.9808640750628942,0.06617775115390949,7.9319610595703125 -ALG13,17,0.007547727786004543,0.06699619442224503,0.9902480375055959,0.13530294798146558,5.719427585601807 -ALG14,19,0.007994959130883217,0.06791862845420837,0.9893325913195367,-0.08097806822534387,5.886437892913818 -ALG1L,21,0.006370738614350557,0.060911521315574646,0.9917084188557636,-0.07334633205176952,5.25459623336792 -ALG2,24,0.006016116589307785,0.05841308459639549,0.9919735739354169,0.04828362707933887,5.106256008148193 -ALYREF,32,0.004331962671130896,0.049789607524871826,0.9942878466844716,0.08029978263093332,4.332981109619141 -ANAPC1,13,0.01071059238165617,0.0783885046839714,0.9856349262089559,0.09060910412443819,6.813200950622559 -ANAPC10,3,0.054291363805532455,0.18016691505908966,0.932581637427162,0.029357138934374177,15.33945083618164 -ANAPC11,7,0.01996791921555996,0.10679923743009567,0.9749919101144499,-0.11368350804716476,9.302739143371582 -ANAPC13,11,0.013520990498363972,0.08954466134309769,0.9822448302050919,0.011475452469182048,7.655062198638916 -ANAPC15,43,0.0034186472184956074,0.044643957167863846,0.9954966087763796,-0.018801400535370667,3.849209785461426 -ANAPC2,23,0.006495322100818157,0.061276890337467194,0.9912370947421939,0.058062026396026034,5.305725574493408 -ANAPC4,7,0.02291696146130562,0.11607196927070618,0.9698376257314494,0.09202435857883508,9.966048240661621 -ANAPC5,4,0.03386085852980614,0.14191365242004395,0.9563023825870762,0.06805301435253913,12.114163398742676 -ANKLE2,12,0.011533836834132671,0.08157052844762802,0.9846883286587191,-0.0754683011483486,7.070194244384766 -ANKRD11,26,0.005552012473344803,0.05542207136750221,0.9930958350725074,-0.05156211616215593,4.905345916748047 -ANKRD49,15,0.00854994636029005,0.07156184315681458,0.989149766165636,-0.015562531658425617,6.087319850921631 -ANKS6,9,0.012761227786540985,0.0867835134267807,0.9835663251395697,-0.0724108580657472,7.436878204345703 -ANLN,9,0.013727898709475994,0.09016595035791397,0.9826844278821221,-0.07309328515806277,7.713411331176758 -AP2M1,21,0.006136603187769651,0.05960065498948097,0.9919000886054326,0.027436339728930996,5.157134532928467 -AP2S1,13,0.010296973399817944,0.07614157348871231,0.9863486147138469,0.14315088234091614,6.680350303649902 -AQP7,11,0.013555835001170635,0.08872555941343307,0.9819567119564693,-0.032854780303919404,7.664919376373291 -AQR,9,0.05984849855303764,0.1720883995294571,0.9215409391149711,0.31852234632684084,16.105384826660156 -ARCN1,34,0.004993450362235308,0.05388948693871498,0.9932724403517883,0.20183283006897368,4.652054309844971 -ARF4,9,0.014683516696095467,0.09213121980428696,0.9808425366598791,-0.00516257736264334,7.977365493774414 -ARFRP1,14,0.008548984304070473,0.07053353637456894,0.9892437888164726,-0.056249885246610905,6.086977481842041 -ARGFX,9,0.016297539696097374,0.09774208813905716,0.978595519823807,0.019969797114150978,8.404376029968262 -ARGLU1,50,0.0036339364014565945,0.044774580746889114,0.9953095253326797,0.06762625047314108,3.9685614109039307 -ARHGAP11B,21,0.00667208805680275,0.06258358806371689,0.9912125937024941,-0.08358272629915563,5.377437114715576 -ARID3A,3,0.04600442573428154,0.16874578595161438,0.9425130146427865,-0.06050633240120368,14.120311737060547 -ARIH1,6,0.022394416853785515,0.11411619186401367,0.9707821196949744,-0.04400902394064925,9.85177230834961 -ARL2,8,0.02062099799513817,0.10845502465963364,0.9730274824818621,0.09134796686570175,9.453644752502441 -ARL4D,12,0.011461790651082993,0.08211813867092133,0.9851086602745015,-0.031226691299441056,7.048077583312988 -ARMC6,8,0.016836417838931084,0.09971431642770767,0.978586133062653,-0.0659335060518555,8.542191505432129 -ARMC7,11,0.01614833064377308,0.09812456369400024,0.9787361084181025,0.24226936384426856,8.365815162658691 -ARPC2,22,0.006380603648722172,0.0608055479824543,0.991647264545908,-0.034482367921969403,5.258663177490234 -ARPC3,14,0.009536707773804665,0.07524697482585907,0.9876325294250473,-0.03064996869803074,6.429004192352295 -ARPC4,22,0.006727785337716341,0.06281974166631699,0.9909621281564449,0.019483313851818486,5.399835109710693 -ASF1B,29,0.004329347051680088,0.05082758143544197,0.994272567236529,0.01318330117487631,4.3316731452941895 -ATF4,7,0.018820341676473618,0.10469144582748413,0.9759746059074412,0.024495680770507296,9.031464576721191 -ATF5,14,0.023903165012598038,0.10954049229621887,0.9677496408248409,0.36214187006843696,10.178227424621582 -ATL2,12,0.012839747592806816,0.0868348777294159,0.9831451283649119,-0.006722605216117305,7.459722995758057 -ATP1A1,9,0.019927741959691048,0.10569170862436295,0.9736056301458119,0.06866683792388782,9.293375968933105 -ATP2A2,16,0.007766257971525192,0.06740591675043106,0.9901328962053589,0.04955201617900856,5.801634311676025 -ATP6AP1,12,0.011675801128149033,0.08295939862728119,0.9846476754825889,0.019961321847932728,7.11357307434082 -ATP6AP2,17,0.0076084742322564125,0.06688042730093002,0.9898596666881483,0.08967835010110355,5.742397308349609 -ATP6V0B,14,0.009830676950514317,0.07348943501710892,0.9870945114374237,0.03028616895727039,6.527338981628418 -ATP6V0C,6,0.023860475048422813,0.11773849278688431,0.968819422251902,0.17794181765655828,10.169135093688965 -ATP6V0D1,10,0.014394267462193966,0.09177026897668839,0.9811387448703389,0.049896946976863664,7.898402214050293 -ATP6V1A,9,0.013812488876283169,0.0888829305768013,0.9814843019650749,0.1016299844882077,7.7371392250061035 -ATP6V1B2,44,0.006155780516564846,0.055886201560497284,0.9917147202910003,0.29279892969228943,5.165186405181885 -ATP6V1C1,6,0.02052769809961319,0.11025868356227875,0.9737632483271126,-0.02381526567182505,9.432233810424805 -ATP6V1D,14,0.009141410700976849,0.07380060106515884,0.9878941328432483,0.08175874757529664,6.2943525314331055 -ATP6V1E1,19,0.007011131849139929,0.0643051266670227,0.9906261076371622,0.046357812384665145,5.512372016906738 -ATP6V1F,7,0.019343296065926552,0.10650280117988586,0.975397867258346,0.0504528141282524,9.156082153320312 -ATP6V1G1,6,0.024586891755461693,0.11961020529270172,0.9671522173218561,0.1299998899427792,10.322770118713379 -ATP6V1H,9,0.017297664657235146,0.10157852619886398,0.9779852750860405,0.044542660895703856,8.658411026000977 -ATR,14,0.012305594980716705,0.08448682725429535,0.9839781797141429,0.10031824611403778,7.3029069900512695 -ATRIP,19,0.009053817950189114,0.07247238606214523,0.9885361906138947,0.09750708598039652,6.264123439788818 -AURKA,6,0.026824887841939926,0.12117133289575577,0.9639319315780047,0.17791488133486627,10.782349586486816 -AURKAIP1,7,0.01921664923429489,0.10614003986120224,0.9751942297830809,-0.004611582808579947,9.126059532165527 -AURKB,4,0.046435218304395676,0.16670635342597961,0.9427981324112957,0.08396020384073355,14.186269760131836 -BANF1,16,0.009113206528127193,0.07309540361166,0.9878066684978523,-0.07126465565383515,6.284634590148926 -BANP,11,0.012464779429137707,0.08618389070034027,0.9835916817622137,0.013329639139759977,7.349989891052246 -BAP1,20,0.01054340973496437,0.07546579092741013,0.9868202306968215,-0.01493430916797509,6.759817600250244 -BARD1,14,0.008799473755061626,0.07163199782371521,0.9888393959643647,-0.056757610753835445,6.175509929656982 -BCAR1,24,0.005690652411431074,0.05725277587771416,0.9924134251039937,0.00584301119183653,4.966214656829834 -BCAS2,22,0.013426066376268864,0.08497907221317291,0.9824010299362979,0.2878109967494461,7.628143310546875 -BCL2L1,18,0.007736790459603071,0.06764713674783707,0.9897822960784431,-0.04183147846382494,5.790616989135742 -BCLAF1,5,0.02781074494123459,0.12957631051540375,0.9646356695328601,-0.03978486243494111,10.9786958694458 -BCR,56,0.020682796835899353,0.09400644898414612,0.9748400346619602,0.25716725708513094,9.46780014038086 -BCS1L,7,0.01747480221092701,0.10268959403038025,0.9774659747418198,0.02798588086879986,8.702630996704102 -BDP1,33,0.0067057302221655846,0.06050741299986839,0.9909787271109655,0.3649838954388669,5.390977382659912 -BET1,13,0.010156122967600822,0.07756375521421432,0.9863864663596223,-0.00914271089414955,6.6345038414001465 -BGLAP,23,0.006957449950277805,0.06338714808225632,0.9905992579174824,0.24240142919267052,5.4912285804748535 -BIRC5,4,0.030487749725580215,0.13393716514110565,0.9615000110728869,0.06499278963018296,11.494951248168945 -BMS1,3,0.051433734595775604,0.17313836514949799,0.9363992915786705,-0.0347686888354112,14.9302978515625 -BNIP1,11,0.012842416763305664,0.08537252247333527,0.9829690958897068,0.13345765750550945,7.460497856140137 -BOP1,26,0.00518438033759594,0.05547386035323143,0.9931753603776232,0.08852616628892111,4.740158557891846 -BORA,5,0.02191859297454357,0.11407896131277084,0.9715156313352747,0.12097430808518206,9.746546745300293 -BPTF,5,0.02463100105524063,0.12194496393203735,0.9699851645878566,0.0017475221790514135,10.332025527954102 -BRCA1,11,0.01140858605504036,0.08126005530357361,0.9852489215755833,0.0744827554220114,7.031700611114502 -BRCA2,14,0.009517275728285313,0.07550643384456635,0.9872581908514059,0.08186340854354582,6.422450065612793 -BRD4,11,0.011008874513208866,0.07985154539346695,0.9859361451319679,-0.08546619223866067,6.907420635223389 -BRD8,7,0.02141139842569828,0.1106811985373497,0.9724812666140755,0.14884467521112965,9.633119583129883 -BRF1,16,0.009831620380282402,0.07508523762226105,0.987075383027041,0.122508016733323,6.527652263641357 -BRF2,18,0.010048852302134037,0.07577089220285416,0.9867502829976127,-0.05472498416809682,6.5993733406066895 -BRIP1,13,0.011633962392807007,0.0834774449467659,0.9847100763421254,-0.04489552930375565,7.100816249847412 -BRIX1,13,0.014280482195317745,0.09103881567716599,0.9806727025034055,0.30714480498940505,7.867122173309326 -BRK1,18,0.006568075157701969,0.06145884469151497,0.9914205981553894,0.06593307357176467,5.335357189178467 -BTF3,8,0.01873019151389599,0.10388672351837158,0.9754671526158155,0.10890048332465288,9.009808540344238 -BTF3L4,14,0.009363595396280289,0.07436306774616241,0.9877301914280343,-0.07302485574365959,6.370386123657227 -BUB1,65,0.002222471870481968,0.03607209771871567,0.997030370457556,0.0538576147512798,3.1035776138305664 -BUB1B,14,0.009563026949763298,0.07448374480009079,0.9874886862290954,0.1032289622501697,6.437869071960449 -BUB3,41,0.005154861137270927,0.05279067903757095,0.9931466119690359,0.20724688620137074,4.726644515991211 -BUD13,18,0.016107549890875816,0.09480652958154678,0.9807735186945109,0.22079395495677331,8.355244636535645 -BUD31,3,0.043758541345596313,0.1614222377538681,0.9463528282068314,0.004637544380523225,13.771328926086426 -BYSL,10,0.02038004994392395,0.10865025222301483,0.9727070088073067,0.15117799117463215,9.3982515335083 -C12orf45,10,0.016242295503616333,0.09673938155174255,0.9784259318766505,0.062339946561743666,8.390119552612305 -C12orf60,11,0.015405731275677681,0.09243208169937134,0.979893303730505,0.11085799207680326,8.171195030212402 -C14orf178,15,0.017198028042912483,0.09668095409870148,0.977101638745286,0.2597869028078412,8.633438110351562 -C16orf86,19,0.007055935449898243,0.06522524356842041,0.9908764004738382,0.03923730186039124,5.529956817626953 -C17orf58,6,0.020380346104502678,0.11063820868730545,0.9742185139894453,-0.04165237808337527,9.398320198059082 -C19orf25,19,0.008781684562563896,0.07065650075674057,0.9884317680693283,-0.12339859249078877,6.169264316558838 -C19orf53,17,0.009123197756707668,0.0723043754696846,0.9877477602340954,-0.06420492293804037,6.288079261779785 -C1D,16,0.00806032307446003,0.06885946542024612,0.9895523477392973,-0.01487914921246681,5.910451889038086 -C1QBP,21,0.008110464550554752,0.06617829203605652,0.9890393610403645,0.08876154085810392,5.928807258605957 -C1QTNF4,20,0.006864179857075214,0.06314270943403244,0.990847648521951,-0.006353409908045279,5.454297065734863 -C1orf109,24,0.00577967194840312,0.05819856375455856,0.9923439510138145,0.008576315383030381,5.004907608032227 -C6orf15,17,0.008048242889344692,0.06880912184715271,0.9892003759458995,0.1658486851507057,5.9060211181640625 -C7orf26,8,0.019080661237239838,0.1036122739315033,0.9744182982422231,0.2100526913890548,9.093711853027344 -C7orf50,10,0.012773136608302593,0.08710359781980515,0.9839618529168016,-0.0680664282452757,7.440347671508789 -C9orf16,21,0.00683615542948246,0.06219237297773361,0.9908970747946788,-0.0007300058910170687,5.443151473999023 -C9orf78,20,0.006831009406596422,0.06230577081441879,0.9910989603112718,-0.08609913955663592,5.441102504730225 -CA5A,19,0.007935592904686928,0.06792835146188736,0.989461427707026,0.008683002474455722,5.864542484283447 -CACNB3,31,0.004284722730517387,0.0499381385743618,0.994524871039065,-0.050443750429271045,4.309290885925293 -CACTIN,6,0.02298167534172535,0.11733659356832504,0.9704122254006486,0.060420420159340835,9.980108261108398 -CADM4,38,0.003718117019161582,0.04625122994184494,0.9952584364424045,-0.07998042561635409,4.014264106750488 -CAMLG,10,0.012403384782373905,0.08500763773918152,0.9835823605177291,0.01429035551689753,7.33186674118042 -CAP1,2,0.05674498900771141,0.18410630524158478,0.9306392192669138,-0.021374025053784624,15.682244300842285 -CAPZB,9,0.01341930776834488,0.0882553905248642,0.9822594034458444,-0.04749240942096013,7.626223087310791 -CARS,2,0.09705902636051178,0.2432570904493332,0.8809631753327426,0.2078250854458442,20.509845733642578 -CASP8AP2,21,0.011216862127184868,0.07402036339044571,0.9849511199987377,0.09809668306775715,6.972365379333496 -CBLL1,10,0.01775427721440792,0.10097063332796097,0.9758416309578403,0.01544321386679969,8.771946907043457 -CBLN1,22,0.006201356183737516,0.06005789712071419,0.9919220354856175,-0.09396031181691984,5.184272289276123 -CBX1,15,0.008812055923044682,0.07190902531147003,0.9882560957838622,0.005357890074845288,6.1799235343933105 -CCAR1,5,0.0272845309227705,0.12753057479858398,0.964822707914769,0.059880221026386266,10.874335289001465 -CCDC115,21,0.00658445106819272,0.062342435121536255,0.9913281923542258,-0.04189724163540107,5.342004299163818 -CCDC130,12,0.01064610667526722,0.07834286242723465,0.9860409101278513,0.03355739288028212,6.792659759521484 -CCDC137,20,0.007357602473348379,0.06561125814914703,0.9905138990834564,0.0231049611797796,5.646932601928711 -CCDC144A,11,0.01345409918576479,0.08966420590877533,0.98254228282885,-0.03407086589383789,7.636102676391602 -CCDC144NL,8,0.015740565955638885,0.09633241593837738,0.9794072210080725,0.0819229914081084,8.259515762329102 -CCDC174,11,0.018238823860883713,0.1008637398481369,0.9759414770672803,0.1807094514617499,8.890841484069824 -CCDC59,13,0.011500195600092411,0.08159051835536957,0.9846647438808019,0.07734638536116853,7.059875965118408 -CCDC78,14,0.01257880125194788,0.08397818356752396,0.9830669586699589,0.05429536876696867,7.383531093597412 -CCDC84,8,0.019031710922718048,0.10415326803922653,0.9750085665444937,0.19039609931356127,9.082038879394531 -CCDC86,26,0.008671213872730732,0.06927812099456787,0.9886078667824352,0.28225258687322935,6.130337715148926 -CCNA2,3,0.0429440513253212,0.1625068187713623,0.9463949132319578,0.007082734184110824,13.642563819885254 -CCNB1,16,0.009573479183018208,0.07521525025367737,0.9872033506962595,0.033743195148326625,6.4413862228393555 -CCND1,14,0.008479264564812183,0.07156919687986374,0.9890766122283816,0.05921714502994119,6.062106609344482 -CCNH,8,0.025749007239937782,0.12183890491724014,0.964963750656074,0.10926716179868329,10.563909530639648 -CCNK,32,0.0053954897448420525,0.054415348917245865,0.9927807125486001,0.15313178205105693,4.835705757141113 -CCNL1,17,0.008036711253225803,0.06817232817411423,0.9896115730763473,0.006726780999472924,5.901788234710693 -CCP110,4,0.03292644023895264,0.14179809391498566,0.9590559010968781,-0.09590368271189754,11.945844650268555 -CCT2,20,0.01414966769516468,0.08458966761827469,0.9810417741745613,0.3483528284270427,7.831006050109863 -CCT3,24,0.010707643814384937,0.07418424636125565,0.9859867032057231,0.25468930286482394,6.812263011932373 -CCT4,25,0.01107934582978487,0.07661832869052887,0.9849781910103511,0.3668138785546828,6.9294939041137695 -CCT5,6,0.031161397695541382,0.1317942589521408,0.959096772543184,0.14860695472598284,11.62125301361084 -CCT6A,13,0.012130632065236568,0.08107016980648041,0.9836652806049418,0.2720029292990897,7.2508039474487305 -CCT7,15,0.015021526254713535,0.08875061571598053,0.9797019791951049,0.2535885043644954,8.068660736083984 -CCT8,11,0.014230211265385151,0.08943608403205872,0.9810053815014648,0.2097867152496798,7.853262901306152 -CD2BP2,13,0.009997227229177952,0.07665647566318512,0.9871096864951704,-0.02210784334387919,6.582399368286133 -CD3EAP,9,0.018623562529683113,0.1028490960597992,0.9749371771878201,0.19903943742451288,8.984126091003418 -CD8B,11,0.010520590469241142,0.0790482833981514,0.9863136791223743,0.00331261043928138,6.752499103546143 -CDAN1,13,0.012241329066455364,0.08469442278146744,0.9834927756932547,0.03744386471982456,7.283812046051025 -CDC123,11,0.012311790138483047,0.08396130800247192,0.9834427523004436,0.19861209696489288,7.304745197296143 -CDC16,17,0.009304235689342022,0.07273898273706436,0.9878957910022095,0.15477397530796924,6.350162029266357 -CDC20,11,0.013531670905649662,0.0871841236948967,0.9826877702156323,0.10060884391654543,7.658084869384766 -CDC23,17,0.008615334518253803,0.06945966929197311,0.9887609579071702,0.026852771461528032,6.11055326461792 -CDC26,5,0.02274838089942932,0.11633272469043732,0.9713530224001184,-0.009905587402485407,9.92932415008545 -CDC27,36,0.0037638735957443714,0.046130985021591187,0.9949622008898773,0.08543827317549457,4.038889408111572 -CDC37,18,0.007572390604764223,0.06679655611515045,0.9899929245195993,-0.0655627947516,5.728764533996582 -CDC40,10,0.012284833006560802,0.08590751141309738,0.9838623652527785,0.05059362803025095,7.296743869781494 -CDC42,7,0.018831763416528702,0.1056598648428917,0.9751629287244441,0.11195309177112572,9.034205436706543 -CDC45,9,0.01328447088599205,0.08892374485731125,0.9828433292898584,-0.003913887904691293,7.587812423706055 -CDC5L,10,0.027776505798101425,0.12324268370866776,0.964422679088338,0.26784639628013673,10.971936225891113 -CDC6,19,0.010805702768266201,0.07679968327283859,0.9853682092736449,0.2658530138344715,6.843384742736816 -CDC7,9,0.01478793565183878,0.09299349039793015,0.9806051520103627,0.027202806822890984,8.005680084228516 -CDC73,17,0.02410680428147316,0.11338350176811218,0.9705500752962963,0.14304560529652194,10.221491813659668 -CDCA5,6,0.023800378665328026,0.11907057464122772,0.9688446825548165,0.03599484022375222,10.156320571899414 -CDCA8,8,0.019214589148759842,0.10660550743341446,0.9740690176686528,0.10561441460973857,9.125569343566895 -CDIPT,24,0.006071694660931826,0.05979900434613228,0.9919892874422345,0.1999121791218534,5.129787921905518 -CDK1,34,0.00665279570966959,0.05963372811675072,0.9911132306943143,0.24874031664661964,5.369657039642334 -CDK11A,27,0.0045869131572544575,0.05236801505088806,0.9939159564827276,0.059097877678976604,4.458663463592529 -CDK2,11,0.013042493723332882,0.08781324326992035,0.9826472887090288,0.020154324921367487,7.518388748168945 -CDK7,2,0.076559878885746,0.2175167053937912,0.9066087295654098,-0.007375777872997232,18.215665817260742 -CDK9,14,0.014430089853703976,0.09030653536319733,0.9814557602228333,0.0715696726955466,7.908224105834961 -CDT1,3,0.07504286617040634,0.20326422154903412,0.8985608197545062,0.2630037487910186,18.03429412841797 -CEBPZ,12,0.01111766416579485,0.07971245050430298,0.985379869729687,0.21722984362984268,6.941466331481934 -CENPA,7,0.01757996715605259,0.10221074521541595,0.9780819163198662,0.013577339292470675,8.728777885437012 -CENPC,13,0.012539991177618504,0.08560173958539963,0.9839163170322924,0.15226649882914253,7.37213134765625 -CENPE,17,0.01055737491697073,0.07652082294225693,0.985726181244504,0.13290840521554131,6.764293670654297 -CENPH,7,0.01976928859949112,0.10724319517612457,0.9750697361415135,-0.031161810228814393,9.256354331970215 -CENPI,4,0.043350715190172195,0.16121190786361694,0.9453701255049763,0.05952640632402932,13.707005500793457 -CENPJ,28,0.008636057376861572,0.06611718237400055,0.9884983314931297,0.2759192371338281,6.1178975105285645 -CENPK,9,0.01843087002635002,0.10417770594358444,0.9755742857093043,0.09195178956461801,8.937527656555176 -CENPN,14,0.00939902663230896,0.07454031705856323,0.9876980501320789,0.026464192499027754,6.38242769241333 -CENPT,7,0.021512918174266815,0.11259979009628296,0.971524187796703,0.06791921265646295,9.655929565429688 -CENPW,7,0.018753355368971825,0.10335826128721237,0.9766725793952705,-0.10672266980285346,9.01537799835205 -CEP152,16,0.008787637576460838,0.07181021571159363,0.988320603069638,0.08438457121608363,6.1713547706604 -CEP192,20,0.00864451751112938,0.07050101459026337,0.9885917050095895,0.12174319390552556,6.120893478393555 -CEP68,7,0.018432771787047386,0.10355547815561295,0.976349951991166,0.10920753064599949,8.93798828125 -CEP85,13,0.010917684994637966,0.07900331914424896,0.9854065158473825,0.03255828640189257,6.878753185272217 -CEP97,13,0.010184784419834614,0.07811582088470459,0.9868396418595246,0.08322444027424049,6.643858432769775 -CFDP1,15,0.008900316432118416,0.0719185471534729,0.988808906900243,-0.12385788140163374,6.210794448852539 -CFL1,3,0.03564678505063057,0.14513790607452393,0.9576389550823176,-0.04556693634521155,12.42952823638916 -CHAF1A,4,0.04062042012810707,0.15220877528190613,0.9489835343242822,0.09296704013730052,13.268342018127441 -CHAF1B,9,0.02924823760986328,0.12465804070234299,0.9618830682042829,0.1447305875857234,11.258856773376465 -CHCHD1,9,0.0147565808147192,0.09339173138141632,0.9808807356161388,0.09156177409779133,7.997188568115234 -CHCHD2,9,0.013384629040956497,0.0872490406036377,0.9829122019651312,0.022499998550087216,7.616363048553467 -CHCHD4,14,0.011375222355127335,0.08092182129621506,0.9853910101494733,0.22225589605017673,7.021410942077637 -CHD4,5,0.030184131115674973,0.13390597701072693,0.9599999057887837,0.15710360464897888,11.43757152557373 -CHEK1,29,0.00838257186114788,0.06888259947299957,0.9903208124105746,0.1372829682876478,6.027442932128906 -CHERP,6,0.02494778297841549,0.12061222642660141,0.9683973733476385,0.10157757961917634,10.39825439453125 -CHMP1A,8,0.014527383260428905,0.09347396343946457,0.9813139780183976,0.012672575356966675,7.934839248657227 -CHMP2A,10,0.03673338145017624,0.13411471247673035,0.9504945127980404,0.2116540340246448,12.617545127868652 -CHMP3,13,0.04503387585282326,0.14197103679180145,0.9383710672156395,0.2834530098325536,13.970569610595703 -CHMP4B,7,0.020572395995259285,0.11167100071907043,0.972230884819425,0.12279804090778486,9.442498207092285 -CHMP5,11,0.016304705291986465,0.09803806245326996,0.9789434641724734,0.035180389008768166,8.406224250793457 -CHMP6,63,0.006149865221232176,0.05412572994828224,0.9920307686188197,0.27037373919317076,5.162703990936279 -CHMP7,24,0.005501494277268648,0.05712941288948059,0.9926848912426375,0.021094917679146924,4.8829779624938965 -CHORDC1,26,0.0050054145976901054,0.05402704328298569,0.9935179575510855,-0.04255273278121499,4.657624244689941 -CHTF18,9,0.015881149098277092,0.09612690657377243,0.9792935191771928,0.060445152786781996,8.296319007873535 -CHTF8,5,0.02606777846813202,0.12348851561546326,0.966507059543254,0.0384218279170184,10.629098892211914 -CHTOP,15,0.009804481640458107,0.07596088200807571,0.9874153358185116,-0.08920001235768595,6.518636703491211 -CIAO1,25,0.00574201624840498,0.057663168758153915,0.9924818037473447,0.01275535095589398,4.988576889038086 -CIAPIN1,15,0.012017321772873402,0.08163496106863022,0.9838218440290125,0.1456540467711338,7.216860771179199 -CINP,8,0.018429413437843323,0.10445236414670944,0.9757815871863079,0.023997577179535854,8.937173843383789 -CKAP5,31,0.004567732103168964,0.0519305095076561,0.9939983819119871,-0.06442245661004643,4.449331283569336 -CLASRP,12,0.012392416596412659,0.08597123622894287,0.9844917114915588,0.14264023253614674,7.328624248504639 -CLCC1,4,0.04288698360323906,0.161157488822937,0.9451004976252899,0.00852113338113242,13.63349437713623 -CLK2,9,0.014354381710290909,0.09235324710607529,0.9817377925119616,-0.07185630124109982,7.887451648712158 -CLNS1A,6,0.024271948263049126,0.11900738626718521,0.9694034070126679,0.08419870944585148,10.256443977355957 -CLP1,5,0.028505908325314522,0.1277405172586441,0.963343283137863,0.11498414227358475,11.115062713623047 -CLSPN,14,0.010602695867419243,0.07729993760585785,0.9856324176709306,0.0763848852334703,6.778796672821045 -CLTC,15,0.0098286597058177,0.0756203904747963,0.9868111476024347,0.12132205708669054,6.526669025421143 -CMPK1,18,0.007843263447284698,0.06746003031730652,0.9894931017816271,0.07721104747113904,5.830326080322266 -CMTR1,15,0.020419742912054062,0.09898678213357925,0.9731704852840769,0.021330821344649826,9.407400131225586 -CNIH4,6,0.0229480117559433,0.11669015139341354,0.9698506609410291,0.05040956928054503,9.972796440124512 -CNN2,19,0.006670043803751469,0.06210426986217499,0.9912309725021492,-0.014222940383528846,5.376613616943359 -CNOT1,27,0.024950461462140083,0.11026454716920853,0.9686242932775813,0.22729973246567145,10.398812294006348 -CNOT2,9,0.01912936009466648,0.10485663264989853,0.9740446739036056,0.15378740982059574,9.105308532714844 -CNOT3,33,0.009633198380470276,0.0713261142373085,0.9871099158478861,0.17461874754214177,6.461445331573486 -COA5,17,0.00820416584610939,0.06982322037220001,0.9896147978287436,-0.011234941934876098,5.962957382202148 -COASY,24,0.005835831165313721,0.05825291574001312,0.9922964883847079,-0.05937069207624073,5.0291643142700195 -COG1,4,0.0292523056268692,0.132112056016922,0.963175292038148,-0.08348659111074995,11.259639739990234 -COG2,35,0.004447328858077526,0.050241753458976746,0.9941098285486384,-0.03806711935348432,4.390298366546631 -COG3,5,0.03402969613671303,0.14106100797653198,0.9553368284956044,0.1303654407570881,12.144328117370605 -COG4,13,0.010166537947952747,0.07661908119916916,0.986914009161931,-0.028254935299586826,6.637904644012451 -COG6,19,0.008768304251134396,0.07159408926963806,0.988367889364558,0.03078388195446583,6.164562702178955 -COG8,4,0.04947654530405998,0.1719999760389328,0.9368506156905348,0.1839316750602355,14.643474578857422 -COPA,25,0.006290513090789318,0.05966833606362343,0.9916053228315386,0.19834163505798064,5.22140645980835 -COPB1,9,0.028484459966421127,0.12378429621458054,0.9618583703735725,0.34290582943781245,11.110879898071289 -COPB2,13,0.010881171561777592,0.08033191412687302,0.9854357647773314,0.1359711470133785,6.867240905761719 -COPE,15,0.00956574734300375,0.07386400550603867,0.987525807803299,-0.04030701490340177,6.438785076141357 -COPG1,26,0.005207505542784929,0.05543972924351692,0.9931699631713043,0.006539471183058806,4.75071907043457 -COPS2,4,0.044404659420251846,0.16319973766803741,0.9456374762734135,0.12143718907403288,13.872628211975098 -COPS3,8,0.022414693608880043,0.11520327627658844,0.9705390142199213,0.14568476070608075,9.856230735778809 -COPS4,26,0.006750303320586681,0.06236898526549339,0.9911863881086763,0.210566241805908,5.408864498138428 -COPS5,10,0.01446773111820221,0.09219173341989517,0.9815952763899172,-0.015355783560785874,7.918531894683838 -COPS6,23,0.007562835235148668,0.0649786964058876,0.9900485177980203,0.13244255197988114,5.725148677825928 -COPS8,23,0.007579904515296221,0.06609731167554855,0.9901687478662898,0.02934923346809194,5.7316060066223145 -COPZ1,20,0.008055047132074833,0.06870628893375397,0.9892372538933418,0.12404489585605072,5.908517360687256 -COQ2,6,0.019758226349949837,0.10882985591888428,0.9750091000426638,-0.0624993249078238,9.253765106201172 -COQ4,16,0.008988362736999989,0.07222967594861984,0.9884740750811961,-0.07615725022143517,6.241439342498779 -COQ5,12,0.010460923425853252,0.07829247415065765,0.9863558619674933,-0.0377936518383259,6.733323097229004 -COX10,10,0.015126874670386314,0.09407422691583633,0.9803609641094126,0.12534218661322394,8.096905708312988 -COX11,11,0.014829476363956928,0.09200232475996017,0.98041315003284,0.01909104338894704,8.016916275024414 -COX15,11,0.016960956156253815,0.09907909482717514,0.9787511071751074,0.21000838477688294,8.573726654052734 -COX17,15,0.009326677769422531,0.07303453236818314,0.9876620907723115,-0.047303855852912746,6.357816219329834 -COX5A,9,0.014720418490469456,0.09136030077934265,0.9803702625918274,0.14997410754771237,7.987383842468262 -COX5B,12,0.013649219647049904,0.08780402690172195,0.9818636568757286,0.05819044198752204,7.691275596618652 -COX6C,11,0.012548975646495819,0.0845603197813034,0.9836898645165948,0.055821076504458896,7.374772071838379 -COX7B,12,0.010776778683066368,0.07856390625238419,0.9860719522534239,0.00843705094039476,6.834219455718994 -COX7C,17,0.009817213751375675,0.07354464381933212,0.9871264761528228,0.05084885389688049,6.522867679595947 -CPAMD8,24,0.005780093837529421,0.058082278817892075,0.9923291202067829,-0.03015304215897969,5.005090236663818 -CPNE7,10,0.012841364368796349,0.08633936196565628,0.9839528770454514,-0.023172402176196878,7.460192680358887 -CPOX,10,0.012388649396598339,0.08536788076162338,0.9839812191222311,-0.030098230264037257,7.327509880065918 -CPSF1,5,0.02822510153055191,0.12792843580245972,0.9626987976047447,-0.07554344032604567,11.0601806640625 -CPSF2,2,0.11550808697938919,0.26397693157196045,0.8605169746278949,0.044798263457121165,22.37436294555664 -CPSF3,62,0.008148512803018093,0.061622243374586105,0.9893971361435094,0.12344246241913562,5.942698001861572 -CPSF4,6,0.030695822089910507,0.13213199377059937,0.9596787482944623,0.128700605444305,11.534110069274902 -CPSF6,5,0.040085598826408386,0.1523406058549881,0.9459612133888353,0.11364178749350407,13.180705070495605 -CRCP,16,0.008918981067836285,0.07222964614629745,0.9881102407827619,0.07369368524264197,6.217303276062012 -CRKL,12,0.010692628100514412,0.07954613864421844,0.9860649888158177,0.05177613873108138,6.807485103607178 -CRLS1,11,0.011782522313296795,0.08448977768421173,0.9853506886766882,0.017798098376778863,7.14600944519043 -CRNKL1,26,0.007042958401143551,0.06345725804567337,0.9912644284459929,0.1352551348596054,5.524869441986084 -CS,22,0.006500997114926577,0.06005767732858658,0.9918933794458408,-0.06331071635126928,5.308043479919434 -CSE1L,43,0.02075829543173313,0.09633121639490128,0.9718617366211294,0.19836032290014083,9.485064506530762 -CSH2,18,0.007265095133334398,0.06479179114103317,0.9905120858543297,0.004782693815248225,5.611320972442627 -CSNK1A1,17,0.011012221686542034,0.07963982224464417,0.9852350395209102,0.20102940976634895,6.90847110748291 -CSNK2B,13,0.014758712612092495,0.09020056575536728,0.9803827244198852,0.06127435731332818,7.997766017913818 -CSTF1,14,0.012163014151155949,0.08304063975811005,0.9835318827302097,0.08200442609573949,7.2604756355285645 -CSTF3,6,0.03125189617276192,0.1353764533996582,0.9607371847785194,0.06316043658915732,11.638115882873535 -CTBP2,11,0.012167864479124546,0.08437091112136841,0.9842486660869385,-0.032386501838638504,7.261922359466553 -CTCF,7,0.02265108749270439,0.11493463814258575,0.9705955027150197,0.040422537650895075,9.908068656921387 -CTDP1,11,0.015708239749073982,0.0949549749493599,0.9789627354405187,0.009142524407812588,8.251030921936035 -CTNNBL1,7,0.0221983902156353,0.11220086365938187,0.9714724844838066,0.016366058060904368,9.808558464050293 -CTPS1,2,0.05528148263692856,0.1842549443244934,0.9362719997706522,-0.031470838200691124,15.478693008422852 -CTR9,21,0.02347881905734539,0.10941535979509354,0.9737407854648884,0.13445174847039823,10.087477684020996 -CTU2,12,0.010882765986025333,0.08038758486509323,0.9859104663929377,0.040545111066358046,6.867743968963623 -CUL1,23,0.0066504101268947124,0.06206871196627617,0.9911556496512716,-0.1108834061455592,5.368693828582764 -CUL2,10,0.016351476311683655,0.09808827936649323,0.9787157819805024,0.16093133152963277,8.4182710647583 -CUL3,12,0.010676570236682892,0.07900143414735794,0.9864492748255179,-0.019201140921655632,6.802371501922607 -CUL7,4,0.04079669713973999,0.15701811015605927,0.9480104949949096,0.12794811352999033,13.297100067138672 -CWC15,19,0.01618790626525879,0.09324249625205994,0.9794169884771539,0.26041684692551337,8.376060485839844 -CWC22,4,0.03058939427137375,0.13584226369857788,0.9621248274624292,-0.012602035624895636,11.514097213745117 -CWC25,15,0.008124789223074913,0.0699605941772461,0.989380329481487,0.027472941519254366,5.934040546417236 -CWF19L2,20,0.011736302636563778,0.0799933597445488,0.9848212824957857,0.142850033493051,7.131979942321777 -CXXC1,12,0.014952205121517181,0.09146712720394135,0.9798235064863775,0.22674475088071258,8.05002212524414 -CYC1,10,0.012290380895137787,0.08429068326950073,0.9837847965243175,0.11160665304092891,7.298390865325928 -CYCS,17,0.00832119956612587,0.06750795990228653,0.9891193072570136,-0.05839750187457579,6.005337715148926 -CYFIP1,11,0.0155991455540061,0.0957501158118248,0.9795158077453042,0.1424292635221638,8.222329139709473 -CYP2A13,14,0.011141238734126091,0.08073166757822037,0.9857035390896398,0.18298808646536502,6.948822021484375 -CYP4F11,12,0.010583894327282906,0.07927445322275162,0.986202811502309,0.0018962699681642525,6.7727837562561035 -CYS1,17,0.008082454092800617,0.06837550550699234,0.9893402484202536,-0.013704342198746076,5.918560028076172 -DAD1,8,0.022763052955269814,0.11146336793899536,0.9700085870288172,0.032077891555455644,9.932526588439941 -DAP3,16,0.00803782232105732,0.06767430156469345,0.9898572968274121,0.017198533471515145,5.902196884155273 -DARS,7,0.032182034105062485,0.13556283712387085,0.960670295439314,0.2403060287908832,11.810035705566406 -DBF4,13,0.010253546759486198,0.077600859105587,0.9870537417747729,-0.10301801351141696,6.666248321533203 -DBR1,10,0.0189361609518528,0.10574611276388168,0.9777557718291559,0.21028636231042308,9.059211730957031 -DCAF6,13,0.00904667004942894,0.07273188978433609,0.9882261287456896,-0.02289738175500336,6.261651039123535 -DCLRE1B,11,0.011184613220393658,0.0813104584813118,0.9855547660732964,-0.04483432128658499,6.962335109710693 -DCTN1,19,0.0072094169445335865,0.06432802975177765,0.9906610407891816,0.025728319887981597,5.58977746963501 -DCTN2,12,0.013098887167870998,0.0872543454170227,0.9824021190477966,0.07633126871936635,7.534625053405762 -DCTN3,20,0.0058861845172941685,0.05864514410495758,0.9924372276387399,0.09235357460366844,5.050814151763916 -DCTN4,7,0.01941116526722908,0.10740691423416138,0.9742135200238249,0.10056705706759239,9.172131538391113 -DCTN5,14,0.01131448708474636,0.08122843503952026,0.9847555557510603,0.13378528118572644,7.002641677856445 -DCTN6,7,0.01737283542752266,0.10093216598033905,0.977888920818555,-0.024216960968023327,8.677204132080078 -DCUN1D5,19,0.008912207558751106,0.07110095769166946,0.9880516991843409,0.1752182100956953,6.214942455291748 -DDB1,17,0.008038091473281384,0.06868202984333038,0.9892557198107297,0.10937440721243076,5.902295112609863 -DDN,25,0.005469483323395252,0.05685753747820854,0.9927001074903504,0.1500996165976719,4.868751049041748 -DDOST,8,0.018361009657382965,0.10327047109603882,0.9763332495911352,0.07757621255744242,8.920572280883789 -DDX1,10,0.011833074502646923,0.08227292448282242,0.9848551730337487,-0.024406437987673574,7.161323070526123 -DDX10,11,0.016991473734378815,0.09857819974422455,0.9772640113874627,0.11339437250134717,8.581436157226562 -DDX11,13,0.010855766013264656,0.08008064329624176,0.9860398009191715,0.04193694341328611,6.859219074249268 -DDX18,13,0.0198943130671978,0.10462801903486252,0.9733009570096517,0.2975642692578666,9.285577774047852 -DDX19A,11,0.014800095930695534,0.09270365536212921,0.98031248397067,-0.014038600021493226,8.008971214294434 -DDX19B,29,0.004794955253601074,0.05307953059673309,0.9936277516334839,0.011438938911145826,4.558655261993408 -DDX20,12,0.01457477081567049,0.09081350266933441,0.981237131686071,0.10818335925303188,7.947770595550537 -DDX21,9,0.01672760397195816,0.09719282388687134,0.9785928057778311,0.07897295991851304,8.514543533325195 -DDX23,16,0.013178586959838867,0.08665511757135391,0.9826902895593761,0.11710407297748017,7.557512283325195 -DDX24,28,0.009123110212385654,0.06936890631914139,0.9886788521902636,0.32380599697030577,6.28804874420166 -DDX27,5,0.030739586800336838,0.13554395735263824,0.9611537741353363,0.11706727686358173,11.542329788208008 -DDX3X,3,0.03776867687702179,0.15055245161056519,0.9526003219805157,0.051265015231201035,12.79411792755127 -DDX41,29,0.015287528745830059,0.09012597799301147,0.9818919313977972,0.17383407197940187,8.139787673950195 -DDX42,5,0.029363665729761124,0.13059435784816742,0.9623015403660614,0.030230762082797992,11.281051635742188 -DDX46,22,0.007676434703171253,0.06565287709236145,0.9899601600032748,0.2195552589864015,5.767986297607422 -DDX47,32,0.016689203679561615,0.09268156439065933,0.9794103699532276,0.2485512954599154,8.50476360321045 -DDX49,5,0.033447299152612686,0.14157117903232574,0.9561656575060148,0.04450858884823823,12.039958953857422 -DDX5,15,0.009234024211764336,0.07177913188934326,0.9885880739975059,-0.05133247269862006,6.3261566162109375 -DDX51,13,0.012375231832265854,0.08528585731983185,0.9837917695686812,0.18677638435474836,7.323541164398193 -DDX52,11,0.015109035186469555,0.09340666979551315,0.9797211621619432,0.00997683898090606,8.09212875366211 -DDX54,9,0.018377337604761124,0.10393974184989929,0.9756810687502879,0.21691331228513086,8.924538612365723 -DDX55,17,0.00819108821451664,0.06851963698863983,0.9892457071740082,0.09855040969661376,5.958202362060547 -DDX56,18,0.015472088009119034,0.08872052282094955,0.9801507502863726,0.3245607382178063,8.188775062561035 -DDX59,16,0.008899521082639694,0.07182732224464417,0.988229168930511,0.07447756135664117,6.210517406463623 -DDX6,7,0.02925017476081848,0.12772199511528015,0.9618469577988381,0.2801656317129501,11.25922966003418 -DENR,25,0.006664230488240719,0.06164083257317543,0.991081510609969,0.1437512498371802,5.374269485473633 -DERL2,13,0.010237964801490307,0.0767744779586792,0.9862887531861482,0.14220394923363763,6.6611809730529785 -DHDDS,9,0.024554558098316193,0.11393924057483673,0.9684760503801935,0.14014746725095537,10.315980911254883 -DHFR,13,0.00962214544415474,0.07523586601018906,0.9875950414643024,-0.0643666787828743,6.457737922668457 -DHODH,23,0.0059454734437167645,0.058856748044490814,0.9920863229839283,0.022386441726137556,5.076187610626221 -DHPS,6,0.020513668656349182,0.10980384796857834,0.9746197342049486,-0.035014862996705774,9.429010391235352 -DHX15,36,0.022040968760848045,0.10287706553936005,0.9721603823433818,0.36802364550111133,9.773717880249023 -DHX16,33,0.0176350437104702,0.09638587385416031,0.97955512264377,0.2037135378913066,8.74244213104248 -DHX33,14,0.014113102108240128,0.0901549682021141,0.9810002828346845,0.176467458005647,7.820880889892578 -DHX36,27,0.0076646446250379086,0.06523089110851288,0.9898366899863936,0.09791285767247315,5.763555526733398 -DHX37,21,0.009080746211111546,0.07146753370761871,0.9877770356769268,0.15534112361039487,6.27343225479126 -DHX8,11,0.03259476646780968,0.13482829928398132,0.9618053450556219,0.2545183419693501,11.88552474975586 -DHX9,9,0.014883769676089287,0.09273820370435715,0.9806172506890678,0.15437294325802758,8.03157901763916 -DICER1,16,0.009750258177518845,0.07510875910520554,0.9868810426486825,0.01752229689417775,6.500586032867432 -DIMT1,10,0.022452300414443016,0.11252474784851074,0.9693138639949554,0.1464596132035645,9.864496231079102 -DIS3,5,0.035302456468343735,0.141639843583107,0.9526837822218033,0.0842834754312918,12.36935043334961 -DKC1,6,0.027919529005885124,0.127865269780159,0.9642027907440465,-0.05411913564088104,11.000146865844727 -DLD,31,0.01281535904854536,0.07851695269346237,0.9826202912183863,0.225912820458589,7.452635288238525 -DMAP1,8,0.031902629882097244,0.12668892741203308,0.958835710825542,0.1356513287585402,11.758655548095703 -DMRTA2,10,0.012392479926347733,0.08483957499265671,0.9837071729137696,0.01745718505740949,7.328642845153809 -DNAJA1,29,0.005084632895886898,0.053607527166604996,0.9931384549864826,0.0808587981578446,4.694336891174316 -DNAJA3,12,0.017032083123922348,0.098175048828125,0.9778798394044359,0.2074156735288722,8.59168529510498 -DNAJC11,4,0.03617148473858833,0.14784346520900726,0.9519500342323176,0.13822110178699298,12.520671844482422 -DNAJC17,3,0.0569748692214489,0.18440553545951843,0.9327627121495803,0.11259281266943044,15.713977813720703 -DNAJC19,40,0.021499134600162506,0.09908142685890198,0.9716045170776001,0.4430444957399697,9.652835845947266 -DNAJC8,27,0.005257114302366972,0.05402132123708725,0.9930305384308874,0.09801159418115296,4.773293972015381 -DNAJC9,9,0.014352180063724518,0.09152892231941223,0.9822981144570387,-0.10508697691702809,7.886846542358398 -DNLZ,14,0.009259858168661594,0.07385297864675522,0.987837541222247,0.027345879644085216,6.335000514984131 -DNM1,14,0.010385952889919281,0.07746147364377975,0.9862159889299502,0.036346427693500376,6.709151744842529 -DNM1L,17,0.011548592709004879,0.08162899315357208,0.9861322743276556,0.13310805096204953,7.074716091156006 -DNM2,1,0.12399814277887344,0.2801773250102997,0.8757595733883902,0.12968877431344,23.18206024169922 -DNMT1,17,0.00993682723492384,0.075038842856884,0.9870927083086274,-0.13838752811111735,6.5624847412109375 -DNTTIP2,27,0.014174547046422958,0.08566940575838089,0.9817610937674033,0.2638682128180798,7.837888240814209 -DOHH,15,0.011538438498973846,0.08089170604944229,0.9844493738944852,0.09297768002023249,7.071604251861572 -DOLK,14,0.009683072566986084,0.07554890215396881,0.9871543066747589,-0.015526222505927193,6.478151321411133 -DONSON,75,0.004912941250950098,0.049341022968292236,0.9936091703227198,0.28146117181870434,4.6143999099731445 -DPAGT1,13,0.009522243402898312,0.0749272033572197,0.987519307372165,-0.0017526998093684374,6.424126625061035 -DPH1,33,0.004537842236459255,0.05168827250599861,0.9940818641340121,0.04545380720925052,4.434750080108643 -DPH2,17,0.008398453705012798,0.07021420449018478,0.9887622306947924,0.09541109128841875,6.0331501960754395 -DPH3,6,0.020129255950450897,0.10911543667316437,0.9751890837055216,-0.07404218757327871,9.340246200561523 -DPH6,10,0.012679723091423512,0.08588350564241409,0.98334730722611,0.031571587340851415,7.41309118270874 -DPM2,16,0.007720068097114563,0.0681065246462822,0.9897736675795272,0.0693078535754552,5.784355640411377 -DPPA2,6,0.028756877407431602,0.13136649131774902,0.9644111915308243,0.05328101742633256,11.163884162902832 -DPY19L2,5,0.02635781466960907,0.12518630921840668,0.9671562112957838,0.020705528957221702,10.688066482543945 -DR1,12,0.011915798299014568,0.08436983823776245,0.985173066322373,-0.10805145725380678,7.1863112449646 -DRAP1,22,0.006602107081562281,0.06106971576809883,0.9912918841548778,0.11378465261028214,5.3491621017456055 -DRG1,19,0.007345302030444145,0.06571545451879501,0.9905252396743721,0.027808856706176648,5.642210483551025 -DSN1,25,0.011642280034720898,0.07870256155729294,0.9857448019537937,0.3128732141796463,7.103353977203369 -DTL,13,0.01229444332420826,0.0835144892334938,0.9833315964336599,0.19845933913058358,7.299597263336182 -DTYMK,12,0.010130256414413452,0.07705136388540268,0.9866128178904273,0.0025930605383404563,6.626049995422363 -DUT,5,0.02762659452855587,0.12647342681884766,0.9634520556139078,0.12802589896921074,10.942288398742676 -DYNC1H1,29,0.008753318339586258,0.06879008561372757,0.9890466127984127,0.2575991420837141,6.159292221069336 -DYNC1I2,30,0.00495097367092967,0.05339871346950531,0.9933263003218582,0.13674308051411657,4.63222599029541 -DYNLL1,7,0.021472007036209106,0.11101928353309631,0.9720218974158199,0.01300724104432832,9.646744728088379 -DYNLRB1,28,0.004701188299804926,0.052259452641010284,0.9936552580500448,0.12642847908886307,4.513862133026123 -E2F6,15,0.009424538351595402,0.07483632862567902,0.9873073744859818,-0.016881436909233064,6.391083717346191 -E4F1,14,0.011722457595169544,0.08108039945363998,0.9845437594937156,0.057696766357863895,7.127771854400635 -EARS2,19,0.006881460547447205,0.06291855871677399,0.9911675065667666,-0.0314429688931701,5.461158275604248 -EBNA1BP2,8,0.022543704137206078,0.11481305956840515,0.9697885176441775,0.24820070851729556,9.884553909301758 -ECD,7,0.022329876199364662,0.11370596289634705,0.9713421274385509,0.08429716107864116,9.837564468383789 -ECT2,5,0.016809798777103424,0.0993758961558342,0.9793361706803881,0.0028653125949321145,8.535435676574707 -EDC4,16,0.009713895618915558,0.0752020999789238,0.9872968982591113,-0.07313183584244304,6.488453388214111 -EEF1A1,3,0.06966911256313324,0.20054428279399872,0.9169848462482852,0.025032189752548693,17.376590728759766 -EEF1B2,3,0.03479883447289467,0.1448892205953598,0.9581861064291606,0.008819389257455998,12.280804634094238 -EEF1G,17,0.012467455118894577,0.07907004654407501,0.9838557333038136,0.12942356235518301,7.350779056549072 -EEF2,32,0.00629022903740406,0.0586649589240551,0.9916159618434929,0.12435018848015861,5.221288681030273 -EFTUD2,12,0.02016485668718815,0.10676278173923492,0.9748329103358411,0.18240341169107763,9.348502159118652 -EGLN2,36,0.0037784858141094446,0.04718868434429169,0.9950046738309571,0.04547174784397941,4.046721458435059 -EIF1,9,0.013638721778988838,0.08732470124959946,0.9830092011789725,-0.03483794918993245,7.68831729888916 -EIF1AD,4,0.027685800567269325,0.12974829971790314,0.9647744699028141,-0.04471385403368553,10.95400619506836 -EIF1AX,14,0.0127993393689394,0.08452139794826508,0.9837246298295309,0.25948421694645174,7.447975158691406 -EIF2B1,18,0.010344521142542362,0.07585545629262924,0.9865770863529161,0.16979506193149854,6.695756435394287 -EIF2B2,32,0.015393438749015331,0.08552256971597672,0.9797157100001659,0.39702271548854856,8.167935371398926 -EIF2B3,11,0.019825691357254982,0.10359024256467819,0.9743015964116798,0.28851899939161413,9.269549369812012 -EIF2B4,33,0.01697857864201069,0.09018994122743607,0.9781205178806052,0.44239748918793437,8.578178405761719 -EIF2B5,46,0.014271625317633152,0.081490159034729,0.981607552745144,0.41732928489855603,7.864682197570801 -EIF2S1,25,0.02634100802242756,0.1097802072763443,0.9639218550883301,0.46565687326379646,10.68465805053711 -EIF2S2,12,0.013453825376927853,0.08683037012815475,0.9818743050409682,0.21073958590775507,7.6360249519348145 -EIF2S3,5,0.07546675950288773,0.20626074075698853,0.9022522695487348,0.3394049797345397,18.08515739440918 -EIF3A,38,0.010118911042809486,0.07098019868135452,0.9866168774616395,0.38328308738685524,6.622337818145752 -EIF3B,4,0.033242326229810715,0.14101292192935944,0.9577851385827086,0.14124511405922338,12.003010749816895 -EIF3CL,5,0.03215252608060837,0.1360379308462143,0.9602323513742239,0.043842042904151814,11.804619789123535 -EIF3D,18,0.015033603645861149,0.08867421746253967,0.9811010537034759,0.1868868795766588,8.071904182434082 -EIF3E,23,0.010838041082024574,0.07527920603752136,0.9860609955098568,0.1861978051662533,6.853617191314697 -EIF3F,6,0.023015854880213737,0.11521265655755997,0.9693807667408721,0.17201174880234962,9.987527847290039 -EIF3G,6,0.031863387674093246,0.13371127843856812,0.9585779500505598,0.2512919549156446,11.751422882080078 -EIF3H,37,0.007666163146495819,0.06262822449207306,0.9903400772305393,0.17259202578410324,5.764125823974609 -EIF3I,6,0.023953886702656746,0.118714340031147,0.9698056907191924,0.02562178096709792,10.189021110534668 -EIF3J,8,0.019233200699090958,0.10624858736991882,0.9764038052227297,-0.01515829467454034,9.129988670349121 -EIF3L,5,0.024529730901122093,0.11871698498725891,0.9691165523967342,-0.06100412194528513,10.31076431274414 -EIF3M,24,0.009489445015788078,0.06928686052560806,0.9875530389022138,0.1798321845051745,6.413053512573242 -EIF4A1,4,0.030942494049668312,0.13593249022960663,0.9610728110580972,-0.018456715958573656,11.580361366271973 -EIF4A3,1,0.12801317870616913,0.2853442132472992,0.8630304237120904,-0.05632523298562876,23.554386138916016 -EIF4B,4,0.032265763729810715,0.1387154757976532,0.9584047273948697,0.028071747467834623,11.82538890838623 -EIF4E,35,0.009345050901174545,0.0688217431306839,0.9876021908551663,0.17674303918037482,6.36407470703125 -EIF4G1,37,0.007781848777085543,0.06280698627233505,0.989657197060125,0.25580674670198095,5.807455062866211 -EIF4G2,34,0.004874296020716429,0.05270456522703171,0.9935378994099655,0.10373424794411655,4.596215724945068 -EIF5,17,0.010643210262060165,0.07808531075716019,0.9856952361830449,0.16894552652570574,6.791735649108887 -EIF5A,14,0.012970239855349064,0.08447503298521042,0.9829948688736219,0.08015541572207184,7.497534275054932 -EIF6,13,0.012762482278048992,0.08516637235879898,0.9833965906189235,0.1543686233718323,7.437243938446045 -ELAC2,9,0.013024752959609032,0.08746381103992462,0.9833707287252552,0.07143261131592814,7.5132737159729 -ELL,9,0.0155172199010849,0.09604565054178238,0.9798307526793414,0.048077632673714746,8.200709342956543 -ELOVL1,17,0.008049276657402515,0.06865619122982025,0.9891828314903047,0.036919778053464615,5.906400203704834 -ELP2,19,0.007684860844165087,0.06663751602172852,0.9902246786798297,-0.12707316852393777,5.771151065826416 -ELP3,12,0.012730101123452187,0.08657747507095337,0.9837322412710463,-0.026378855753499086,7.427803039550781 -ELP5,17,0.008213064633309841,0.06965421140193939,0.9893068364815633,0.049331720631113594,5.966189861297607 -ELP6,9,0.013494342565536499,0.08919120579957962,0.983999200428735,-0.08251426379570416,7.647514343261719 -EMC1,9,0.016415083780884743,0.09887456148862839,0.9787502784996466,0.11108029203162757,8.434630393981934 -EMC3,19,0.007403852883726358,0.06619493663311005,0.9901501599789354,0.005385935057363977,5.664653301239014 -EMC4,5,0.025150831788778305,0.12283247709274292,0.9678923640352824,-0.07459913697925567,10.440484046936035 -EMC7,20,0.0062392656691372395,0.06022903695702553,0.9918355769367911,0.06941401679593588,5.200093746185303 -ENO1,7,0.02138979360461235,0.10914967209100723,0.9713972850989925,0.0036222989314481703,9.628259658813477 -EP400,8,0.029049554839730263,0.12355729192495346,0.9602379936085005,0.18256020025638225,11.220550537109375 -EPRS,8,0.02488003671169281,0.11831168085336685,0.9666703106911732,0.24001258337308012,10.384126663208008 -ERAL1,11,0.013902395963668823,0.09035036712884903,0.9818091207916221,-0.00834571325862615,7.762279987335205 -ERCC2,11,0.015392748638987541,0.09397058933973312,0.9791423951315095,-0.014546955201857406,8.16775131225586 -ERCC3,17,0.009500719606876373,0.07432767748832703,0.9872031522946194,-0.002053078611837336,6.4168620109558105 -ERH,8,0.014178973622620106,0.09112124890089035,0.9813459066964478,0.105210917030992,7.839111328125 -ERVW-1,17,0.00866550300270319,0.07116248458623886,0.9889997908622471,-0.04400315029131561,6.128318786621094 -ESF1,15,0.011056050658226013,0.08033255487680435,0.9851707726546833,0.0872358891206389,6.922204971313477 -ESPL1,12,0.012363875284790993,0.084475077688694,0.9834971412831193,0.13660643559477792,7.3201799392700195 -ESPN,42,0.0033150645904242992,0.044051360338926315,0.99567285659894,-0.0940242242409264,3.790447235107422 -ETF1,5,0.04800495505332947,0.16304458677768707,0.9379860001694162,0.3370198802521456,14.42405891418457 -EWSR1,5,0.023341577500104904,0.11718985438346863,0.9707045479951142,0.023421366286608115,10.057952880859375 -EXOC1,18,0.007240399718284607,0.06499645859003067,0.9903580030441814,0.01407681759728157,5.601775646209717 -EXOC2,9,0.014082773588597775,0.09197304397821426,0.9820061646263863,0.011742934727037767,7.812473297119141 -EXOC3,24,0.006211746018379927,0.05981822684407234,0.9917945439187527,-0.0913388722238277,5.188613414764404 -EXOC4,11,0.012325088493525982,0.08522645384073257,0.9842745458954316,-0.005549750888261422,7.308689117431641 -EXOC5,11,0.016119815409183502,0.09690912812948227,0.9782540191339801,0.2336105193548591,8.358426094055176 -EXOC7,3,0.08750444650650024,0.23112083971500397,0.888165434247216,0.1494031501847357,19.47419548034668 -EXOC8,2,0.06428144127130508,0.19831308722496033,0.9244145628718374,-0.09283568831306889,16.69118881225586 -EXOSC1,7,0.01968389004468918,0.10658977180719376,0.9738312396239155,0.06845156932190438,9.23634147644043 -EXOSC10,9,0.01635938696563244,0.09785321354866028,0.9778883403387818,0.10958263754269247,8.420308113098145 -EXOSC2,22,0.02225683629512787,0.09857243299484253,0.9696730962618993,0.15204819237040743,9.821462631225586 -EXOSC3,6,0.044639572501182556,0.15665370225906372,0.9387404971940375,0.13040773673841088,13.909274101257324 -EXOSC4,10,0.029928885400295258,0.12534625828266144,0.9590276389550695,0.18584653265732717,11.38910961151123 -EXOSC5,4,0.049604631960392,0.16704218089580536,0.9346073178044813,0.1450679980875225,14.6624174118042 -EXOSC6,7,0.02465786598622799,0.12070959806442261,0.9664759379431997,0.19302292295508736,10.337657928466797 -EXOSC7,12,0.012737029232084751,0.08540670573711395,0.9830363131252113,0.010600816711437143,7.429823875427246 -EXOSC8,12,0.022725438699126244,0.10596051812171936,0.9691818509906379,0.20255903884349813,9.924315452575684 -EXOSC9,4,0.06153273209929466,0.18827560544013977,0.9170132094696327,0.1536399516214338,16.330427169799805 -F8A1,3,0.04596545174717903,0.1657882183790207,0.9414842703570384,0.0728630462721237,14.114327430725098 -FAF2,17,0.007782995700836182,0.06666410714387894,0.9899742334559768,-0.023417388581403135,5.807882785797119 -FAM133B,3,0.04307336360216141,0.16334103047847748,0.9479989928111172,-0.035042158588093494,13.663087844848633 -FAM136A,5,0.04650473967194557,0.16420365869998932,0.9370572174577959,0.24505537972417227,14.196885108947754 -FAM207A,11,0.013607299886643887,0.08966074138879776,0.9820385590234814,0.061383116804682454,7.6794562339782715 -FAM229A,13,0.009849969297647476,0.07667171955108643,0.9875477673144776,-0.06134777091969851,6.533740997314453 -FAM32A,39,0.005359407048672438,0.05472704395651817,0.9928708185417305,0.2477259662293242,4.819509506225586 -FAM50A,4,0.06002464145421982,0.18550674617290497,0.9222556866444823,0.18455824085314876,16.129066467285156 -FAM72D,12,0.01468546874821186,0.09289630502462387,0.9803098211931108,0.025731748731751526,7.977895259857178 -FARS2,4,0.044472306966781616,0.16128724813461304,0.9422179924544567,0.2141270600997479,13.883190155029297 -FARSA,14,0.018366819247603416,0.09988768398761749,0.9766981678433467,0.2737802642350502,8.921982765197754 -FARSB,9,0.02238970622420311,0.11226300150156021,0.9708477182395614,0.1919416053811888,9.850735664367676 -FASTKD5,18,0.007831822149455547,0.06721911579370499,0.9896728658424663,0.07320913079971046,5.826072692871094 -FAU,5,0.036079440265893936,0.1446021944284439,0.9521349241211676,0.18475392806838495,12.504730224609375 -FBL,13,0.02623181976377964,0.11702609807252884,0.9654225322645292,0.34411913370579206,10.662490844726562 -FBLIM1,14,0.0102440370246768,0.0778210237622261,0.9865776560683238,-0.0014748215858019517,6.663156986236572 -FBXO5,32,0.004399776924401522,0.049916304647922516,0.9942737028352919,-0.05436153482746158,4.366764545440674 -FCF1,23,0.01554179098457098,0.08955184370279312,0.980718322355833,0.20431312075237648,8.207200050354004 -FDPS,1,0.11932840943336487,0.2759461998939514,0.880861745050358,0.016349442555800744,22.741355895996094 -FDXR,14,0.010231325402855873,0.07729636132717133,0.9863243261427581,-0.07951501955852545,6.659021377563477 -FEN1,14,0.010229451581835747,0.07728226482868195,0.9868554787324412,-0.0662364113768675,6.658411502838135 -FGFR1OP,12,0.012041905894875526,0.08368679136037827,0.984237609504785,-0.0407541702352773,7.224238395690918 -FIP1L1,2,0.09762852638959885,0.24336324632167816,0.8874965461763739,0.13507912445994968,20.569931030273438 -FNBP4,16,0.007977515459060669,0.06812021881341934,0.9900753536119966,0.05596636692027305,5.880012512207031 -FNTA,8,0.01795431785285473,0.10274995863437653,0.9761702105377971,0.02874090772465675,8.821224212646484 -FNTB,10,0.012211919762194157,0.08514615893363953,0.9842621764836545,-0.07816244830275865,7.275057315826416 -FOLR3,31,0.004834738094359636,0.053253959864377975,0.9934940489227282,0.13176261091276292,4.577527046203613 -FOXD4,10,0.013276541605591774,0.08968450129032135,0.9826521361412566,0.011760238137915645,7.58554744720459 -FOXL2,17,0.007946091704070568,0.06789230555295944,0.9894057471320701,-0.005556395887806933,5.868420600891113 -FRG2,5,0.02496315725147724,0.12230104953050613,0.96859499472841,-0.0021724089340774714,10.401457786560059 -FTSJ3,20,0.010809785686433315,0.07674822956323624,0.9854679559669864,0.2102230342834053,6.844677448272705 -FXN,18,0.007914227433502674,0.06776591390371323,0.9894690186095815,0.149637213744016,5.856642246246338 -GAB2,68,0.012231101281940937,0.06614185869693756,0.9839980404302777,0.2180014990390632,7.280768871307373 -GABPA,11,0.0131558608263731,0.08786193281412125,0.983098125990513,-0.035707414031463706,7.5509934425354 -GAK,16,0.008861513808369637,0.0721876323223114,0.988256824906614,0.0836524789275686,6.19724178314209 -GAPDH,19,0.008935045450925827,0.06986498832702637,0.9883371686600095,0.1377949232438292,6.222900390625 -GAR1,14,0.012410729192197323,0.08431416004896164,0.9834542314264947,0.06711428247203768,7.334036827087402 -GARS,25,0.005951817613095045,0.05817104130983353,0.9922087219648241,0.09375425395073601,5.078895092010498 -GATA1,11,0.0904206782579422,0.2126055657863617,0.8811922313909805,0.283742391114751,19.796039581298828 -GBF1,16,0.023580867797136307,0.10928618907928467,0.969815948847589,0.2612006406160609,10.109375 -GCOM1,6,0.02001575566828251,0.10903716087341309,0.9741574068685818,-0.007849827231937199,9.313876152038574 -GEMIN4,8,0.01984928362071514,0.10853873938322067,0.9741001361620406,0.15470373223674827,9.275063514709473 -GEMIN5,14,0.012713736854493618,0.08426526933908463,0.9837439564000495,0.09290480096471777,7.423027515411377 -GEMIN6,9,0.014673558063805103,0.09231431782245636,0.9810969409911342,0.0010905470381539488,7.974659442901611 -GEMIN7,13,0.009367447346448898,0.0744447410106659,0.987650542410183,0.03889577512737818,6.371696472167969 -GEMIN8,8,0.014540924690663815,0.09330778568983078,0.9813047886257306,-0.06585217246130753,7.938536643981934 -GET1,8,0.016353953629732132,0.09798504412174225,0.9789903914030013,-0.0018569186062086089,8.418910026550293 -GET3,12,0.010828125290572643,0.0792870745062828,0.9858184935057297,-0.003042738361600754,6.850481033325195 -GFER,22,0.0071947029791772366,0.06420841813087463,0.9903770582588776,0.1208210138717596,5.584070682525635 -GFM1,24,0.005709301680326462,0.05714285001158714,0.9925564654754734,0.06998204112815186,4.9743452072143555 -GGPS1,16,0.007826379500329494,0.06767324358224869,0.9898219684876647,-0.04844078537507381,5.824047565460205 -GINS1,18,0.019555814564228058,0.09726336598396301,0.9733984039717322,0.27001821341300053,9.206241607666016 -GINS2,15,0.010797165334224701,0.07815457135438919,0.9857560330413735,0.10230066198691115,6.8406805992126465 -GINS3,10,0.01404870580881834,0.09024887531995773,0.9811611659823908,0.12976371163428269,7.803018569946289 -GINS4,6,0.02355390600860119,0.11809807270765305,0.9681632317737089,0.20211091197167066,10.103594779968262 -GIT2,10,0.013098304159939289,0.08793041110038757,0.9823777350504034,0.16347161262643287,7.534457683563232 -GJA3,8,0.015387735329568386,0.09608637541532516,0.9800164395626597,-0.055899677187424744,8.166421890258789 -GLE1,5,0.03396742418408394,0.13837656378746033,0.9554998292004881,0.13282082053468738,12.133210182189941 -GLI4,15,0.00935332290828228,0.0733010470867157,0.9878560404291566,-0.04573358320601053,6.366891384124756 -GLMN,8,0.017029784619808197,0.09968684613704681,0.9781342285727194,0.0003026631243833219,8.591105461120605 -GLRX3,13,0.010014982894062996,0.0764085203409195,0.986985743883294,0.003707417576272492,6.588242053985596 -GLRX5,10,0.013167556375265121,0.08774840831756592,0.9831527047863541,-0.046175245603344614,7.554348945617676 -GMPPB,17,0.008108863607048988,0.06744690984487534,0.9892486779231849,0.011538841679787288,5.928221702575684 -GMPS,13,0.009965392760932446,0.07575234025716782,0.9867108846662523,0.17940637344806729,6.571910858154297 -GNB1L,7,0.018139487132430077,0.10372206568717957,0.9770466573181549,-0.01824343589312141,8.866597175598145 -GNL2,35,0.008665667846798897,0.06666714698076248,0.988429331738876,0.41695223129006276,6.1283769607543945 -GNL3,4,0.029614929109811783,0.13175800442695618,0.9634658617868729,0.08070353491036239,11.329215049743652 -GNL3L,7,0.024658696725964546,0.1196824461221695,0.9671677666672456,0.20608696043289265,10.3378324508667 -GNPNAT1,11,0.013666733168065548,0.0900605171918869,0.9819808399472774,-0.03103419216427046,7.6962080001831055 -GOLGA6L1,19,0.007368162274360657,0.06479325145483017,0.9903805601711487,-0.05433986443179345,5.6509833335876465 -GOLT1B,28,0.006165540311485529,0.05983773246407509,0.9917811972030216,0.15136287333253204,5.1692795753479 -GON4L,23,0.0058565582148730755,0.058998458087444305,0.9924350109484665,0.046590205204256024,5.038087368011475 -GOSR2,9,0.013934561051428318,0.09073050320148468,0.9825517053551321,-0.08063527797475152,7.77125358581543 -GPKOW,10,0.014245525002479553,0.09148464351892471,0.982111858664138,0.002608586464591343,7.857487201690674 -GPN1,8,0.02935490384697914,0.12997137010097504,0.9616800716588303,0.15218470576189808,11.27936840057373 -GPN2,5,0.03499991074204445,0.1424618512392044,0.9522051878795409,0.10750775383250187,12.316232681274414 -GPN3,27,0.018453449010849,0.09629034996032715,0.9761676755347476,0.221884468831204,8.943000793457031 -GPR61,11,0.010758375748991966,0.07953746616840363,0.986558634641195,-0.07876192024710817,6.8283820152282715 -GPS1,43,0.005033369641751051,0.05257469415664673,0.9934982379307851,0.17876120661777664,4.670612335205078 -GRB2,17,0.010213701985776424,0.07582029700279236,0.9863783254935452,-0.048664479189259584,6.653283596038818 -GRPEL1,13,0.019815275445580482,0.102175273001194,0.9735643697566648,0.2725688177616864,9.26711368560791 -GRSF1,17,0.010072755627334118,0.07643835991621017,0.9870334252837475,0.19195806384569128,6.607217311859131 -GRWD1,19,0.007781164720654488,0.06691064685583115,0.9896960384326954,0.009688814781779841,5.807199954986572 -GSDMA,11,0.011183220893144608,0.08102725446224213,0.9853089966124909,0.08661844447862793,6.961901664733887 -GSPT1,11,0.01996464654803276,0.10295504331588745,0.9739438079910805,0.31472064865866933,9.301977157592773 -GTF2A1,12,0.01673455350100994,0.09694389253854752,0.9772818680962999,0.10931451576696995,8.516311645507812 -GTF2A2,18,0.011015965603291988,0.0766151174902916,0.9850719022049407,0.12603780376505871,6.909644603729248 -GTF2B,3,0.08389059454202652,0.2191355973482132,0.8945684598688457,0.13754892780496808,19.067821502685547 -GTF2E1,18,0.0101516367867589,0.07594744116067886,0.9867108732177439,0.2055358559672415,6.63303804397583 -GTF2E2,7,0.021711235865950584,0.11295264959335327,0.971621473182975,0.058058157653972806,9.700334548950195 -GTF2F1,4,0.03666216507554054,0.14726372063159943,0.952703364136934,-0.014049151791578524,12.605308532714844 -GTF2F2,6,0.02748832106590271,0.12550850212574005,0.9639091530986891,0.08171822395299597,10.91486930847168 -GTF2H1,44,0.01118407491594553,0.07664618641138077,0.9858091463756009,0.2939077239500877,6.962167739868164 -GTF2H2,15,0.010284459218382835,0.07726525515317917,0.9863677334174289,-0.0476726431312875,6.676290035247803 -GTF2H2C,12,0.014897650107741356,0.09273700416088104,0.9813150867631157,0.1035619514456173,8.035323143005371 -GTF2H3,7,0.02086830511689186,0.11143652349710464,0.9726942239090455,0.040258411758331496,9.510165214538574 -GTF2H4,11,0.015061783604323864,0.09198430180549622,0.9798285962765128,-0.026425255422039,8.079465866088867 -GTF3A,9,0.016411930322647095,0.09735716134309769,0.9780382505979474,0.12504811576638633,8.433818817138672 -GTF3C1,16,0.008478252217173576,0.07048498094081879,0.9888582243570244,-0.015393007128764713,6.061744213104248 -GTF3C2,12,0.012369388714432716,0.08432114124298096,0.9833711932520505,0.08340329268568611,7.321811676025391 -GTF3C3,13,0.011608650907874107,0.08169706910848618,0.9848996275423133,-0.027724299069120238,7.093087673187256 -GTF3C4,13,0.01006248313933611,0.07638252526521683,0.9870963251289171,-0.08480557724212864,6.603847980499268 -GTF3C5,16,0.008237605914473534,0.06975969672203064,0.9896309344629827,-0.0867554048477982,5.975096702575684 -GTF3C6,25,0.005332718603312969,0.05484490841627121,0.993194093236784,-0.03934661721141635,4.807494640350342 -GTPBP4,7,0.02004941925406456,0.10762833058834076,0.9755850258933966,0.17886301674535798,9.321704864501953 -GUCD1,28,0.007083875592797995,0.06303338706493378,0.9905652950232889,0.16973362083569157,5.540894985198975 -GUK1,12,0.01367664523422718,0.089110367000103,0.9831709055936009,0.11224361222082717,7.698998928070068 -H2AFX,14,0.009604470804333687,0.07560240477323532,0.9873596608362905,0.07533096265277901,6.451804161071777 -H2AFZ,10,0.014545978046953678,0.0895589217543602,0.9818876917938656,-0.16937584755699275,7.939916133880615 -H3F3A,23,0.006306863389909267,0.05966554954648018,0.9916831613620505,-0.07767941847747412,5.2281880378723145 -HAMP,4,0.04197177290916443,0.15884272754192352,0.9461736077828655,0.08706276116234993,13.4872407913208 -HAPLN2,12,0.011505959555506706,0.08220551162958145,0.9849673624676264,-0.0958055786343581,7.061645030975342 -HARS,29,0.02192104235291481,0.10180702060461044,0.9719128284906116,0.37763995138581086,9.747091293334961 -HARS2,6,0.02137119509279728,0.11313595622777939,0.9735776150501522,0.006848761842840462,9.624072074890137 -HAUS1,6,0.022101804614067078,0.11359624564647675,0.9715259828499264,0.04239256494715869,9.787196159362793 -HAUS3,17,0.008014350198209286,0.06840725988149643,0.9894015822410854,-0.015075820385161793,5.893571376800537 -HAUS4,19,0.00833671074360609,0.06915275752544403,0.9889128288730549,0.1746549146377123,6.010931968688965 -HAUS5,16,0.008309578523039818,0.07082412391901016,0.9890585310612017,0.0779832195476814,6.001142501831055 -HAUS6,4,0.03454381600022316,0.14241406321525574,0.9546266900629975,0.16069142935792147,12.235722541809082 -HAUS7,24,0.0054124691523611546,0.056762754917144775,0.9927806274548172,0.08955304097332001,4.843308925628662 -HAUS8,12,0.013638600707054138,0.08901753276586533,0.9819090587490437,0.04580783835767634,7.688283443450928 -HBS1L,15,0.022584697231650352,0.10955343395471573,0.970329542795887,0.39512809549777245,9.893537521362305 -HCFC1,5,0.03437451645731926,0.14245864748954773,0.9584472604126991,0.08979316986250395,12.20570182800293 -HCRTR1,11,0.010877164080739021,0.08008723706007004,0.985844927070006,0.0011757277198687487,6.865976333618164 -HDAC3,6,0.019725900143384933,0.10758504271507263,0.9752994828486662,0.13652251144531327,9.246191024780273 -HDAC7,17,0.009581510908901691,0.07298458367586136,0.9870649520463587,0.10867246702293941,6.444087982177734 -HEATR1,24,0.016744129359722137,0.09229305386543274,0.9783100805451737,0.35852340021096113,8.518747329711914 -HECTD1,13,0.00950800720602274,0.07345011830329895,0.9878923543307461,-0.0495197231853943,6.419322967529297 -HGS,10,0.0167896319180727,0.09833835065364838,0.9777176462840241,0.2613990152716636,8.530314445495605 -HINFP,16,0.013957482762634754,0.08617621660232544,0.9812262890379974,0.23405254534782946,7.777643203735352 -HIRA,34,0.004108595196157694,0.04923649877309799,0.9945242524480971,-0.0052055325919327005,4.21979284286499 -HIST1H2AB,9,0.014736823737621307,0.09336254745721817,0.9809212737656208,-0.0030743438448901385,7.991832733154297 -HIST1H2AE,13,0.01016688346862793,0.07780064642429352,0.986597437182269,0.024694903421573537,6.638017177581787 -HIST1H2AI,5,0.027491658926010132,0.12700261175632477,0.9645672514623503,-0.05948844035722032,10.915532112121582 -HIST1H2BB,13,0.010242314077913761,0.07686243206262589,0.9866674886090993,-0.040760830404782046,6.662596702575684 -HIST1H2BC,7,0.01896129921078682,0.10512781143188477,0.9755982410612901,-0.018763044675483807,9.06522274017334 -HIST1H2BE,5,0.025072095915675163,0.12215637415647507,0.967787323306942,0.06384228534154972,10.424128532409668 -HIST1H2BJ,5,0.02928009256720543,0.1321663111448288,0.963437980135208,0.06157532898016382,11.264986038208008 -HIST1H2BL,21,0.006391423754394054,0.06111186742782593,0.9917685760348556,-0.021844758168611946,5.263119697570801 -HIST1H2BM,15,0.008779243566095829,0.0723918229341507,0.9885765385263653,0.005304310279627319,6.168406963348389 -HIST1H2BN,10,0.01520796213299036,0.09444177895784378,0.9794658216277912,0.12825346291998982,8.118578910827637 -HIST2H2AA3,6,0.019985569640994072,0.10925599187612534,0.974312989207603,-0.045522302119915446,9.30685043334961 -HIST2H2AC,8,0.019000425934791565,0.10671301186084747,0.9754119462801738,0.06771279149656853,9.07457160949707 -HIST2H2BE,3,0.0410136803984642,0.15557725727558136,0.9497143499354873,-0.04897426068094547,13.332415580749512 -HIST2H2BF,12,0.010945050045847893,0.07966704666614532,0.9858382812915858,-0.06505665142656417,6.887369155883789 -HIST2H3A,8,0.01673835515975952,0.09925279766321182,0.9783202118343801,0.04540705558274863,8.517278671264648 -HIST2H3D,19,0.007299675606191158,0.06552384793758392,0.9905591595749282,-0.048430622733296645,5.624659538269043 -HJURP,11,0.011522580869495869,0.08303771913051605,0.9848548801467586,0.11728878597494026,7.066743850708008 -HLA-C,5,0.025880241766572,0.12285546213388443,0.9655070958737326,0.10628107630537344,10.59079647064209 -HMGA1,12,0.012918779626488686,0.08609170466661453,0.9830959899384576,0.08953472306494978,7.4826459884643555 -HMGB1,24,0.005595288705080748,0.05714430660009384,0.992565265046632,0.05067950289138315,4.924427032470703 -HMGCR,16,0.0078824358060956,0.06800900399684906,0.9896309432422902,0.062159865663367134,5.84486722946167 -HMGCS1,23,0.007294586859643459,0.0634557381272316,0.9902327813551353,0.12300007309217424,5.6226983070373535 -HMGN2,10,0.01361836213618517,0.08906299620866776,0.9820322250978991,0.0017725592835499688,7.6825761795043945 -HNRNPC,34,0.012791091576218605,0.0787065401673317,0.9828672637421485,0.35358168854800553,7.445575714111328 -HNRNPH1,16,0.010443021543323994,0.07758509367704391,0.9865128930099843,0.11899930224808708,6.727559566497803 -HNRNPK,3,0.03886623680591583,0.1530599594116211,0.9529971797209945,0.11097026484360312,12.97868537902832 -HNRNPL,8,0.016230693086981773,0.09737105667591095,0.9793785122116417,0.08219560299680666,8.38712215423584 -HNRNPM,26,0.005582396872341633,0.056710053235292435,0.9925240761274193,0.17642394057976088,4.918750762939453 -HNRNPR,15,0.009189322590827942,0.07281086593866348,0.987814937432909,0.025988948090044834,6.310825347900391 -HNRNPU,10,0.015347553417086601,0.09239180386066437,0.9802810924764843,0.07069229487658146,8.155752182006836 -HOXC10,9,0.014671703800559044,0.09226162731647491,0.9808986265816688,-0.05879204679039581,7.974155902862549 -HS6ST1,17,0.007608854677528143,0.06662768125534058,0.9901344003813105,-0.07935955883524694,5.7425408363342285 -HSCB,8,0.017251521348953247,0.10136803239583969,0.9778412218935884,0.08035376396269214,8.646854400634766 -HSD17B10,2,0.0659058466553688,0.19936466217041016,0.9198588527981073,0.0162357806619698,16.900766372680664 -HSD17B12,14,0.013507251627743244,0.08805171400308609,0.9817636846882258,0.23037387420213515,7.6511712074279785 -HSF1,13,0.01126245316118002,0.08016178011894226,0.9853090818587196,7.37723687096241e-05,6.986520767211914 -HSP90B1,12,0.011298436671495438,0.08067850768566132,0.9855217906767517,0.008384678657427737,6.997673034667969 -HSPA14,12,0.010707813315093517,0.07962604612112045,0.9864627443840894,-0.055397629619518454,6.81231689453125 -HSPA5,44,0.022581640630960464,0.09754900634288788,0.9709123124822502,0.3625385323751908,9.892867088317871 -HSPA8,25,0.01563967391848564,0.0863928273320198,0.9808220639257635,0.2550739709751614,8.233003616333008 -HSPA9,42,0.024492312222719193,0.10392890125513077,0.9678041110541896,0.4492647228121781,10.302896499633789 -HSPD1,15,0.010409240610897541,0.07636062055826187,0.9865054708808677,0.0029656967511841855,6.71666955947876 -HSPE1,3,0.05386529490351677,0.17673379182815552,0.9357402493251155,0.16487459453840203,15.279142379760742 -HTATSF1,9,0.014281395822763443,0.09182242304086685,0.9810656071715622,0.13919372290075746,7.867373466491699 -HUS1,20,0.00686808954924345,0.06265757232904434,0.9913131477717582,-0.10351287642182216,5.455850124359131 -HUWE1,10,0.015677358955144882,0.09622897952795029,0.979785077207128,0.009906694575114726,8.242916107177734 -HYOU1,3,0.05014905706048012,0.16668853163719177,0.9354623106164803,0.12026838854874888,14.742659568786621 -HYPK,12,0.011341598816215992,0.08212674409151077,0.9851399143531571,0.047580984800340544,7.011026382446289 -IARS,6,0.027549663558602333,0.12700307369232178,0.9645163403595317,0.25229322310150565,10.927042007446289 -IARS2,21,0.011950628831982613,0.07726960629224777,0.9842919391097437,0.2506406283133335,7.196806907653809 -ICE1,43,0.005945398472249508,0.056442756205797195,0.9920687007388853,0.21477938435854307,5.076155662536621 -IFITM2,15,0.00956062600016594,0.07444893568754196,0.9874563893269839,-0.010910394469883612,6.437060356140137 -IFITM3,2,0.07752563059329987,0.21985489130020142,0.910316425954893,0.039442863013100946,18.3301944732666 -IGBP1,28,0.008321699686348438,0.06826864928007126,0.9891801479986165,0.24619206670352667,6.005517959594727 -IK,29,0.005610223393887281,0.05685114115476608,0.9924478876920435,0.08875883751571906,4.930994510650635 -ILF2,5,0.025243934243917465,0.12124378979206085,0.9700862359520545,-0.06745047000871847,10.459790229797363 -ILF3,14,0.011844971217215061,0.08213011175394058,0.9845779887534629,0.09227263324149981,7.164921760559082 -IMP3,2,0.05707945674657822,0.18683801591396332,0.9289517990930134,0.023693622919189045,15.728394508361816 -IMP4,17,0.020737793296575546,0.10445356369018555,0.9738973102589064,0.24272207266905269,9.480380058288574 -IMPDH2,7,0.022740647196769714,0.11439060419797897,0.969627094941069,0.15528735710188907,9.92763614654541 -INCENP,6,0.02105768769979477,0.11112885922193527,0.9723726287265738,0.01182068948309854,9.553220748901367 -ING3,7,0.020132269710302353,0.10911303013563156,0.9741102794927049,-0.03643283243097805,9.34094524383545 -INO80,15,0.013098984025418758,0.08463943749666214,0.9823634485623008,0.1012609133255048,7.5346527099609375 -INO80B,11,0.01528406422585249,0.0936274379491806,0.9803585786375465,0.06804163131568121,8.13886547088623 -INTS1,21,0.007318185176700354,0.06527884304523468,0.99073762208226,-0.08868769667307107,5.631786346435547 -INTS2,36,0.03484145924448967,0.1264888495206833,0.952719613193873,0.24183241587767157,12.288323402404785 -INTS3,10,0.015390745364129543,0.09530526399612427,0.9800351850766612,0.12586568319349947,8.167220115661621 -INTS4,5,0.04118041694164276,0.15346212685108185,0.9447619741800152,0.17512519387604375,13.359488487243652 -INTS5,7,0.03568737953901291,0.13811782002449036,0.9538574642149114,0.14375929008372376,12.436603546142578 -INTS6,14,0.01184756588190794,0.08297055959701538,0.9840523179133941,-0.011636624317490714,7.165706634521484 -INTS7,8,0.02564321458339691,0.11963900178670883,0.9659410459459771,0.04519688240729953,10.542186737060547 -INTS8,33,0.02278437837958336,0.10410420596599579,0.969222822640467,0.21519391813571423,9.937177658081055 -INTS9,23,0.019164446741342545,0.09829414635896683,0.9748075025705738,0.24355249691800984,9.113655090332031 -IPO11,7,0.01862594485282898,0.10514136403799057,0.9751407783327396,0.00363302248857522,8.984700202941895 -IPO13,5,0.03486451879143715,0.14022961258888245,0.955445274775408,0.06558962281336234,12.292388916015625 -IPO7,7,0.020031709223985672,0.10864546149969101,0.9747176021710048,0.04755380578736189,9.317587852478027 -IPO9,18,0.008583417162299156,0.06983932852745056,0.9884749972885004,-0.02352927510136244,6.099223613739014 -ISCA2,23,0.005669299513101578,0.05647600069642067,0.9926116299249872,-0.022531289187464016,4.956888675689697 -ISCU,6,0.02047252468764782,0.10978082567453384,0.9730792803873713,0.12132713705959076,9.419549942016602 -ISG20L2,8,0.017345553264021873,0.10067392140626907,0.9775782006516968,0.20137278721194812,8.670388221740723 -ISY1,3,0.04210938513278961,0.15835388004779816,0.9475364809293751,-0.006620551288350979,13.509332656860352 -JAZF1,21,0.007367369253188372,0.06620539724826813,0.9906924397637259,0.05497016556548942,5.650679588317871 -JMJD6,20,0.0061067682690918446,0.06034564599394798,0.9920222431456522,0.11386065960604956,5.144583225250244 -KANSL1,5,0.041689835488796234,0.15732115507125854,0.9510052666327429,0.07915839475037684,13.441864967346191 -KANSL2,16,0.016801608726382256,0.09624006599187851,0.9790700262471003,0.16730031679213397,8.533355712890625 -KANSL3,18,0.011380521580576897,0.07967644184827805,0.9860315484562241,0.020021698464621148,7.023046493530273 -KARS,6,0.02408747375011444,0.1200595498085022,0.9690613862890947,0.12118382491458998,10.217392921447754 -KAT5,12,0.010527441278100014,0.07821717858314514,0.9858943699063314,0.13661437445021546,6.754696846008301 -KAT8,28,0.006850348319858313,0.06184990704059601,0.9917039429759641,-0.022771833202266667,5.448799133300781 -KATNB1,18,0.0072770738042891026,0.06453755497932434,0.9904552663297673,-0.0681754491030393,5.615944862365723 -KCMF1,16,0.008908289484679699,0.07198992371559143,0.9882016563049062,0.03530146595102452,6.213576316833496 -KCNA10,14,0.00966187845915556,0.07485198974609375,0.9871334418483347,0.020998011006934658,6.471056938171387 -KCTD10,7,0.02034977823495865,0.10797727853059769,0.9735160823684712,0.008492687111961994,9.39126968383789 -KDM2A,11,0.010929836891591549,0.07959876954555511,0.9857089601956845,0.05957693810173734,6.882580757141113 -KDM8,5,0.028906086459755898,0.13027778267860413,0.9629652858017733,0.1809474797675313,11.192809104919434 -KIAA1143,20,0.007213477045297623,0.0648755431175232,0.9905579995538539,0.2203424768773945,5.59135103225708 -KIF11,98,0.0016616054344922304,0.03095211647450924,0.9977805828880524,-0.014091507388758166,2.683542013168335 -KIF14,12,0.010088086128234863,0.07726892083883286,0.9867474846963371,0.009166252676264332,6.61224365234375 -KIF18A,4,0.030444428324699402,0.13507582247257233,0.9627879035046547,-0.06472153352578133,11.48678207397461 -KIF18B,12,0.010361596010625362,0.07866692543029785,0.9861289575238757,0.04105371693942281,6.701280117034912 -KIF20A,9,0.01837734878063202,0.10042450577020645,0.9758729838642791,0.15227175374310656,8.924540519714355 -KIF23,8,0.017314275726675987,0.10008928924798965,0.977350346708477,-0.025360348526621133,8.662567138671875 -KIF4A,13,0.009490055032074451,0.07475385069847107,0.9878391594814244,0.020606097471459247,6.413259983062744 -KIN,16,0.012353315018117428,0.08379728347063065,0.9840046337443039,0.23979549939630862,7.317053318023682 -KLC2,12,0.012874494306743145,0.08666779845952988,0.9827228734222134,0.07699553498584497,7.469809532165527 -KLF7,18,0.007222708780318499,0.06564127653837204,0.99042934524562,0.07513674774849285,5.59492826461792 -KLHL11,13,0.009750045835971832,0.07575540244579315,0.9873865441126091,0.033097936486464,6.500515460968018 -KPNA6,8,0.01942947320640087,0.10704514384269714,0.974496402263616,-0.024761758457773186,9.1764554977417 -KPNB1,36,0.006049176678061485,0.056899938732385635,0.9920159773672997,0.026757999630460296,5.120266437530518 -KRI1,5,0.03307123854756355,0.13867437839508057,0.9566458872957836,0.07882984096320966,11.97208309173584 -KRR1,45,0.009494820609688759,0.06800874322652817,0.987708668595985,0.18794915465569934,6.41486930847168 -KRT10,18,0.00780020747333765,0.06784022599458694,0.9900204187288819,-0.005225596877883865,5.814301490783691 -KRT17,9,0.01600557193160057,0.0970001295208931,0.9786629779866904,0.08481128779946054,8.328754425048828 -KRT8,34,0.004306837450712919,0.0493290051817894,0.9942533599256882,0.06819751021373159,4.32039737701416 -KRTAP4-2,13,0.010150839574635029,0.07769758254289627,0.9867264649577002,0.11959741350314672,6.632778167724609 -KRTAP4-7,17,0.008723482489585876,0.0709637776017189,0.9884872100298052,0.009247084725822313,6.148785591125488 -KTI12,5,0.025753561407327652,0.12269420921802521,0.9673422361175678,0.127218283776064,10.564845085144043 -KXD1,10,0.012722055427730083,0.08589573204517365,0.9834496619126954,0.00816187390936485,7.425455570220947 -LAMTOR2,6,0.03811514005064964,0.14802774786949158,0.9495603638830906,0.15187341093765333,12.852665901184082 -LAMTOR3,4,0.02708456479012966,0.12820684909820557,0.9671456027107196,-0.06476598182298189,10.834413528442383 -LAMTOR4,13,0.011434538289904594,0.08074909448623657,0.9848412158580386,0.050619942935874275,7.039694309234619 -LARS,10,0.024414274841547012,0.11387845873832703,0.9670805770830747,0.30777578664914973,10.286469459533691 -LARS2,15,0.008818784728646278,0.07253414392471313,0.9882801864571027,0.04374381028704095,6.182282447814941 -LAS1L,9,0.014553166925907135,0.09270267933607101,0.9810395203888663,0.0755613633058975,7.941877841949463 -LCE1C,15,0.010254730470478535,0.07713206112384796,0.9863536358187336,0.025656588581315733,6.666633605957031 -LCE1E,13,0.009111640974879265,0.07325202971696854,0.9884093197404143,0.0409875900902478,6.284095287322998 -LCN10,8,0.016296330839395523,0.09819534420967102,0.9797079266938584,0.024957820282802162,8.404064178466797 -LENG8,12,0.011317102238535881,0.08181317150592804,0.9852858759561248,0.0863437557247823,7.003450870513916 -LETM1,13,0.013003679923713207,0.08624936640262604,0.9831149281968462,0.18317051129746367,7.507193088531494 -LIMS1,28,0.005099808797240257,0.054984819144010544,0.9932823887512877,0.011585472243404527,4.7013373374938965 -LIN52,2,0.04798813909292221,0.1713898926973343,0.9439659791283838,0.08029328904600644,14.42153263092041 -LIN54,30,0.004487618338316679,0.05106987804174423,0.9939727238112767,0.09598351049546022,4.410140514373779 -LONP1,13,0.0115001006051898,0.08041827380657196,0.9850948729532132,0.1405084483490352,7.0598464012146 -LPA,12,0.010343759320676327,0.07774364948272705,0.9866147493754222,-0.043524100115727986,6.695509910583496 -LRPPRC,11,0.01723126322031021,0.0916885957121849,0.9789755229882178,0.1771103413606529,8.641776084899902 -LRR1,12,0.015506705269217491,0.09397115558385849,0.9795270212409887,0.012713676224973179,8.197930335998535 -LRRC37A3,12,0.010402821935713291,0.07874838262796402,0.986503604792239,-0.058289369691618895,6.714598655700684 -LRRC37B,11,0.012329376302659512,0.0846107006072998,0.9836373546854563,0.10251454087301526,7.309959888458252 -LSG1,11,0.013997151516377926,0.08859822899103165,0.9817091035611617,0.1618575193212284,7.788687229156494 -LSM10,13,0.012281249277293682,0.08101753890514374,0.9840693903134049,-0.012121290517540415,7.295679092407227 -LSM11,13,0.015425722114741802,0.08983837068080902,0.9791090520936074,0.05328684266363073,8.176495552062988 -LSM12,7,0.01872776262462139,0.10571908950805664,0.9761176387899573,-0.005683022077080083,9.009223937988281 -LSM2,35,0.01573745720088482,0.08874166756868362,0.9801133510535257,0.2836393226025929,8.258700370788574 -LSM3,8,0.027534591034054756,0.12328310310840607,0.965280465104132,0.20904050159091392,10.924052238464355 -LSM4,9,0.02193480357527733,0.11174137145280838,0.9734794164398193,0.16172045434604806,9.750150680541992 -LSM5,20,0.00813169777393341,0.06797144562005997,0.9893928452064086,0.07783999394211752,5.936563014984131 -LSM6,46,0.01436646282672882,0.08338075876235962,0.981174121655359,0.2999155626352304,7.8907694816589355 -LSM7,14,0.014013346284627914,0.08943448215723038,0.9818030640908086,0.13937538822185375,7.793191909790039 -LST1,15,0.008601195178925991,0.07113724201917648,0.9887374463463798,0.024600485123454644,6.105536937713623 -LTB4R2,19,0.00696781137958169,0.06398061662912369,0.9910106744402752,-0.0911949987249352,5.4953155517578125 -LTBP3,15,0.00856020301580429,0.07137347012758255,0.9888676017529155,-0.019515070101683546,6.090970516204834 -LTV1,32,0.009723749943077564,0.07128425687551498,0.987446902034368,0.24338766496629696,6.491743564605713 -LUC7L3,5,0.02460474893450737,0.11913285404443741,0.9687017954127143,0.06013453101582469,10.326519012451172 -LURAP1,7,0.017562439665198326,0.10250899195671082,0.9781045415185473,-0.011185928393871369,8.72442626953125 -LYRM4,7,0.018553514033555984,0.10513994842767715,0.9762666516946511,-0.005469918008786039,8.967214584350586 -MAD2L1,16,0.009448803029954433,0.07297088205814362,0.9874588854146779,0.15178345510687488,6.399306297302246 -MAD2L2,20,0.006593949161469936,0.06219777464866638,0.9912119026587526,0.06815333975405057,5.345856189727783 -MAGOH,24,0.009725949726998806,0.0709451213479042,0.9868789366197552,0.19384561580382975,6.4924774169921875 -MAK16,22,0.009143981151282787,0.0707167387008667,0.9879901555534903,0.27787277716522585,6.295237064361572 -MANF,14,0.009755276143550873,0.07551898807287216,0.9871947337781608,-0.043974542506472133,6.50225830078125 -MARS,9,0.019879966974258423,0.10630717873573303,0.9739488837375766,0.25569024675845076,9.28222942352295 -MARS2,7,0.016016671434044838,0.09798973053693771,0.979052401674707,0.07316409890803474,8.331642150878906 -MASTL,11,0.01634589582681656,0.09631064534187317,0.978582049908045,0.16579582033703016,8.41683578491211 -MAT2A,5,0.03474430367350578,0.14176665246486664,0.9559589557814286,0.081184839058966,12.27117919921875 -MAU2,6,0.021918993443250656,0.1141497939825058,0.9713667412228806,0.12214727720349863,9.746636390686035 -MAX,18,0.026539742946624756,0.11475000530481339,0.9676144014302185,0.2549454284519244,10.724889755249023 -MBTPS1,7,0.018885962665081024,0.10580754280090332,0.9758155818412259,0.06975851349107844,9.047196388244629 -MBTPS2,22,0.007562499027699232,0.06375440210103989,0.9900047636871484,0.08713535400241836,5.725021839141846 -MCL1,8,0.01508916076272726,0.09558447450399399,0.9808088561554134,-0.00033934999289684836,8.08680534362793 -MCM10,22,0.009454289451241493,0.07013442367315292,0.9871998897686568,0.14670745313530356,6.401163101196289 -MCM2,13,0.012583093717694283,0.08445112407207489,0.9830892006224095,0.1415890194558063,7.384790420532227 -MCM3,13,0.011027052067220211,0.07949911803007126,0.9853177703245775,0.055310965955621655,6.913121223449707 -MCM3AP,11,0.01467717532068491,0.0925373062491417,0.9801008457041172,0.08838840842737318,7.975642681121826 -MCM4,8,0.023396406322717667,0.11580684036016464,0.9682959435980101,0.12297229513654924,10.069757461547852 -MCM5,9,0.01530276145786047,0.09418007731437683,0.9801477639311504,0.1071353815345027,8.143842697143555 -MCM6,10,0.01316099800169468,0.08649343252182007,0.983220156329158,0.006308185729603284,7.552467346191406 -MCM7,16,0.007815191522240639,0.06779059022665024,0.9896104279551843,0.013652133800799155,5.819883346557617 -MCMBP,15,0.011740789748728275,0.08103740215301514,0.9846131909654544,0.08430234218865722,7.13334321975708 -MCRS1,7,0.015883062034845352,0.09775537997484207,0.9805694518060072,-0.010825027796014611,8.296817779541016 -MDN1,9,0.015247801318764687,0.09484154731035233,0.9805802623955137,0.1716156505159713,8.129204750061035 -MED1,26,0.030860165134072304,0.12589401006698608,0.9614304159815951,0.3752542543501173,11.564945220947266 -MED10,51,0.010531796142458916,0.07364987581968307,0.9864959783512467,0.31041409068170034,6.756094455718994 -MED11,23,0.018514711409807205,0.09942406415939331,0.9759716544055904,0.2961747157758694,8.957832336425781 -MED12,58,0.03538604825735092,0.12895424664020538,0.9549100042077825,0.43099605037673033,12.383986473083496 -MED14,11,0.0193039420992136,0.10485896468162537,0.9739927173468225,0.13704513225310877,9.146763801574707 -MED17,15,0.032536063343286514,0.13158351182937622,0.9569337153523775,0.30423791087120555,11.874817848205566 -MED18,2,0.09593687951564789,0.24235734343528748,0.8793347994593477,0.08019840381664134,20.390939712524414 -MED19,55,0.019832370802760124,0.09795065224170685,0.9762884629957701,0.39973971697129945,9.271110534667969 -MED20,8,0.03868570923805237,0.14543673396110535,0.9474146136536808,0.18053553685549814,12.948508262634277 -MED21,7,0.05129515007138252,0.16917692124843597,0.9318013892779122,0.31040997486679106,14.910170555114746 -MED22,29,0.016852587461471558,0.09139727801084518,0.9773818174803865,0.47487475634150567,8.546293258666992 -MED26,11,0.014288470149040222,0.09200169891119003,0.981076526598897,-0.01401395833530421,7.869321823120117 -MED27,2,0.1271718442440033,0.2786775231361389,0.845818634163054,0.13280476058152446,23.476856231689453 -MED28,8,0.038081761449575424,0.1466253250837326,0.9487624247302169,0.28308193771588763,12.847037315368652 -MED29,10,0.0253969244658947,0.11970802396535873,0.9664492174176352,0.20100433635938147,10.491437911987305 -MED30,29,0.03512084484100342,0.13220100104808807,0.9554113154376266,0.3800689520955573,12.337492942810059 -MED31,4,0.036531928926706314,0.14892321825027466,0.951544982050083,0.02103335474352863,12.582901000976562 -MED4,6,0.043237414211034775,0.15874412655830383,0.9424375696150608,0.22866815123864442,13.689081192016602 -MED6,24,0.023436974734067917,0.11094403266906738,0.9710482588766758,0.39423556097229684,10.078484535217285 -MED7,27,0.02929588407278061,0.12230638414621353,0.9635814977667385,0.3796029174087256,11.268024444580078 -MED8,11,0.019401537254452705,0.10618984699249268,0.9738039228590627,0.159411003608086,9.169856071472168 -MED9,12,0.062203895300626755,0.17884552478790283,0.9191039104580385,0.3841403275778425,16.419246673583984 -MEPCE,12,0.009552210569381714,0.07564520090818405,0.9876871018325609,-0.08159320291688274,6.434227466583252 -METAP1,10,0.012689989060163498,0.08689882606267929,0.983452993048536,-0.02743083285442176,7.416091442108154 -METAP2,16,0.012458172626793385,0.08173909783363342,0.9834455304059267,0.08244552581608264,7.348042011260986 -METTL1,21,0.006553774233907461,0.06127062067389488,0.9912351934741567,0.021520043566830266,5.329545974731445 -METTL14,12,0.014338626526296139,0.09157007187604904,0.9805880752190246,0.016797128010930088,7.883121967315674 -METTL16,4,0.044738803058862686,0.16419029235839844,0.9452983075544321,-0.01428818263857071,13.924724578857422 -METTL17,12,0.01146024838089943,0.08136968314647675,0.9850174599266296,0.017699591469163323,7.047603607177734 -METTL23,12,0.011830048635601997,0.08369400352239609,0.9850355231676855,0.020097520091889653,7.160407066345215 -METTL3,14,0.011931126937270164,0.08274190127849579,0.9842799297726563,-0.09594479268665926,7.190931797027588 -MEX3A,17,0.00818157009780407,0.06836923211812973,0.9894734291192182,0.04976272603186509,5.954739570617676 -MFAP1,19,0.0158770103007555,0.0930509939789772,0.9787118653850094,0.2515720468432644,8.29523754119873 -MFN2,13,0.010370426811277866,0.07797373831272125,0.9862805148816318,0.07657681045382266,6.704135417938232 -MICOS10,8,0.017248056828975677,0.09983953088521957,0.9773783549881965,0.19787292627251946,8.64598560333252 -MIOS,17,0.0084474952891469,0.06932298094034195,0.9887554225254844,0.09335813128703659,6.050739765167236 -MIPEP,9,0.017175033688545227,0.10112998634576797,0.9775475283980152,0.14157073274716708,8.627664566040039 -MIS12,8,0.01774325780570507,0.10366589576005936,0.9766406806530744,0.06127698342700385,8.7692232131958 -MIS18A,13,0.010102600790560246,0.07674933969974518,0.9871464407035282,-0.038437489628350016,6.616998672485352 -MIS18BP1,10,0.01321498118340969,0.08858143538236618,0.9828774443793248,-0.0691722381111813,7.567940711975098 -MKRN1,16,0.00932406447827816,0.0735841616988182,0.9874309115298283,0.11441585113489378,6.356925010681152 -MLST8,8,0.018439127132296562,0.10483495146036148,0.9752682053943067,0.08427401559753928,8.939528465270996 -MMGT1,19,0.006296148523688316,0.06127814203500748,0.9919124141014328,0.013299440272871275,5.223744869232178 -MMS19,13,0.010409185662865639,0.07905323058366776,0.9865776983050674,0.0991728169752501,6.716651916503906 -MMS22L,5,0.02322506718337536,0.11776230484247208,0.9693458257248416,0.09212329412162486,10.032817840576172 -MNAT1,9,0.020188109949231148,0.10886983573436737,0.9731754650292392,0.15546637062708984,9.353890419006348 -MOB4,12,0.015763824805617332,0.09514801949262619,0.9791000047859153,0.1870012317381423,8.265616416931152 -MOCS3,12,0.01029650866985321,0.07752933353185654,0.9868403990215807,-0.06474070550299474,6.68019962310791 -MOG,13,0.010934102348983288,0.0801311805844307,0.9857476995931507,-0.04467630218614121,6.883923053741455 -MOK,13,0.01127877738326788,0.08222868293523788,0.9853295851122204,0.020581883953170805,6.99158239364624 -MPHOSPH10,5,0.0389118529856205,0.14902174472808838,0.9498242762830741,0.09979038937989926,12.986299514770508 -MRGBP,8,0.019698189571499825,0.10648825764656067,0.9746245983351755,-0.002421534680282706,9.239693641662598 -MRPL1,17,0.007215934339910746,0.06446537375450134,0.9908331364226592,0.056790346432183024,5.59230375289917 -MRPL10,10,0.01713378168642521,0.10010292381048203,0.9776795585252257,0.08102656597670194,8.617297172546387 -MRPL13,8,0.02544992044568062,0.12107732892036438,0.9671153392658314,0.16109039705240508,10.502378463745117 -MRPL14,21,0.007328343112021685,0.06338350474834442,0.9903745435312468,0.12718700555046356,5.635693550109863 -MRPL15,4,0.03467829152941704,0.14209534227848053,0.9566808594662889,0.030695292640667466,12.259514808654785 -MRPL16,9,0.015736708417534828,0.09690335392951965,0.9797670236647367,0.14246848005682555,8.258504867553711 -MRPL17,10,0.015584561042487621,0.09559252858161926,0.9804737443177304,0.05693922577930945,8.218484878540039 -MRPL18,18,0.00824410654604435,0.06784936040639877,0.9892055875312924,0.06602330852767943,5.977454662322998 -MRPL19,19,0.007404561620205641,0.06460428982973099,0.9905095470267348,0.18165694650830788,5.664924621582031 -MRPL2,14,0.011172118596732616,0.07999585568904877,0.9850498997731507,0.11086318038078817,6.9584455490112305 -MRPL20,12,0.013888603076338768,0.0880640298128128,0.981960183282605,0.068772266821141,7.75842809677124 -MRPL21,2,0.0735807791352272,0.2130471020936966,0.9140519793278803,-0.050297168434069665,17.85774803161621 -MRPL22,9,0.015223025344312191,0.09317895025014877,0.9797060863929667,0.022081713130762994,8.122597694396973 -MRPL24,20,0.007341038901358843,0.06465121358633041,0.990177154209421,0.0859594612500536,5.640572547912598 -MRPL27,10,0.013613206334412098,0.08810118585824966,0.982047096435328,0.04524064205818905,7.681122303009033 -MRPL28,5,0.0293795857578516,0.1329781711101532,0.9636780578224041,0.06387362482510772,11.284109115600586 -MRPL3,17,0.009218246676027775,0.07269503176212311,0.9882064322783665,0.10896549376452473,6.320750713348389 -MRPL32,6,0.021194826811552048,0.111275315284729,0.9740991961188848,0.06055725715552507,9.584278106689453 -MRPL33,15,0.010214620269834995,0.07619219273328781,0.9872114766072193,0.13472717963881584,6.65358304977417 -MRPL34,2,0.07572582364082336,0.21505746245384216,0.9086351423613499,-0.03851549461978849,18.116172790527344 -MRPL35,17,0.009703055955469608,0.07414349168539047,0.987569828430345,0.1026520117577673,6.4848313331604 -MRPL36,19,0.00878748670220375,0.0706305280327797,0.9881406917963887,0.14543050278633993,6.17130184173584 -MRPL37,11,0.014227855950593948,0.09133461862802505,0.9820060913562466,0.09763677395446631,7.8526129722595215 -MRPL38,15,0.009197622537612915,0.07313359528779984,0.9878116033378278,0.05253342542851323,6.313675403594971 -MRPL39,16,0.010434290394186974,0.07585394382476807,0.9864317994497382,0.1262175413599646,6.724746227264404 -MRPL4,13,0.009190094657242298,0.07267577201128006,0.9880986530723169,0.15296329493659475,6.31109094619751 -MRPL41,7,0.019689247012138367,0.10742712765932083,0.9749829554010128,0.12486982818262762,9.237597465515137 -MRPL43,9,0.01590714603662491,0.09552823007106781,0.9789584253667447,-0.0015721790101644478,8.303106307983398 -MRPL44,8,0.01533575914800167,0.09582488238811493,0.9804731166564926,0.05320289965423742,8.152617454528809 -MRPL45,18,0.007984102703630924,0.06853900104761124,0.989572646220636,0.049806681313274155,5.882440090179443 -MRPL49,4,0.04445965215563774,0.16207124292850494,0.939613782866979,0.1700859673854686,13.88121509552002 -MRPL50,17,0.007177624851465225,0.06391391158103943,0.9906583196962327,0.028062477658244392,5.577438831329346 -MRPL51,11,0.016124777495861053,0.0972306951880455,0.9793129793085742,0.13317967834325642,8.359712600708008 -MRPL53,16,0.009478098712861538,0.07379420846700668,0.9874160027752967,0.1643616830988648,6.409217834472656 -MRPL54,14,0.009044917300343513,0.0722963809967041,0.9883440190158268,-0.03377606546754149,6.261044025421143 -MRPL55,17,0.00837615318596363,0.06844180822372437,0.9892111363886849,0.09997976764574815,6.025135040283203 -MRPL9,12,0.011038459837436676,0.08022063225507736,0.9855269253442975,0.1343030518161205,6.916696071624756 -MRPS10,12,0.01268106047064066,0.08602064102888107,0.983740194757846,0.13704429864004916,7.413482189178467 -MRPS11,9,0.01603643223643303,0.09709422290325165,0.9797379398279507,0.09516390988549044,8.336779594421387 -MRPS12,5,0.02906133607029915,0.13274236023426056,0.9632005532809491,0.06491447128931385,11.22282600402832 -MRPS14,9,0.019226443022489548,0.10175788402557373,0.9764510833655335,0.09956271044277576,9.128384590148926 -MRPS16,4,0.03394824266433716,0.1418549120426178,0.9577987425800485,0.08076443938048726,12.12978458404541 -MRPS18A,12,0.011854605749249458,0.08274489641189575,0.9843790332327353,0.0846914363917317,7.167835235595703 -MRPS18B,6,0.02800525352358818,0.12883228063583374,0.9640276492070036,-0.009700857013464037,11.017021179199219 -MRPS2,11,0.011818967759609222,0.08370419591665268,0.9848857032925792,0.024596203424645698,7.157052993774414 -MRPS21,14,0.009903128258883953,0.07462912797927856,0.9871217341119232,0.1246136042585729,6.5513482093811035 -MRPS22,17,0.009532967582345009,0.07546533644199371,0.9878138650741503,0.1280392493457076,6.427742958068848 -MRPS23,12,0.013648937456309795,0.08851242810487747,0.9823380752262727,0.020475595272617074,7.691195964813232 -MRPS24,15,0.010480777360498905,0.07809736579656601,0.986464662645582,0.1219028554716723,6.739709854125977 -MRPS25,10,0.013898561708629131,0.09013652801513672,0.9823400085650087,-0.03681231891084758,7.761209011077881 -MRPS26,12,0.01186091173440218,0.08336833864450455,0.9846219000053987,0.15723809642242664,7.169741153717041 -MRPS27,8,0.01635969989001751,0.09815720468759537,0.9783731003218941,0.14485622772939274,8.420388221740723 -MRPS28,16,0.009086406789720058,0.07208358496427536,0.9880594961995669,0.12476939887248417,6.275387763977051 -MRPS31,10,0.013422494754195213,0.08788111805915833,0.9828020058065977,0.03645374175738375,7.627128601074219 -MRPS34,14,0.0153657803311944,0.09318786859512329,0.9796740695187621,0.10667435297851903,8.16059398651123 -MRPS35,20,0.007510511204600334,0.06621652096509933,0.9901357469401999,0.2049019947620585,5.7053093910217285 -MRPS5,8,0.01653285324573517,0.09752850979566574,0.9791615490941386,-0.09919629319836723,8.46483325958252 -MRPS6,15,0.008435367606580257,0.07036980241537094,0.9891034461236791,0.10527035523307014,6.046394348144531 -MRPS7,13,0.010180667042732239,0.07764974236488342,0.9868984276449585,0.08343398564495474,6.642515182495117 -MRPS9,7,0.026364494115114212,0.1236041933298111,0.9657807697820465,0.27377549003092966,10.689420700073242 -MRTO4,9,0.014095007441937923,0.0913131833076477,0.9822844121979333,0.043079566860144136,7.815865516662598 -MSRB1,21,0.0071578118950128555,0.06421085447072983,0.9904342354201142,0.05618145987543932,5.569736003875732 -MST1,14,0.009583570063114166,0.07439114153385162,0.9873306596465666,0.01844616312526036,6.4447808265686035 -MSTO1,10,0.013551526702940464,0.08979079872369766,0.9824429412311474,-0.06355720691666267,7.663701057434082 -MT2A,11,0.013496054336428642,0.08944466710090637,0.9817576735876276,0.051497368947204546,7.6479997634887695 -MTBP,19,0.0077103180810809135,0.06625021994113922,0.989784241712978,0.07357942067186823,5.780702114105225 -MTG2,14,0.009524212218821049,0.07467488199472427,0.9874920290617354,-0.0052413821597874334,6.424790382385254 -MTOR,10,0.024739686399698257,0.11729602515697479,0.9690678431424756,0.14944707027278634,10.354796409606934 -MTPAP,9,0.020527279004454613,0.10283365845680237,0.9749935262606473,0.02560764679408537,9.432137489318848 -MTRNR2L1,5,0.03581584617495537,0.14446285367012024,0.9536983729124001,0.006326384067826463,12.458968162536621 -MVD,11,0.014715355820953846,0.09108485281467438,0.9802750466165453,0.1323331510222926,7.98600959777832 -MVK,13,0.011464360170066357,0.0815822035074234,0.9847048381083011,0.10082965480213008,7.048867702484131 -MYBBP1A,11,0.01714194566011429,0.0988011360168457,0.9769541942726037,0.27163887916768625,8.619349479675293 -MYBL2,10,0.01608901284635067,0.09633700549602509,0.9792194417543288,0.12573876764584657,8.350436210632324 -MYC,8,0.01808796636760235,0.10326488316059113,0.9770603452323804,0.04025591240300713,8.853996276855469 -MYCBP,21,0.006220642011612654,0.0596788227558136,0.9916971030465701,0.07190396139329848,5.192327499389648 -MYO1H,13,0.01087254285812378,0.08045196533203125,0.9858982064309897,0.012848226027318674,6.864517688751221 -MZF1,13,0.011350656859576702,0.08176358044147491,0.9849019746095313,-0.05318397482753635,7.013825416564941 -MZT1,45,0.004019094631075859,0.047022946178913116,0.9945967761031498,0.023537388009244303,4.173578262329102 -N6AMT1,14,0.010790898464620113,0.07917365431785583,0.9866355708169169,-0.06534493159796419,6.838695049285889 -NAA10,13,0.011733971536159515,0.08333979547023773,0.9841508784900762,0.03221804933587595,7.1312713623046875 -NAA15,13,0.013286860659718513,0.08824235945940018,0.9822566055610842,0.10663603439301492,7.588494777679443 -NAA20,18,0.008335327729582787,0.06832277029752731,0.9888061724956124,0.09541696006784711,6.010433197021484 -NAA25,17,0.011095086112618446,0.07964378595352173,0.9850125731042146,0.16089615479365474,6.934414386749268 -NAA35,20,0.007497953716665506,0.06570258736610413,0.9899413524594263,0.11917168088155292,5.700538158416748 -NAA38,37,0.012924907729029655,0.08023998141288757,0.9832678900115375,0.2855053251985419,7.484420299530029 -NAA50,11,0.0155534902587533,0.09312910586595535,0.9799356406196,0.03673195262562629,8.210288047790527 -NACA,36,0.013012616895139217,0.07912877202033997,0.9828644974054862,0.33216748330157747,7.509772300720215 -NAE1,9,0.0162818506360054,0.09682633727788925,0.9790287023990769,-0.04101899028483795,8.40032958984375 -NAF1,10,0.014287012629210949,0.09108728170394897,0.9816059315414793,-0.007741170580899387,7.86892032623291 -NAGLU,27,0.004779974929988384,0.053468018770217896,0.9936336073687403,0.1097632577402931,4.551528453826904 -NAPA,1,0.13169226050376892,0.2873697280883789,0.8601306323169516,0.045526607941726964,23.890464782714844 -NAPG,6,0.02434154786169529,0.11893995851278305,0.9684147920986371,0.0512195555116784,10.271137237548828 -NARS,17,0.013772489503026009,0.08514486253261566,0.9813870696459708,0.32472949756845304,7.72592830657959 -NASP,19,0.007552028633654118,0.06475597620010376,0.9899222257508742,0.08733171148416691,5.7210564613342285 -NAT10,3,0.057744864374399185,0.18492159247398376,0.9275447884953152,0.10757518917107675,15.819805145263672 -NBAS,10,0.01447763480246067,0.09160005301237106,0.9806786745638045,0.08355511520797435,7.921241760253906 -NBEAL1,10,0.012439972721040249,0.0847940519452095,0.9836922928918544,0.10983320873144557,7.342672348022461 -NBPF12,15,0.008131083101034164,0.06868194043636322,0.9897506200616769,-0.057805554219282115,5.936338424682617 -NBPF15,6,0.017305394634604454,0.09879443049430847,0.9787641202945275,-0.05290396519737054,8.660346031188965 -NBPF3,14,0.00922822393476963,0.07473226636648178,0.9878072896452592,0.02267283513915835,6.324169635772705 -NCAPD2,5,0.030495963990688324,0.13359488546848297,0.9605090056155505,0.11817544760795888,11.496500015258789 -NCAPD3,8,0.01862921565771103,0.1054878830909729,0.975785438685804,-6.376632149340353e-05,8.985489845275879 -NCAPG,8,0.016863897442817688,0.09926102310419083,0.9772476785491203,0.1385779152508411,8.54916000366211 -NCAPG2,9,0.018328428268432617,0.1041387990117073,0.975577335537839,-0.02286143690507107,8.912654876708984 -NCAPH,7,0.0195067897439003,0.10655920207500458,0.9743185644971124,0.1830087828944286,9.194695472717285 -NCAPH2,15,0.00947897881269455,0.07373746484518051,0.9876369568103998,-0.036252997441542006,6.409516334533691 -NCBP1,6,0.03611057624220848,0.1431465893983841,0.9523991690871607,0.024880973708671907,12.510125160217285 -NCBP2,99,0.009405253455042839,0.06347265839576721,0.9875080069345988,0.09934477350152283,6.3845415115356445 -NCKAP1,7,0.027537938207387924,0.1265580952167511,0.9633472334110202,0.1121961603719383,10.924715995788574 -NCL,30,0.00470556179061532,0.0517389178276062,0.9941253167988979,-0.03314866063441268,4.515961647033691 -NCOA4,14,0.010594063438475132,0.07876323908567429,0.9860334304856191,0.054055772213372974,6.776036739349365 -NDC1,4,0.04797565937042236,0.1660510152578354,0.9379746494606946,-0.07149899822354097,14.419657707214355 -NDC80,22,0.008800213225185871,0.07127323001623154,0.9883554066937266,0.27741625917125157,6.175769329071045 -NDOR1,4,0.06311619281768799,0.19271312654018402,0.9216480081428674,0.10528073177256794,16.539213180541992 -NDUFA11,8,0.01739376410841942,0.10132638365030289,0.9771893173308487,0.08290318452200743,8.682429313659668 -NDUFA2,18,0.014277036301791668,0.08833764493465424,0.9808713254736745,0.2868458695905828,7.866173267364502 -NDUFA4,8,0.018399402499198914,0.10154817998409271,0.9768599422676103,-0.014371421986619394,8.92989444732666 -NDUFA6,17,0.007594432216137648,0.06636222451925278,0.9899737633212519,0.0974980152180227,5.737095832824707 -NDUFA8,11,0.011843427084386349,0.0827559232711792,0.9843529746976007,-0.029450579028654604,7.164454936981201 -NDUFAB1,16,0.00924007035791874,0.07184354215860367,0.9878151922496257,0.01135966102015515,6.3282270431518555 -NDUFAF3,8,0.017305558547377586,0.10193746536970139,0.9783770293252096,0.15587654956690658,8.660386085510254 -NDUFAF7,2,0.057269662618637085,0.18846093118190765,0.9319904301938337,-0.045753019870552365,15.75457763671875 -NDUFB10,16,0.009980587288737297,0.07520845532417297,0.9870598905744257,0.08846870267355611,6.576919078826904 -NDUFB3,14,0.008976442739367485,0.0713597983121872,0.9884495739355591,-0.003339849876749043,6.23729944229126 -NDUFB4,16,0.008949161507189274,0.07192710041999817,0.9883295624824133,-0.027485826381698393,6.227814197540283 -NDUFB6,19,0.007344440557062626,0.06511522829532623,0.9904834238076251,-0.003598890601486466,5.641879558563232 -NDUFB7,8,0.013741662725806236,0.08980721980333328,0.9826441964253042,-0.018389165120253237,7.7172770500183105 -NDUFB8,20,0.007287272252142429,0.06449036300182343,0.9904840891539665,-0.016294050507521855,5.619878768920898 -NDUFS3,11,0.012552464380860329,0.08603610098361969,0.9830759496037267,0.022724448346548798,7.375797271728516 -NDUFS5,19,0.006932092364877462,0.06350603699684143,0.9907588965434839,0.0703008525109275,5.481212139129639 -NEDD1,14,0.009116620756685734,0.07293450087308884,0.98830783225123,-0.0037497209341327653,6.285811901092529 -NEDD8,19,0.01606632024049759,0.09169884771108627,0.9788750025403186,0.0866376180218459,8.344545364379883 -NELFA,9,0.022874590009450912,0.11356630176305771,0.9696565566848573,0.07796456727288838,9.956830978393555 -NELFB,14,0.012200496159493923,0.08401492238044739,0.9834512915106837,0.10067946635507445,7.271653652191162 -NELFCD,20,0.008482334204018116,0.06906966120004654,0.988561981126936,-0.024567735525049574,6.063203811645508 -NELFE,8,0.016366906464099884,0.09833730757236481,0.9790101435037643,0.07752995793467411,8.422243118286133 -NEMF,7,0.01691451482474804,0.09903960675001144,0.9796844190425757,0.07991645118438512,8.561980247497559 -NEPRO,4,0.04400591179728508,0.1580437570810318,0.9444548324190243,0.08041852892158587,13.810199737548828 -NFATC2IP,12,0.00990526657551527,0.07689400017261505,0.9877552339231046,-0.026127015949313555,6.5520548820495605 -NFKBIE,5,0.03198745846748352,0.13651590049266815,0.9594654046372671,-0.05412643466867646,11.774279594421387 -NFRKB,23,0.007832497358322144,0.06673343479633331,0.9894849061377476,0.01956154363031118,5.826323509216309 -NFS1,8,0.02018776908516884,0.10883420705795288,0.9736434117108156,0.1605598997352038,9.353811264038086 -NFYB,3,0.03637267276644707,0.145546555519104,0.9573030738530452,-0.09060451920968286,12.555442810058594 -NFYC,3,0.055610883980989456,0.1788257509469986,0.9255205337940521,0.17507701033889636,15.524741172790527 -NHLRC2,9,0.011943431571125984,0.08403713256120682,0.9845823739932289,-0.031518488890437235,7.194638729095459 -NHP2,6,0.024321911856532097,0.12037689238786697,0.9681030855291428,0.08560132990786591,10.266993522644043 -NIFK,24,0.010343782603740692,0.07359005510807037,0.9861497425585347,0.3355792212196418,6.695517539978027 -NIP7,28,0.006814620923250914,0.061251651495695114,0.9908732743899284,0.24556854679306486,5.434571743011475 -NISCH,15,0.008433562703430653,0.06981547176837921,0.9888622838857208,0.06367780912914305,6.04574728012085 -NKAP,13,0.015636101365089417,0.09317447245121002,0.9802655761614936,0.1455669618163318,8.232062339782715 -NLE1,42,0.007614282891154289,0.06291602551937103,0.9899390008137718,0.3658916906400214,5.744588851928711 -NMD3,7,0.01905033364892006,0.1055305078625679,0.9768018176916651,0.0810896336645979,9.086482048034668 -NMT1,3,0.039478350430727005,0.1537514477968216,0.9530200610359597,-0.02114067670544787,13.080488204956055 -NOB1,6,0.028390511870384216,0.12899550795555115,0.962322996410922,-0.01081604400243015,11.092541694641113 -NOC2L,3,0.06082926318049431,0.19453519582748413,0.925493229236339,0.10149852542645171,16.2368106842041 -NOC3L,9,0.017664244398474693,0.10152462124824524,0.9770226951346263,0.1266212169658307,8.749676704406738 -NOC4L,4,0.04884608834981918,0.16889023780822754,0.9366250695480777,0.10507084869797681,14.549878120422363 -NOL10,3,0.05121641233563423,0.17538122832775116,0.9334275326179914,0.05519989793556016,14.898721694946289 -NOL11,27,0.013096379116177559,0.0813756212592125,0.983352631452292,0.2831137681039049,7.533904075622559 -NOL12,10,0.015438081696629524,0.09423351287841797,0.9805488295600165,0.12592925962559906,8.179771423339844 -NOL6,12,0.023486187681555748,0.11352869123220444,0.969072064192434,0.2495173167411409,10.08906078338623 -NOL7,8,0.01643184944987297,0.0999128594994545,0.9787189872780927,0.06477716787688213,8.438936233520508 -NOL8,8,0.018799737095832825,0.10478925704956055,0.9755394777860219,0.23181565161365758,9.026518821716309 -NOL9,4,0.024618735536932945,0.12074285000562668,0.970958431659372,-0.0014319857669036673,10.329453468322754 -NOLC1,12,0.012230372987687588,0.08366188406944275,0.9845485335588446,-0.09434718054398922,7.280551910400391 -NOM1,8,0.016587624326348305,0.09845760464668274,0.9777304420652393,0.12566525622492125,8.478841781616211 -NOMO3,19,0.007322230376303196,0.06544574350118637,0.9901645708786543,0.023412353298797255,5.633342266082764 -NOP10,14,0.01198673341423273,0.08177062124013901,0.9843444934247856,-0.001353863139754105,7.207669734954834 -NOP14,4,0.04611080884933472,0.1668829321861267,0.940156355420519,0.03542085554755899,14.136627197265625 -NOP16,15,0.014657558873295784,0.0907968282699585,0.9806861887680212,0.23339160975027423,7.970311164855957 -NOP2,6,0.03777207434177399,0.1466301530599594,0.9500886540403763,0.2861448703241077,12.794692039489746 -NOP56,10,0.012587146833539009,0.08661559224128723,0.9835440876952035,0.06750311193883304,7.385979652404785 -NOP58,5,0.03276633098721504,0.13542047142982483,0.9569904440706691,0.12928322150521565,11.916764259338379 -NOP9,8,0.01673022098839283,0.09886697679758072,0.9776953440042241,0.0768677134929175,8.515209197998047 -NOS1AP,6,0.018976910039782524,0.10525544732809067,0.9759274256528437,-0.00446006026169013,9.068953514099121 -NPAT,2,0.08077067881822586,0.2138313204050064,0.9005420575705989,0.10346621957886849,18.70989418029785 -NPB,19,0.007581210695207119,0.06661980599164963,0.9899877288299187,-0.029684257525698858,5.732100009918213 -NPEPPS,16,0.008204313926398754,0.06923272460699081,0.9892951363924003,0.03913441271128835,5.963010787963867 -NPLOC4,6,0.03773542493581772,0.14436906576156616,0.9499297172045016,0.22269595805055303,12.788484573364258 -NPM1,2,0.08323860168457031,0.22811979055404663,0.9029301145920388,0.05165463825612875,18.993581771850586 -NPM3,6,0.024304894730448723,0.12005054950714111,0.9690714945445381,-0.01326501776933175,10.263401985168457 -NR2C2AP,8,0.015631932765245438,0.09566708654165268,0.9798552473094656,-0.05745337488035721,8.230965614318848 -NRBP1,7,0.029675951227545738,0.13296130299568176,0.9605577291664472,0.12737235415909515,11.34088134765625 -NRDE2,51,0.009582022204995155,0.06772485375404358,0.9875140871759888,0.23837223437627839,6.4442596435546875 -NRF1,7,0.02060506120324135,0.109522745013237,0.9730098954653996,0.05162489129572722,9.449991226196289 -NSA2,12,0.010773786343634129,0.07947532832622528,0.9861132680103584,0.061088224625991026,6.83327054977417 -NSF,17,0.007460808381438255,0.06671755015850067,0.9902728598674003,-0.024338921828906443,5.686399936676025 -NSL1,7,0.020759526640176773,0.11054927855730057,0.9733266806961732,0.12904351989252125,9.485345840454102 -NSMCE1,6,0.025555042549967766,0.12219548225402832,0.9666153085644889,0.011833132201002938,10.524045944213867 -NSMCE2,12,0.010552467778325081,0.07888692617416382,0.9866152716356541,-0.06048919396724873,6.762721061706543 -NSRP1,23,0.008547459729015827,0.06933534145355225,0.9890714968652131,0.21831041956433098,6.086434841156006 -NSUN4,15,0.008105129934847355,0.06841867417097092,0.9892185527340298,-0.01238499392853266,5.926856994628906 -NUBP1,15,0.008392860181629658,0.0702061876654625,0.9888647034735493,0.11405409298490067,6.031140327453613 -NUBP2,16,0.007971728220582008,0.06888166815042496,0.9894528348350577,0.03891304040989682,5.877880096435547 -NUDC,11,0.01425568200647831,0.0903824046254158,0.9810453817686587,0.05664399095812129,7.860287666320801 -NUDCD3,11,0.017947128042578697,0.10045544803142548,0.9762494993028575,0.1014048603664889,8.8194580078125 -NUDT21,6,0.03748467564582825,0.14241476356983185,0.9487299851799466,0.26457769062047304,12.74592399597168 -NUF2,51,0.0042847138829529285,0.04886464774608612,0.9945960761481886,0.2536601364097205,4.309286594390869 -NUFIP1,7,0.023752156645059586,0.11840475350618362,0.9698221160786705,0.10250762548692977,10.146025657653809 -NUMA1,29,0.007343321572989225,0.06395221501588821,0.9903873951920359,0.1506023920952899,5.641449928283691 -NUP107,31,0.009302434511482716,0.06969306617975235,0.9876936130241926,0.2357612905496542,6.349546909332275 -NUP133,8,0.018991658464074135,0.10463929921388626,0.975334552360958,0.07107046520721624,9.072477340698242 -NUP153,8,0.017626915127038956,0.10187960416078568,0.9770378482734343,0.06487343281609544,8.740426063537598 -NUP155,1,0.10776880383491516,0.26329630613327026,0.885146959733922,-0.00910898324906925,21.61180305480957 -NUP160,29,0.008246900513768196,0.06686270236968994,0.9888639478466548,0.21756250199606145,5.978466987609863 -NUP205,73,0.002450497355312109,0.03747120499610901,0.9967049299127188,0.11964636786742225,3.258903980255127 -NUP214,13,0.01065424457192421,0.07794767618179321,0.9860298097277356,-0.07105731752580745,6.795255184173584 -NUP35,7,0.02204538881778717,0.11461592465639114,0.9719674137104383,0.006473923055996942,9.774697303771973 -NUP43,3,0.05555848404765129,0.18303139507770538,0.9352906361382446,0.020655683471056337,15.517424583435059 -NUP54,29,0.011547254398465157,0.07652481645345688,0.9849205776144221,0.2560443423089403,7.074305534362793 -NUP62,10,0.0344046950340271,0.13410329818725586,0.9536297127425538,0.2657512853574339,12.211058616638184 -NUP85,11,0.014033234678208828,0.08990408480167389,0.981174641402226,0.05316768800001068,7.798720359802246 -NUP88,3,0.0619041807949543,0.19087107479572296,0.9203496551083926,0.11482431950767993,16.379642486572266 -NUP93,16,0.008959543891251087,0.0718778669834137,0.9880785321011137,0.1085287975230135,6.2314252853393555 -NUP98,51,0.009342489764094353,0.06791003048419952,0.9877819066362464,0.4256643378882819,6.3632025718688965 -NUS1,13,0.010391538962721825,0.07709364593029022,0.9863907171762368,0.04911884066134676,6.71095609664917 -NUTF2,6,0.03294535353779793,0.13827194273471832,0.9592194262552263,0.14423823945197187,11.949275016784668 -NVL,11,0.011202524416148663,0.08076301962137222,0.985746667688147,0.14585827832879886,6.967907905578613 -NXF1,25,0.0054021873511374,0.056761737912893295,0.9928879459678224,-0.02660536442730759,4.8387064933776855 -OGFOD1,12,0.013661741279065609,0.08804072439670563,0.9818184289224119,0.11018685708759415,7.694802761077881 -OGT,3,0.03755439445376396,0.1489671915769577,0.9537348920190326,-0.06397267673636531,12.757772445678711 -OIP5,13,0.01026647537946701,0.07753926515579224,0.9864761349211701,-0.054628527980678084,6.670450210571289 -OLFML3,18,0.0075955623760819435,0.06658501923084259,0.9898347753361934,0.0556834050697552,5.737522125244141 -OPA1,6,0.023941028863191605,0.1178426444530487,0.9696338999146586,0.07259752218197792,10.186285972595215 -OR1D2,13,0.009939995594322681,0.07678698748350143,0.9870154200586004,-0.015696565102717795,6.563530921936035 -OR1E2,3,0.034666188061237335,0.1445278376340866,0.9602492454250171,0.0537975261100337,12.25737476348877 -OR2A1,19,0.007285825442522764,0.06543086469173431,0.9902880194281054,-0.041769553425938986,5.619320869445801 -OR2T29,11,0.01451553963124752,0.0922984853386879,0.9815477916145958,0.16417830648183598,7.931604385375977 -OR2T5,10,0.014256763271987438,0.09223740547895432,0.9811060950884292,0.05044711478849423,7.860585689544678 -OR4F4,6,0.020554766058921814,0.1108468621969223,0.9744658597439654,0.04600279152334797,9.438450813293457 -OR4K1,2,0.058276183903217316,0.18993128836154938,0.9279429954756542,-0.05559950099630669,15.89241886138916 -OR4N2,12,0.010860658250749111,0.08058114349842072,0.986235023215124,-0.08768662397769926,6.860764503479004 -OR7A10,4,0.03124045953154564,0.13681119680404663,0.9626028816429195,-0.07125501091747465,11.635985374450684 -OR7A17,11,0.011732283979654312,0.08365755528211594,0.9850697662302561,-0.09538192160701585,7.130758762359619 -ORC1,53,0.005076564848423004,0.05222509056329727,0.9934053829727676,0.26336598882387846,4.690611362457275 -ORC3,18,0.007499186787754297,0.06657927483320236,0.9900447152208224,0.08497078280077847,5.7010064125061035 -ORC4,9,0.01379650179296732,0.08994246274232864,0.9833675403567498,0.009954074155792617,7.732660293579102 -ORC5,9,0.01648028753697872,0.09912879765033722,0.9787982771508967,0.02200815943848423,8.451364517211914 -ORC6,8,0.017533574253320694,0.10120365768671036,0.9769875647344504,0.11070356182225291,8.717252731323242 -OSBP,7,0.02074742503464222,0.1111040860414505,0.9737758537662855,0.019528117646908195,9.48258113861084 -OSGEP,7,0.017887892201542854,0.10251987725496292,0.9766020332654691,0.0002505225679348508,8.804891586303711 -OSTC,23,0.006245937664061785,0.06000605225563049,0.9920954346508023,-0.027761920104432482,5.202873706817627 -OTOP1,11,0.011842571198940277,0.08292632550001144,0.9847400725836035,-0.050564965915616274,7.164196014404297 -OTX1,11,0.01205460075289011,0.0841473713517189,0.9841377222079517,-0.06571386542019989,7.228045463562012 -OVCA2,7,0.0179340448230505,0.10242585092782974,0.977147391428373,-0.0998270220358979,8.816243171691895 -OXA1L,35,0.009852152317762375,0.07001367956399918,0.9869782397186354,0.29942359564123555,6.534464359283447 -PABPC1,2,0.135708749294281,0.28422561287879944,0.8228019305651257,0.10386428995943374,24.252046585083008 -PABPN1,15,0.014283133670687675,0.08747807145118713,0.980918176819776,0.09478304091749631,7.867852687835693 -PAF1,24,0.03502664715051651,0.13412071764469147,0.9615128517136947,0.24345076858469133,12.320937156677246 -PAFAH1B1,12,0.011295625939965248,0.0822390764951706,0.9853268373414622,0.04188056021010852,6.99680233001709 -PAK1IP1,2,0.08233935385942459,0.22684694826602936,0.9096172366205812,0.06469850205432004,18.890705108642578 -PALB2,9,0.015256647020578384,0.09516466408967972,0.9802201782777762,-0.09163956953355655,8.131562232971191 -PAM16,5,0.031920477747917175,0.13684608042240143,0.9587223275015775,0.11271351642210996,11.761944770812988 -PARN,13,0.009989403188228607,0.07746005058288574,0.9867819906112956,0.026114998736877355,6.5798234939575195 -PARS2,6,0.023037848994135857,0.11680875718593597,0.9700084572416199,0.10670128738977107,9.99229907989502 -PAXBP1,13,0.014353032223880291,0.08916523307561874,0.9806561480600161,0.2381789226539597,7.887080669403076 -PC,23,0.0061868722550570965,0.05994287505745888,0.991922239715059,-0.021355287413840945,5.1782145500183105 -PCBP1,11,0.013335869647562504,0.08666934072971344,0.9827125282258075,-0.032266127091680905,7.602477073669434 -PCBP2,17,0.007326698862016201,0.0646706223487854,0.9907225988627203,-0.04506250595384704,5.635060787200928 -PCF11,50,0.007699794601649046,0.061836127191782,0.9901591980349692,0.22330540333071752,5.776756286621094 -PCID2,30,0.005109772086143494,0.05398854613304138,0.9934699084887871,-0.0641719716988798,4.70592737197876 -PCNA,13,0.02136598713696003,0.10879160463809967,0.9722721553290913,0.19846647245616253,9.622899055480957 -PDAP1,5,0.030798668041825294,0.13390956819057465,0.9595639060841855,0.04260698990741479,11.55341625213623 -PDCD11,10,0.028580982238054276,0.12492431700229645,0.9621341703240749,0.2536758657099629,11.12968921661377 -PDCD2,8,0.016971882432699203,0.10103873163461685,0.9776839011362648,-0.04138198979426472,8.57648754119873 -PDCD5,20,0.006801146548241377,0.06196051463484764,0.9909950987701276,0.10787100243656149,5.429195880889893 -PDCD6,21,0.006275153253227472,0.06006366387009621,0.9916351182329939,0.06786969532332199,5.215027332305908 -PDCD6IP,25,0.006424485705792904,0.06070520728826523,0.9914896507104564,-0.0587090754017527,5.27671480178833 -PDCD7,10,0.01751878298819065,0.10163535177707672,0.9784511017461012,0.10305536830068499,8.71357536315918 -PDPK1,8,0.02174975536763668,0.1139744222164154,0.9727489301985157,0.0718348828064657,9.70893669128418 -PDRG1,11,0.01320278737694025,0.08785974234342575,0.9823007429511041,0.06798988571766233,7.56444787979126 -PEF1,5,0.03161764144897461,0.13537025451660156,0.9598312882234853,0.05427106422863043,11.706018447875977 -PELO,28,0.0065328083001077175,0.060667552053928375,0.9912654527278364,0.25793172792800945,5.321013927459717 -PELP1,7,0.024247299879789352,0.11866316199302673,0.9689153508249936,0.10998645668346814,10.25123405456543 -PES1,7,0.025358552113175392,0.11977542191743851,0.9661020503348071,0.22828074009916474,10.483509063720703 -PET117,9,0.013127496466040611,0.08822362869977951,0.9827435473370997,-0.04719404900481544,7.542849063873291 -PEX1,6,0.023184889927506447,0.11804451048374176,0.9698540284029754,0.017517886349346246,10.024136543273926 -PFDN1,4,0.03530038148164749,0.14613334834575653,0.9547146232206742,-0.059548454993643314,12.368987083435059 -PFDN2,8,0.022245941683650017,0.11216475814580917,0.9702849735421062,0.11763698444483907,9.819058418273926 -PFDN4,4,0.037291910499334335,0.1488974392414093,0.9511175787654027,0.06620534953144121,12.713109016418457 -PFDN5,11,0.012599020265042782,0.08485943078994751,0.9834799490271423,0.0447413662634234,7.389462471008301 -PFDN6,11,0.012156413868069649,0.08556009083986282,0.9838038251629709,-0.007411411781013026,7.258504867553711 -PFN1,26,0.006174604408442974,0.057323142886161804,0.9916877802554585,0.03982742430127705,5.173077583312988 -PGAM1,16,0.011073416098952293,0.07989033311605453,0.9856867115528869,0.04242752998316923,6.927639007568359 -PGD,11,0.015017644502222538,0.09203360974788666,0.9798875818338831,0.017076875132359202,8.067618370056152 -PGK1,4,0.03404884412884712,0.1443697065114975,0.9584200509475246,0.034606444699138414,12.147743225097656 -PGPEP1,5,0.03908130154013634,0.15124094486236572,0.9495546801714894,0.15537197760053262,13.014544486999512 -PGS1,11,0.011111032217741013,0.08065367490053177,0.9855246906516261,-0.037540420636531936,6.939395427703857 -PHAX,41,0.007083803880959749,0.06133578345179558,0.990548201542437,0.19154861535572343,5.540866851806641 -PHB,25,0.030575614422559738,0.12002229690551758,0.9602481854323656,0.4636823649220472,11.511504173278809 -PHB2,41,0.025235040113329887,0.10653068125247955,0.9672761887153671,0.47359594266385074,10.457947731018066 -PHF12,11,0.011655624024569988,0.08249097317457199,0.9850045116023155,0.06306866592702683,7.107424259185791 -PHF5A,23,0.013821824453771114,0.08523079007863998,0.9816837121037398,0.3471763440471348,7.7397541999816895 -PI4KA,11,0.012500911019742489,0.08507299423217773,0.983404453191623,-0.013193329383283305,7.360634803771973 -PIAS4,20,0.010805079713463783,0.07612407952547073,0.985872841155354,0.28764766079392556,6.84318733215332 -PIGH,9,0.01432392280548811,0.0915582925081253,0.9814185910677975,0.012752500847461932,7.879078388214111 -PIK3C3,11,0.014100517146289349,0.09136035293340683,0.9812908469437168,-0.04765118413012691,7.8173933029174805 -PIK3R4,7,0.022039324045181274,0.11366834491491318,0.9710193819025144,0.014270556406794515,9.773353576660156 -PKM,15,0.01034076139330864,0.07808392494916916,0.9867648770726904,-0.007968752569461387,6.694539546966553 -PKMYT1,9,0.015867844223976135,0.09681545943021774,0.9791197298351735,0.14708093448193874,8.292841911315918 -PLA2G10,20,0.006433332338929176,0.06201183795928955,0.9914763538620889,0.0680862960466363,5.280346870422363 -PLEKHN1,11,0.010728397406637669,0.08019008487462997,0.9861955625473313,0.054798200730903884,6.818861961364746 -PLK1,6,0.02291681244969368,0.11616937816143036,0.9701460576977177,-0.045270814361284314,9.966015815734863 -PLK4,3,0.04740312322974205,0.16979853808879852,0.9407879716000737,-0.05014337464653125,14.333356857299805 -PLRG1,10,0.01923435740172863,0.10389979183673859,0.9742453950824786,0.15127675379315828,9.130263328552246 -PMF1,27,0.006849630270153284,0.061690740287303925,0.9911516873538369,0.16742223747293275,5.448513507843018 -PMPCA,14,0.012507940642535686,0.0836251899600029,0.9835307146221002,0.18041173908191638,7.362704277038574 -PMPCB,56,0.008185001090168953,0.06250876933336258,0.9895152398146632,0.4052687323454617,5.955987930297852 -PNISR,19,0.008345566689968109,0.069124236702919,0.9888749271162618,0.011525402515386285,6.014123916625977 -PNKP,19,0.0069414954632520676,0.06439132243394852,0.9910984904997193,0.008874569529847559,5.484928607940674 -PNN,8,0.019901257008314133,0.10651975870132446,0.9736681936397761,0.17911910798614353,9.28719711303711 -PNO1,14,0.01909784786403179,0.09963638335466385,0.9746605271518988,0.20254960822864387,9.097805976867676 -PNPT1,13,0.014026964083313942,0.0858176127076149,0.9829299667654802,0.07863408714591526,7.796977996826172 -POGZ,4,0.03886238858103752,0.15185105800628662,0.9523156106785753,0.02974191127723762,12.978043556213379 -POLA1,23,0.008540057577192783,0.06931116431951523,0.9884724134327304,0.0759979446791178,6.083798885345459 -POLA2,24,0.010018028318881989,0.07285969704389572,0.9864625982421878,0.17879635050713757,6.5892438888549805 -POLD1,21,0.012633800506591797,0.07804009318351746,0.9829212418783368,0.13911103012979417,7.399654865264893 -POLD2,8,0.020480044186115265,0.10756427049636841,0.9727044003083029,0.05903357284193407,9.421279907226562 -POLD3,7,0.021815180778503418,0.11177484691143036,0.9707785140831489,0.07184387631604452,9.723527908325195 -POLE,17,0.012431876733899117,0.08365176618099213,0.9833692609914128,0.18630317851594477,7.340282440185547 -POLE2,5,0.021223168820142746,0.11223259568214417,0.9739944087094643,-0.10342899776816897,9.590683937072754 -POLG2,15,0.009622589685022831,0.07397864013910294,0.9875396638907459,-0.06605301449216792,6.457887172698975 -POLR1A,12,0.01670023426413536,0.09607569128274918,0.977700719034737,0.22836561521488563,8.507573127746582 -POLR1B,18,0.020315783098340034,0.10027294605970383,0.9728155727271864,0.2989956172217116,9.383421897888184 -POLR1C,18,0.009173099882900715,0.07339967787265778,0.9876637177130348,0.1572396742983331,6.305253028869629 -POLR1E,21,0.008923474699258804,0.07063060253858566,0.9884054523242237,0.16689724201056158,6.218869686126709 -POLR2A,41,0.0069003039970994,0.06057923659682274,0.9911518800458529,0.18960716424029703,5.468630313873291 -POLR2B,32,0.014491247944533825,0.08559378236532211,0.9810725584648818,0.17780006684147068,7.924964904785156 -POLR2C,19,0.01811605878174305,0.09851383417844772,0.9762932377998107,0.17271717988090987,8.860869407653809 -POLR2D,28,0.0068889399990439415,0.06234245374798775,0.9908044480922606,0.10599952111277651,5.464125156402588 -POLR2E,23,0.007195908110588789,0.06508146226406097,0.9902742243675656,0.1004841801956302,5.584537982940674 -POLR2F,33,0.006655884440988302,0.06078706681728363,0.9912929233771047,0.19274982223580991,5.370903491973877 -POLR2G,15,0.023512203246355057,0.11130673438310623,0.9689675194470597,0.17284827421535556,10.094646453857422 -POLR2H,10,0.049872029572725296,0.1624605804681778,0.9336876673575818,0.21357358572871893,14.701883316040039 -POLR2I,23,0.02219066582620144,0.10672131180763245,0.9719016185628833,0.2202148777219421,9.806852340698242 -POLR2K,2,0.1146555095911026,0.2644241750240326,0.8528397319107894,0.24759115250323452,22.291635513305664 -POLR2L,6,0.035431407392024994,0.14439083635807037,0.9531099312923996,0.09130550974239264,12.391921043395996 -POLR3A,1,0.13522756099700928,0.2918654978275299,0.8585765263291316,0.11627271741563182,24.20901107788086 -POLR3B,12,0.012878851033747196,0.08808217197656631,0.9836652220107103,0.12699503182557742,7.471073627471924 -POLR3C,17,0.009314903058111668,0.07362538576126099,0.9877672372124939,-0.06721594509988532,6.353801250457764 -POLR3D,12,0.0112288948148489,0.08073557913303375,0.9852231180490836,0.03222423548407784,6.976103782653809 -POLR3E,22,0.006508220452815294,0.06199462339282036,0.9917052596138773,-0.02229575814750838,5.310991287231445 -POLR3F,17,0.009425852447748184,0.07375626266002655,0.9876691176327853,0.0799790766429548,6.391529560089111 -POLR3H,11,0.011187401600182056,0.0801207646727562,0.9862016250739432,-0.07122310940990513,6.963203430175781 -POLR3K,11,0.015354945324361324,0.09291667491197586,0.9796056120192775,0.08173460408450604,8.157715797424316 -POLRMT,22,0.011720743030309677,0.07123154401779175,0.9850755232845911,0.22183954827275254,7.127249717712402 -POP1,9,0.015969006344676018,0.09674786031246185,0.9789856457405709,0.04815812976328139,8.319233894348145 -POP4,13,0.0124701838940382,0.08491607010364532,0.9835327909440698,0.06883370618739447,7.351583003997803 -POP5,5,0.03218952566385269,0.1382211595773697,0.9573787310000288,0.10895868598028799,11.811409950256348 -POP7,22,0.006832131650298834,0.06310706585645676,0.9909266645545235,0.07557748297557045,5.441549301147461 -POT1,1,0.1252860277891159,0.27319157123565674,0.8682210493148806,0.15989734985004098,23.30213737487793 -POTEI,12,0.014188617467880249,0.08971115201711655,0.9818154740717782,0.056764858912669476,7.841777324676514 -POU5F1B,8,0.014935536310076714,0.09354916214942932,0.9808336593477709,0.06597993619266763,8.045534133911133 -PPA1,15,0.009444006718695164,0.07281232625246048,0.9873612547004899,0.1312653599762497,6.397680759429932 -PPA2,7,0.022168830037117004,0.11597637832164764,0.9728291937977218,-0.0022002238302525627,9.802026748657227 -PPAN,11,0.014815141446888447,0.09153752774000168,0.9801884022828123,0.17865050295494844,8.013040542602539 -PPARGC1B,14,0.010263333097100258,0.07696506381034851,0.9864169029920857,0.04572684957879654,6.669428825378418 -PPIA,16,0.009040376171469688,0.06858249008655548,0.9884123155652332,0.04659841316284068,6.259472370147705 -PPIE,4,0.0412026084959507,0.15311506390571594,0.9476387830697991,0.13941783734638566,13.363086700439453 -PPIH,12,0.011891797184944153,0.08385700732469559,0.9845977958128452,0.09101034488341053,7.179069995880127 -PPIL1,3,0.042914777994155884,0.16140928864479065,0.9465482075140987,-0.02938774670001068,13.637911796569824 -PPIL2,13,0.010522780008614063,0.0788164809346199,0.9862775152010351,-0.00335637869447045,6.753201484680176 -PPIL4,8,0.03930268809199333,0.1473974585533142,0.9511261281315264,0.2034358631623293,13.051353454589844 -PPME1,22,0.006578418891876936,0.061796531081199646,0.9912852767974532,0.033082007127941924,5.339556694030762 -PPP1CA,25,0.00548557797446847,0.05678851529955864,0.9929569254891186,0.0038284902762152646,4.875909805297852 -PPP1CB,16,0.008385355584323406,0.07000233232975006,0.9890037720357181,0.01094067493613203,6.028442859649658 -PPP1R10,1,0.247700497508049,0.37574097514152527,0.7310902525433842,0.17776876401480537,32.764827728271484 -PPP1R11,7,0.018803292885422707,0.1045302227139473,0.9768001461683492,-0.025263509427771547,9.027373313903809 -PPP1R12A,18,0.00665507186204195,0.06306473910808563,0.9913719952696295,0.034557921347335915,5.370575904846191 -PPP1R15B,13,0.010140258818864822,0.07618185132741928,0.9865979278881889,0.005707385733856988,6.629319667816162 -PPP1R2,8,0.015585879795253277,0.09556736797094345,0.9798648513130033,0.004493750165030732,8.218832015991211 -PPP1R7,7,0.01882670447230339,0.10637231916189194,0.9749762336127312,0.1524054917152246,9.032991409301758 -PPP1R8,7,0.021576279774308205,0.11096909642219543,0.9729198736460573,0.026334285100132828,9.67013931274414 -PPP2CA,19,0.007897559553384781,0.06778453290462494,0.9895474294310136,0.0023139170545268084,5.8504719734191895 -PPP2R1A,45,0.00372559973038733,0.04591447860002518,0.9949887094522177,-0.0017315724583113936,4.018301963806152 -PPP2R3C,5,0.03189975395798683,0.13783612847328186,0.9581818088622126,-0.07184239294706585,11.758127212524414 -PPP4C,17,0.007841953076422215,0.06680737435817719,0.9895237536284326,0.05304184564870651,5.829839706420898 -PPP4R2,18,0.00921291671693325,0.07188767194747925,0.9877554622000558,0.09389570808362915,6.318922519683838 -PPP6C,12,0.010356823913753033,0.07896707952022552,0.9869764945703794,-0.09496226499393619,6.699737071990967 -PPRC1,12,0.01978079229593277,0.10124792903661728,0.973258076201318,0.03956906649411588,9.25904655456543 -PPWD1,3,0.03982309252023697,0.1553751677274704,0.9506606244898669,-0.01611113132851143,13.13747501373291 -PPY,1,0.1678926646709442,0.3226591646671295,0.8220878669321996,-0.02167277029669109,26.974929809570312 -PRC1,18,0.007031848654150963,0.06417114287614822,0.9907351548603622,0.024293552679328034,5.520510196685791 -PRDM10,2,0.06447234004735947,0.19784674048423767,0.9219337320403137,0.045730021716168676,16.715953826904297 -PREB,4,0.04404037073254585,0.16046679019927979,0.9412854809393938,0.15043593905770192,13.815606117248535 -PRELID1,15,0.010605337098240852,0.07792437076568604,0.9858145266331787,0.07641806455795952,6.779641151428223 -PRELID3B,28,0.008763118647038937,0.0679863765835762,0.9886094016932845,0.24864459385856247,6.1627397537231445 -PRIM1,18,0.012790889479219913,0.08125491440296173,0.9827205058213055,0.1770813938612628,7.445516109466553 -PRIM2,15,0.019961636513471603,0.1022954136133194,0.9727800287161067,0.1872690811124683,9.301275253295898 -PRKRA,8,0.014179639518260956,0.09163960814476013,0.9818347313718309,0.04324978065543674,7.839295864105225 -PRKRIP1,18,0.007964706048369408,0.06818001717329025,0.9897398642433473,-0.04092632117716238,5.875290393829346 -PRMT1,6,0.01927437074482441,0.10733506828546524,0.9761116375232135,0.003205345378718419,9.139755249023438 -PRMT5,12,0.010869442485272884,0.08021534979343414,0.9857848413116252,0.07230869448980988,6.863539218902588 -PRODH,38,0.0038105854764580727,0.047808751463890076,0.995074929066032,-0.023651769148339614,4.0638747215271 -PRORP,4,0.026365673169493675,0.1255650371313095,0.9682507608363207,0.07518142815619351,10.68966007232666 -PRPF18,27,0.011604461818933487,0.07941614836454391,0.9848501151863954,0.16390604131908917,7.091807842254639 -PRPF19,11,0.018663646653294563,0.10248719900846481,0.9758307239571665,0.21396455072284268,8.993789672851562 -PRPF3,15,0.011254919692873955,0.07962125539779663,0.984975086487074,0.1700239617270244,6.9841837882995605 -PRPF31,2,0.05174568295478821,0.17927944660186768,0.9372760876791755,0.046052244922820215,14.975506782531738 -PRPF38A,2,0.05016367509961128,0.17615900933742523,0.9431704981032956,0.011974481387653332,14.744807243347168 -PRPF38B,3,0.059341076761484146,0.1908675581216812,0.9236788464307568,0.1269649015727723,16.036964416503906 -PRPF4,11,0.028329523280262947,0.12304897606372833,0.9628189296453693,0.26388712337125547,11.080619812011719 -PRPF40A,12,0.026543086394667625,0.11997037380933762,0.9674303433151777,0.15143823954830216,10.725564956665039 -PRPF4B,21,0.007932523265480995,0.06708269566297531,0.9898662901544861,0.02406770702455601,5.863408088684082 -PRPF6,23,0.008423269726336002,0.06914570927619934,0.9887256813850396,0.18345206593941965,6.042057037353516 -PRPF8,15,0.012805980630218983,0.08537840843200684,0.982777880093158,0.1787994331068519,7.4499077796936035 -PRRC2A,9,0.013933599926531315,0.09076381474733353,0.9823472419722008,0.06174521629977193,7.7709856033325195 -PSAT1,7,0.019680753350257874,0.10741625726222992,0.9735904723884035,0.09689817110879165,9.235604286193848 -PSMA1,17,0.009943169541656971,0.07554842531681061,0.986792090425984,0.09034784282112895,6.564579010009766 -PSMA2,7,0.019992221146821976,0.10558221489191055,0.9730541463710194,0.09352732023706242,9.308398246765137 -PSMA3,37,0.00927810464054346,0.06772680580615997,0.98759921496689,0.1799574492687263,6.341238021850586 -PSMA4,39,0.01374234538525343,0.08051308989524841,0.9816141958134963,0.16334895105399405,7.71746826171875 -PSMA5,3,0.0693141520023346,0.19925053417682648,0.9104786494671134,0.1378539482264696,17.33226776123047 -PSMA6,4,0.038793522864580154,0.1508195549249649,0.9495807521272093,0.060998329272967035,12.966538429260254 -PSMA7,4,0.04144638404250145,0.15506650507450104,0.947629272688395,0.0935758706728891,13.402560234069824 -PSMB1,15,0.009282218292355537,0.07164213061332703,0.9876167682627199,0.08930448654227029,6.342644214630127 -PSMB2,6,0.02729034796357155,0.12449822574853897,0.9636582409402982,0.030595791508284183,10.875494956970215 -PSMB3,3,0.062474124133586884,0.19129067659378052,0.9195253444236171,0.1329470914710053,16.454872131347656 -PSMB4,10,0.015045523643493652,0.09436856210231781,0.9803906874881992,-0.02299917314114909,8.075103759765625 -PSMB5,27,0.009426838718354702,0.06997400522232056,0.9872440803372412,0.11391984402137387,6.3918633460998535 -PSMB6,20,0.023668846115469933,0.1065012514591217,0.9678386620956282,0.19027909263007992,10.128216743469238 -PSMB7,41,0.005510534625500441,0.054694294929504395,0.9925608136957939,0.06144803444048911,4.886988639831543 -PSMC1,31,0.021204784512519836,0.09989631175994873,0.9716608377254122,0.2630274719032701,9.586528778076172 -PSMC2,43,0.010305861011147499,0.0722009614109993,0.9863371350887397,0.26186838371796684,6.683233261108398 -PSMC3,4,0.05105453357100487,0.17098656296730042,0.9357662365879365,0.06939472726212954,14.87515926361084 -PSMC4,24,0.025060506537556648,0.11143243312835693,0.9676350926284688,0.3026419997438863,10.421719551086426 -PSMC5,1,0.13620595633983612,0.29343146085739136,0.8534908638923201,-0.05587785382385805,24.296430587768555 -PSMC6,15,0.013835388235747814,0.08862364292144775,0.9812434377353932,0.1694861927561884,7.7435503005981445 -PSMD1,28,0.008101606741547585,0.06657003611326218,0.9891619300876857,0.23836196022276115,5.925568580627441 -PSMD11,8,0.020261893048882484,0.10798685252666473,0.9725386940181391,0.048737406519240405,9.370967864990234 -PSMD12,27,0.015165353193879128,0.08756724745035172,0.9796144279564906,0.20436937599972924,8.107195854187012 -PSMD13,12,0.021939875558018684,0.10885927081108093,0.97003884768047,0.16254496420451753,9.751277923583984 -PSMD14,1,0.2672516703605652,0.39921772480010986,0.6959392586499734,0.12171410434903598,34.033348083496094 -PSMD2,7,0.05720244720578194,0.17321264743804932,0.9221691796753148,0.2267566325540018,15.745329856872559 -PSMD3,7,0.023493831977248192,0.11764833331108093,0.9681678060655469,0.05196280949841638,10.090702056884766 -PSMD4,35,0.009974142536520958,0.07048960775136948,0.9865029756265438,0.20346984104372964,6.574795246124268 -PSMD6,55,0.01245587132871151,0.07472013682126999,0.9833567268248348,0.21920938559717565,7.347362995147705 -PSMD7,38,0.010436714626848698,0.0723736435174942,0.9862527679600005,0.2501913658822344,6.725527763366699 -PSMD8,2,0.061461035162210464,0.18892431259155273,0.9247485777592291,0.052291821753687205,16.320911407470703 -PSMD9,10,0.013939055614173412,0.08996108919382095,0.9813415634248789,0.002790129152058619,7.7725067138671875 -PSME1,12,0.012214472517371178,0.08359991759061813,0.9837333825785834,0.12840396231164808,7.27581787109375 -PSME2,15,0.009438800625503063,0.0736660361289978,0.9876923670766258,-0.07323437616705557,6.3959174156188965 -PSMG1,11,0.013113575056195259,0.08646905422210693,0.9833268879765104,-0.12268873734888222,7.538848400115967 -PSMG2,17,0.008305027149617672,0.0699470043182373,0.9890349047043784,0.004810393688042816,5.9994988441467285 -PSMG3,44,0.003435317659750581,0.044208429753780365,0.9954340453042293,0.05341270643867671,3.858583450317383 -PSMG4,6,0.03196277469396591,0.1345341056585312,0.9572540017148812,0.10590053282988411,11.769735336303711 -PSTK,38,0.003786061890423298,0.046958547085523605,0.9949964611876552,-0.028769751563342893,4.050776958465576 -PTBP1,15,0.01170316245406866,0.08059170097112656,0.9846390238398576,0.1377498520776237,7.121902942657471 -PTCD1,26,0.005580433644354343,0.056517258286476135,0.9926819172570606,0.15072314635489706,4.917885780334473 -PTCD3,7,0.02212325856089592,0.11494792252779007,0.9715672109473664,0.13153420464785867,9.791945457458496 -PTK2,15,0.008114878088235855,0.06885439157485962,0.9895568589667011,0.04877482387338047,5.930419921875 -PTMA,15,0.010892299935221672,0.07614792138338089,0.9854239452619071,0.01582508218939873,6.870751857757568 -PTPMT1,19,0.007099086884409189,0.0641685351729393,0.9906832100797609,0.00037831652683537634,5.546840667724609 -PTPN23,14,0.008773010224103928,0.07218985259532928,0.988390612524755,0.09710266442261448,6.166215896606445 -PUF60,9,0.01767655462026596,0.10121685266494751,0.9774530850739165,0.00706624284221086,8.752724647521973 -PWP1,7,0.01798846758902073,0.10361846536397934,0.9774319638656283,0.07861543807745708,8.829609870910645 -PWP2,5,0.031155884265899658,0.13574853539466858,0.9594083142775414,0.07446139983241536,11.620223999023438 -PXN,16,0.008164376020431519,0.06816364079713821,0.9895040821409091,-0.026828374286924875,5.948479175567627 -PYROXD1,10,0.011975627392530441,0.08524899929761887,0.9847077482697139,0.06150872478611021,7.204329967498779 -QARS,40,0.015502109192311764,0.08671194314956665,0.9805830711652803,0.4100757368014307,8.196715354919434 -QRSL1,12,0.01108450535684824,0.080377958714962,0.9855740551020207,0.024070552217767018,6.9311065673828125 -RAB18,5,0.022805655375123024,0.11829078197479248,0.9728332044257602,0.025963440205452583,9.941816329956055 -RAB4A,31,0.0046396320685744286,0.051514919847249985,0.9938080642078595,0.037665077557408756,4.484212875366211 -RAB6A,6,0.023182742297649384,0.11724318563938141,0.9702737355977652,-0.07005600767226572,10.023673057556152 -RABGGTA,11,0.01192146260291338,0.08232825994491577,0.984660634114737,-0.03502270286211682,7.188019275665283 -RABGGTB,17,0.00942014716565609,0.0738537609577179,0.9874490394775921,0.08687459377623513,6.389594554901123 -RAC1,20,0.007568646222352982,0.06592664122581482,0.9904454321286525,-0.026819319723038488,5.7273478507995605 -RAC3,22,0.007529749535024166,0.06622185558080673,0.9905161140428919,0.07277343735654668,5.712611675262451 -RACGAP1,17,0.011309679597616196,0.07921213656663895,0.9850145571318212,0.22076351044180118,7.001153469085693 -RAD17,13,0.011504669673740864,0.08153700083494186,0.9847676917440695,0.009723829020353052,7.061248779296875 -RAD21,11,0.01650107465684414,0.09666057676076889,0.9779956047521715,0.12746968890866434,8.456693649291992 -RAD51,24,0.00578902056440711,0.05583707615733147,0.9923939493921116,0.20200520818842882,5.00895357131958 -RAD51C,7,0.0201839879155159,0.10700809210538864,0.9734433984701685,0.13988382181655054,9.352934837341309 -RAD51D,14,0.009424928575754166,0.07438139617443085,0.9878476351008573,0.005126929421578258,6.391215801239014 -RAD9A,4,0.029480498284101486,0.13359671831130981,0.9631386233917206,0.036919521383382205,11.303472518920898 -RAE1,6,0.03317569941282272,0.13827402889728546,0.955304229942338,0.10153885090611581,11.990974426269531 -RAN,20,0.006758503615856171,0.06298855692148209,0.9911700390303358,0.02149557286748949,5.412148952484131 -RANBP2,9,0.015183452516794205,0.09524177014827728,0.9802672833843137,-0.026179870395327053,8.112032890319824 -RANGAP1,13,0.017718063667416573,0.09912550449371338,0.9778424573278722,0.08580130555223474,8.762995719909668 -RARS,41,0.009084882214665413,0.06652223318815231,0.9883935256087,0.28613661942220925,6.274860858917236 -RARS2,15,0.009796648286283016,0.07565073668956757,0.9872283777580011,0.17077213165280877,6.516032695770264 -RBBP4,9,0.01615927927196026,0.09590976685285568,0.9788878899565697,0.0540251715345705,8.368650436401367 -RBBP5,3,0.06331545114517212,0.18650446832180023,0.9177218095145615,0.24027804538539987,16.5653018951416 -RBBP6,8,0.03162338212132454,0.13206717371940613,0.9581731184200697,0.1717439000872653,11.707080841064453 -RBBP8,11,0.010965964756906033,0.07859569042921066,0.9856876750054244,0.09394032835599056,6.893945693969727 -RBM10,13,0.009621953591704369,0.07576602697372437,0.9872007043374645,0.08290807051276695,6.4576735496521 -RBM14,23,0.011230322532355785,0.07317550480365753,0.9851150779230445,0.08460485852004897,6.976547718048096 -RBM17,7,0.020505229011178017,0.10903385281562805,0.9748654085789672,-0.026409959263769392,9.427070617675781 -RBM19,5,0.03349992260336876,0.14073076844215393,0.9554203643773177,-0.003466542081128443,12.04942512512207 -RBM22,1,0.14816419780254364,0.2942613959312439,0.8475651090420953,0.20926854230683015,25.340553283691406 -RBM25,7,0.031840093433856964,0.13447923958301544,0.959137157996239,0.23810396428257966,11.747125625610352 -RBM28,7,0.022081831470131874,0.11183162778615952,0.9733352615079043,0.17758748218472617,9.782773971557617 -RBM39,8,0.02170482836663723,0.1111147403717041,0.971779625285052,0.10946546896261623,9.69890308380127 -RBM4,22,0.0066695306450128555,0.061852604150772095,0.9911634624021296,0.0667334711857309,5.376406192779541 -RBM42,8,0.023149661719799042,0.1159263625741005,0.9713045868626679,-0.029855452069085523,10.016518592834473 -RBM48,5,0.02718076854944229,0.12642934918403625,0.9659082949817596,0.10068353564636862,10.8536376953125 -RBM8A,4,0.043283574283123016,0.15728549659252167,0.9425910449457171,0.16225377107120045,13.696386337280273 -RBMX,14,0.011096598580479622,0.07819678634405136,0.9851319719256355,0.0878183127606386,6.934886932373047 -RBMX2,25,0.0166062843054533,0.09254150092601776,0.9790089365939693,0.3137939662300738,8.483609199523926 -RBMXL1,6,0.02053072117269039,0.11135663092136383,0.9745737456852854,-0.006205938039234941,9.432928085327148 -RBX1,5,0.042430318892002106,0.1557929366827011,0.9442268414662216,0.16378599872550673,13.560714721679688 -RCC1,22,0.007311453577131033,0.06497073173522949,0.9904063442209621,0.17039022066316767,5.629195213317871 -RCL1,10,0.016958387568593025,0.09818998724222183,0.9782813383544662,0.13927567301137908,8.573077201843262 -RCOR1,3,0.047545429319143295,0.17244625091552734,0.9418142146457964,-0.029894969376317766,14.35485553741455 -REV3L,10,0.01359247975051403,0.08957815915346146,0.9819690572155846,0.11161171854097464,7.675272464752197 -REXO2,8,0.0209124106913805,0.10874155163764954,0.9743579267536978,0.0243203487184514,9.520209312438965 -RFC1,15,0.0114730354398489,0.07975535839796066,0.9845740550935219,0.10423361242046367,7.051534175872803 -RFC2,9,0.015031477436423302,0.09298176318407059,0.9802785540476256,0.10406496911261998,8.071333885192871 -RFC3,23,0.00754040852189064,0.06398015469312668,0.989884185499549,0.1538550906543411,5.716653347015381 -RFC4,7,0.02055056020617485,0.10840650647878647,0.9726164234827208,0.06493240713461758,9.437484741210938 -RFC5,11,0.01149750966578722,0.08220486342906952,0.9851821737321299,0.08891550756275456,7.059051990509033 -RFFL,1,0.10148166865110397,0.252122163772583,0.8834022084423088,-0.06966771634729767,20.971923828125 -RFT1,12,0.011079899966716766,0.08082416653633118,0.9855506565517732,0.007283750275794808,6.929666996002197 -RGP1,13,0.00981674250215292,0.07607895880937576,0.9872221165249694,-0.0639585448489531,6.522711277008057 -RGPD6,20,0.006438101641833782,0.06149160861968994,0.9915610081901978,-0.022045766729505485,5.282303333282471 -RHOQ,19,0.006744125857949257,0.06266561150550842,0.9910597885659614,0.07410597762660803,5.406389236450195 -RHPN1,23,0.005725938826799393,0.058052532374858856,0.9927782626732763,-0.07996350126131406,4.981587886810303 -RINT1,15,0.010078679770231247,0.07500061392784119,0.9868433329699189,0.04739524968395098,6.609160423278809 -RIOK1,32,0.011136069893836975,0.07559531182050705,0.9858582676597282,0.16712916637343322,6.94720983505249 -RIOK2,18,0.012956403195858002,0.08352474868297577,0.9825758938331466,0.210170453419568,7.493533611297607 -RMI1,21,0.0072677466087043285,0.06474889814853668,0.9902921980623927,0.04370941650424239,5.612344741821289 -RNASEH2C,10,0.014702853746712208,0.09334595501422882,0.9809839385154436,-0.05151674899647694,7.982616901397705 -RNF103,11,0.015563836321234703,0.09554516524076462,0.9788361453355372,0.08141346464501784,8.213018417358398 -RNF113A,17,0.015470092184841633,0.09208792448043823,0.9809396230620911,0.1709158549670667,8.188246726989746 -RNF123,11,0.011440573260188103,0.0815252885222435,0.9851390985494655,-0.07001639739343035,7.04155158996582 -RNF14,24,0.005434280727058649,0.05658259242773056,0.9928683323060379,-0.016065168864783257,4.853058338165283 -RNF168,6,0.020886903628706932,0.11104544997215271,0.9739151547618844,-0.03754669572669282,9.51440143585205 -RNF20,13,0.01609221287071705,0.09279590100049973,0.9781974869052437,0.3035809471522292,8.351266860961914 -RNF4,3,0.05101054161787033,0.17077764868736267,0.9350733911659745,0.11900564119718655,14.868748664855957 -RNF40,20,0.007864650338888168,0.0671483650803566,0.9894634886643102,0.11115224484734543,5.83827018737793 -RNF8,17,0.008531001396477222,0.07083965092897415,0.9888113051615887,-0.009625529631419572,6.080572605133057 -RNGTT,11,0.012105586007237434,0.08358316868543625,0.9842858191143868,0.02568107629858297,7.24331521987915 -RNMT,12,0.01336990762501955,0.0868874117732048,0.9825860454399389,0.06089216312154097,7.612173080444336 -RNPC3,8,0.01739213615655899,0.10174263268709183,0.9770977752163276,0.17122273017014353,8.682023048400879 -RNPS1,13,0.012457986362278461,0.08509203791618347,0.9834700938986497,0.06936547091505248,7.347987174987793 -ROMO1,8,0.024624977260828018,0.11602712422609329,0.968494973628775,0.22025159281462223,10.33076286315918 -RPA1,47,0.004859255161136389,0.051319289952516556,0.9935469233590083,0.15378419657826004,4.589118957519531 -RPA2,23,0.015216737985610962,0.08888569474220276,0.9798154219803163,0.2612807069503096,8.120920181274414 -RPA3,21,0.012317418120801449,0.08232123404741287,0.9841941672343268,0.1823127549058644,7.3064141273498535 -RPAIN,11,0.01186484657227993,0.08365751802921295,0.9844392513817798,0.05211605884033374,7.1709303855896 -RPAP1,9,0.024961629882454872,0.11899669468402863,0.9671493910432927,0.10201981826926357,10.401140213012695 -RPAP2,40,0.01803075522184372,0.09462413936853409,0.977281575003129,0.20913636152763052,8.839982986450195 -RPAP3,2,0.09079093486070633,0.23431165516376495,0.8792350826865029,0.10956637700498155,19.836530685424805 -RPE,12,0.013227956369519234,0.0878959372639656,0.9830773622337828,0.04103066928837798,7.5716552734375 -RPF1,13,0.010927313007414341,0.0802205428481102,0.9854996452970012,0.12549868719399734,6.881785869598389 -RPF2,7,0.020190401002764702,0.10845973342657089,0.9747687380792855,0.10344890863704298,9.354421615600586 -RPL10,27,0.01558900810778141,0.086457259953022,0.9808005759047235,0.40063507047419783,8.219655990600586 -RPL10A,25,0.010699179023504257,0.07358168065547943,0.9856000008881954,0.34797082445207783,6.809569835662842 -RPL11,38,0.010881701484322548,0.07145842164754868,0.985783005016132,0.4157658433254905,6.867407321929932 -RPL12,25,0.008424244821071625,0.06644511967897415,0.9886032760080838,0.3932564808155566,6.0424065589904785 -RPL13,33,0.00757765956223011,0.06279114633798599,0.9899152807701465,0.3556876753151833,5.730756759643555 -RPL13A,16,0.008116202428936958,0.06822787970304489,0.9892521671440848,0.05465141421467666,5.930903911590576 -RPL14,22,0.02030893787741661,0.09803677350282669,0.9727573200403353,0.4763007892240244,9.381841659545898 -RPL15,17,0.021637413650751114,0.10605350136756897,0.970582654172888,0.41918787893427456,9.683829307556152 -RPL17,15,0.02969767339527607,0.12150074541568756,0.9629023568639271,0.407606852186661,11.345030784606934 -RPL18,31,0.014639090746641159,0.08562923967838287,0.9822046028168692,0.4371050137815493,7.9652886390686035 -RPL18A,7,0.018975170329213142,0.10479092597961426,0.9761573232436729,0.05185359808260719,9.068537712097168 -RPL19,24,0.01621381565928459,0.08917557448148727,0.9792678488596179,0.40985478862778557,8.382760047912598 -RPL21,26,0.011091751046478748,0.07655154913663864,0.9859317722686959,0.4262972172611874,6.933372497558594 -RPL23,40,0.015161839313805103,0.08440300077199936,0.9811954901829849,0.4274479604213322,8.106257438659668 -RPL23A,10,0.03362765908241272,0.13065899908542633,0.9570941202889448,0.3943550634867413,12.072376251220703 -RPL24,46,0.011738060973584652,0.07279651612043381,0.9849862243624584,0.4145419797352702,7.132513999938965 -RPL26,21,0.016596073284745216,0.08980530500411987,0.9794189689226408,0.37952594275320817,8.481000900268555 -RPL26L1,15,0.008482245728373528,0.0703481137752533,0.9890373259560981,-0.016513320536905504,6.063172340393066 -RPL27,4,0.04732588678598404,0.1593390554189682,0.9433053589412574,0.29715403709404703,14.321676254272461 -RPL27A,20,0.02016603574156761,0.10064724087715149,0.9741238840285111,0.42095728913570324,9.348774909973145 -RPL28,5,0.025777742266654968,0.12048444151878357,0.9666760007785339,0.029084440230904115,10.569803237915039 -RPL29,9,0.01807691529393196,0.09908953309059143,0.9762307067966262,0.013628428678776585,8.851289749145508 -RPL3,199,0.001132727600634098,0.024792715907096863,0.9985033372508451,0.15955961281126554,2.2156805992126465 -RPL30,37,0.01127543207257986,0.07338028401136398,0.9857734794949354,0.397564466087842,6.990545272827148 -RPL31,30,0.019975174218416214,0.09905964881181717,0.9746248264873982,0.4949294125083328,9.304429054260254 -RPL32,21,0.009482992812991142,0.06905708461999893,0.9879524704811166,0.2481671059639253,6.410872936248779 -RPL34,33,0.012378571555018425,0.07768357545137405,0.9837180170213741,0.44633426183177954,7.32452917098999 -RPL35,4,0.03266127035021782,0.13848243653774261,0.9580203966658533,0.22132402506029267,11.89764404296875 -RPL35A,13,0.022651242092251778,0.10821833461523056,0.9700490215061109,0.38173424821213187,9.908102035522461 -RPL36,21,0.01586862839758396,0.08927973359823227,0.9793226942497785,0.46311168442553746,8.293046951293945 -RPL36A,9,0.01542435772716999,0.09220843017101288,0.9798456933159966,0.25801830727437175,8.17613410949707 -RPL36AL,15,0.008680928498506546,0.0704004317522049,0.9886834665203593,-0.05506446103757688,6.133770942687988 -RPL37,38,0.0067524961195886135,0.05887856334447861,0.9913125010156718,0.3281900078415027,5.409742832183838 -RPL37A,38,0.013312902301549911,0.07975999265909195,0.9829707337440685,0.4773282618928495,7.595928192138672 -RPL38,19,0.015516195446252823,0.08877091854810715,0.9801529840818828,0.3594815740486639,8.200438499450684 -RPL39,3,0.04824761673808098,0.16934773325920105,0.9396105355700197,0.10804663686812249,14.460468292236328 -RPL4,35,0.012564141303300858,0.0788293108344078,0.9838131875941744,0.47006239264254074,7.3792266845703125 -RPL41,4,0.02960636094212532,0.1305631548166275,0.964027226532695,-0.08363630958161519,11.32757568359375 -RPL5,20,0.012810482643544674,0.08059162646532059,0.9830242733201912,0.4093536972327398,7.451216697692871 -RPL6,28,0.02052372135221958,0.09880608320236206,0.9739164223752635,0.46771039610371584,9.431320190429688 -RPL7,30,0.021232619881629944,0.09941038489341736,0.9746822548055003,0.409842231813922,9.592818260192871 -RPL7A,1,0.2680635452270508,0.39162516593933105,0.7284969119962083,0.24469151150303697,34.08500289916992 -RPL7L1,3,0.041914477944374084,0.15868450701236725,0.9499819918918404,0.15653350570592092,13.478033065795898 -RPL8,39,0.01186749991029501,0.0747583657503128,0.9846860371360305,0.45937625545608707,7.171732425689697 -RPL9,24,0.01138746365904808,0.07794442772865295,0.9855205328642342,0.40865680617713546,7.025187969207764 -RPLP0,19,0.015807928517460823,0.08911588788032532,0.980329368582855,0.2917873855866764,8.27717113494873 -RPLP1,11,0.01392529346048832,0.08818145096302032,0.9813259915442311,0.14271700895131134,7.768669128417969 -RPLP2,22,0.01288837194442749,0.08032633364200592,0.9837104127295436,0.1577276459688467,7.47383451461792 -RPN1,5,0.02707013674080372,0.12491396814584732,0.9659158713031489,0.05692494784872612,10.831527709960938 -RPN2,11,0.011779768392443657,0.08283329755067825,0.9847138301497054,-0.02224249767210938,7.145174503326416 -RPP14,5,0.027803372591733932,0.12605032324790955,0.9630628455804084,0.18699209488261195,10.977241516113281 -RPP21,6,0.019332805648446083,0.10817351937294006,0.9762878069680011,0.0073988819907593395,9.15359878540039 -RPP30,10,0.02002401277422905,0.10598773509263992,0.9733584169340623,0.1955452377703446,9.315797805786133 -RPP38,5,0.032757606357336044,0.14038604497909546,0.956331301108735,0.06784906566303406,11.915178298950195 -RPP40,8,0.02544141747057438,0.12076055258512497,0.9660722542585434,0.05592664235214117,10.50062370300293 -RPS10,10,0.021067356690764427,0.10757876187562943,0.9729316787205814,0.15143022034232667,9.555413246154785 -RPS10-NUDT3,17,0.01989499107003212,0.10138648003339767,0.9743631394524702,0.2507628393515495,9.285736083984375 -RPS11,27,0.013992257416248322,0.0832071304321289,0.982265408042621,0.20610067779106328,7.787325859069824 -RPS12,49,0.01563742756843567,0.08404655009508133,0.9811222404640714,0.21416216495780258,8.232412338256836 -RPS13,13,0.028590073809027672,0.12152276933193207,0.9626312467662909,0.21966542206264067,11.131458282470703 -RPS14,18,0.015711842104792595,0.08956998586654663,0.9791598817804794,0.2166775833074873,8.25197696685791 -RPS15,2,0.09531071782112122,0.2425345629453659,0.8839079952883164,0.10047311953655934,20.32428741455078 -RPS15A,32,0.013190791942179203,0.08005841076374054,0.9834729603205273,0.19295451170792682,7.56101131439209 -RPS16,5,0.03893722593784332,0.1483744978904724,0.9498577493638496,0.15986072504283497,12.990532875061035 -RPS17,23,0.010761935263872147,0.0757761150598526,0.9857039363887932,0.23721958310448898,6.829511642456055 -RPS18,12,0.025175610557198524,0.11385160684585571,0.9680239314828866,0.19547090593982056,10.445626258850098 -RPS19,42,0.012064081616699696,0.07539527863264084,0.98496280651119,0.28655591601287816,7.230886936187744 -RPS19BP1,6,0.03260454535484314,0.13516399264335632,0.9584433216210924,0.05531879265532799,11.887308120727539 -RPS2,6,0.02356129139661789,0.1180303692817688,0.9688030444487874,0.056692523403347496,10.105178833007812 -RPS20,28,0.00890179444104433,0.06951931118965149,0.9891587872456191,0.24513300121709977,6.211310386657715 -RPS21,7,0.03348842263221741,0.13529831171035767,0.9555246898056644,0.16061603195393298,12.047357559204102 -RPS23,12,0.022704271599650383,0.11040160804986954,0.9711212771963615,0.25564100291719255,9.919692993164062 -RPS24,25,0.02038285881280899,0.09938652068376541,0.9737960426877722,0.3347202811904885,9.39889907836914 -RPS25,4,0.03281104564666748,0.137253537774086,0.9591842097143237,0.019720518815864667,11.924893379211426 -RPS26,8,0.03753185644745827,0.1424543708562851,0.9534111342409151,0.22162755073766355,12.75394344329834 -RPS27,18,0.015521945431828499,0.08966029435396194,0.9803914094974022,0.17377343093999928,8.201957702636719 -RPS27A,9,0.020700326189398766,0.1071309968829155,0.9724305829363461,0.23805183710459957,9.471811294555664 -RPS28,31,0.01104841660708189,0.07519280165433884,0.9865214178269499,0.2038540040204108,6.9198150634765625 -RPS29,41,0.006493756547570229,0.05883564054965973,0.9918720918860802,0.20358227040466714,5.305086135864258 -RPS3,27,0.02096366696059704,0.10152353346347809,0.974389161926752,0.3285416038445117,9.531869888305664 -RPS3A,16,0.01647331193089485,0.09240248054265976,0.9793321973899806,0.10874337136446614,8.449575424194336 -RPS4X,18,0.0248737595975399,0.11151523888111115,0.9694504782872558,0.2699573489901783,10.382816314697266 -RPS5,15,0.017759274691343307,0.09732303023338318,0.9780530375395383,0.2875261608029828,8.77318000793457 -RPS6,18,0.022032663226127625,0.1045379638671875,0.9719426879186029,0.28081798441262384,9.771875381469727 -RPS7,27,0.020796610042452812,0.09989479184150696,0.9732266338492697,0.3026907904158633,9.493813514709473 -RPS8,2,0.11985006183385849,0.26540401577949524,0.8447559830031734,0.14086831066603364,22.7910099029541 -RPS9,27,0.016979940235614777,0.09156985580921173,0.979366910007087,0.24168646214984524,8.578523635864258 -RPSA,23,0.010365910828113556,0.07440100610256195,0.9869426768863331,0.21941979172872977,6.7026753425598145 -RPTOR,8,0.02169029787182808,0.11170542240142822,0.9716414578937814,0.01530309624644979,9.695656776428223 -RPUSD3,17,0.00810879934579134,0.06882122159004211,0.9893918323372815,0.02442612667994265,5.928198337554932 -RPUSD4,12,0.009848450310528278,0.07643365859985352,0.9876048877329611,-0.03185418934192305,6.533236503601074 -RRM1,11,0.01685725525021553,0.09490269422531128,0.9772936196983952,0.1292564340566283,8.547475814819336 -RRM2,57,0.003293490270152688,0.04304635897278786,0.9956255919702982,0.18997149059734872,3.778092861175537 -RRN3,4,0.054969556629657745,0.17361053824424744,0.9297699365917336,0.18020005599493058,15.434961318969727 -RRP1,11,0.012585079297423363,0.08605244755744934,0.9839277073089171,0.13518041763314337,7.385373115539551 -RRP12,21,0.01512975338846445,0.08905155956745148,0.9805201309480076,0.14282065474385797,8.097675323486328 -RRP15,3,0.044802237302064896,0.16571402549743652,0.9425847335566795,0.11307456691702343,13.934593200683594 -RRP7A,4,0.045099515467882156,0.16176486015319824,0.9414980200317612,0.13914066556340757,13.98074722290039 -RRP9,11,0.013868287205696106,0.09023286402225494,0.9812627420965663,0.12512422289786113,7.752751350402832 -RRS1,7,0.024276049807667732,0.11850021779537201,0.9677185872688534,0.20677505166114707,10.257309913635254 -RSL1D1,21,0.006925003603100777,0.06287799775600433,0.9909638276960724,0.10371059810093294,5.478409290313721 -RSL24D1,18,0.01132877729833126,0.07889757305383682,0.9848832947194923,0.29196496684586337,7.0070624351501465 -RSRC2,10,0.014597535133361816,0.09264999628067017,0.9804308287771384,0.1357669770443627,7.953974723815918 -RTCB,17,0.008578619919717312,0.07077927887439728,0.9888603388991491,0.02149014239711382,6.097519397735596 -RTEL1,14,0.009101281873881817,0.07352153956890106,0.9879634073602722,0.04143666462296405,6.280521869659424 -RTF1,6,0.04579171538352966,0.1578749716281891,0.9448475937277505,0.184759595124633,14.087628364562988 -RTTN,13,0.010438987985253334,0.07932871580123901,0.9864016320112117,-0.0023547069972071895,6.726259708404541 -RUVBL1,21,0.020438510924577713,0.102946437895298,0.9728778870669516,0.3110437905591691,9.411722183227539 -RUVBL2,2,0.06989631801843643,0.20655521750450134,0.9154284864751068,-0.0007561243642863973,17.404903411865234 -SACM1L,20,0.007229661103338003,0.0654200091958046,0.9902258409534678,0.047860947979803226,5.597620010375977 -SAE1,12,0.01452151034027338,0.08671408891677856,0.9805861997630104,0.07224661447448234,7.9332356452941895 -SAMM50,10,0.020957348868250847,0.10759151726961136,0.9718998514546137,0.31718960220558495,9.53043270111084 -SAP130,12,0.012282371520996094,0.08490390330553055,0.9835960162459019,-0.02633980917884692,7.2960124015808105 -SAP18,5,0.033019550144672394,0.1401723474264145,0.9565991603834435,0.1010130421599674,11.962723731994629 -SAP30BP,4,0.05163859575986862,0.17536424100399017,0.9350927095257372,0.1490152515491533,14.960001945495605 -SARNP,10,0.013337572105228901,0.08849357813596725,0.9828169946126502,-0.07271459742760122,7.602962493896484 -SARS,3,0.07808785885572433,0.21460884809494019,0.9050077423343845,0.17229592193396195,18.396541595458984 -SARS2,17,0.009000828489661217,0.07190059125423431,0.9881649313924954,-0.0275430192676336,6.245765686035156 -SART1,30,0.012396566569805145,0.08146613091230392,0.9851271790195869,0.284189089336762,7.329851150512695 -SART3,16,0.01581176370382309,0.0932055413722992,0.9805785684879756,0.12411290082758032,8.278175354003906 -SASS6,7,0.01728745549917221,0.101926788687706,0.9770753048292761,0.06979723015773516,8.655855178833008 -SBDS,9,0.021682189777493477,0.11230111867189407,0.9722033626803394,0.0024975629794003725,9.693843841552734 -SBNO1,19,0.00829684641212225,0.06985105574131012,0.9889092261526274,-0.07220886132777067,5.9965434074401855 -SCAF1,18,0.011725543066859245,0.081093430519104,0.9844640635762008,0.2504433370830934,7.12870979309082 -SCAP,14,0.009716875851154327,0.07528814673423767,0.9872835924263623,-0.04814905471578206,6.489448547363281 -SCD,3,0.04468098655343056,0.16523930430412292,0.9432739957174988,0.022415587760361247,13.915725708007812 -SCFD1,15,0.009732607752084732,0.07517918199300766,0.9868706463691886,0.0594082806246893,6.494699478149414 -SCNM1,10,0.013040280900895596,0.0881877914071083,0.9831880957484778,-0.09893991511707631,7.5177507400512695 -SDAD1,24,0.009015274234116077,0.06925072520971298,0.9883763369590819,0.31030610223607147,6.250775337219238 -SDC1,17,0.007561325095593929,0.06631799042224884,0.9899333619470522,0.045178495662281756,5.724576950073242 -SDE2,15,0.00903086457401514,0.07259742170572281,0.9883240372559869,0.11012407981525289,6.256178379058838 -SDHA,6,0.024324337020516396,0.1186990737915039,0.9694964559599266,-0.010758743119092423,10.267505645751953 -SDHAF2,12,0.011874820105731487,0.08255708962678909,0.9848023077471891,-0.18255465751349245,7.173943519592285 -SDHC,11,0.012806073762476444,0.08744607120752335,0.9831744198206338,0.012958466709990454,7.449934482574463 -SDHD,5,0.024242490530014038,0.12071869522333145,0.9691019653548116,-0.03972223548609251,10.25021743774414 -SEC13,5,0.026382209733128548,0.12527067959308624,0.9664065182487284,-0.003135237456790068,10.693011283874512 -SEC16A,16,0.008981716819107533,0.07290925085544586,0.9881916712778324,-0.12770471696273303,6.239131450653076 -SEC61A1,18,0.010474711656570435,0.0774896889925003,0.9867035914805353,0.21105590885667216,6.737759113311768 -SEC61B,9,0.014830192551016808,0.0932997390627861,0.9808442810003427,-0.027686912852869422,8.017109870910645 -SEC61G,10,0.020426850765943527,0.1073363721370697,0.9737363839308798,0.19735177193146922,9.409036636352539 -SEC62,1,0.08518067002296448,0.2329595386981964,0.9056646847805861,-0.10517311307558694,19.213876724243164 -SEC63,8,0.017553938552737236,0.10139753669500351,0.9768586346528235,0.05365253103423459,8.72231388092041 -SEH1L,6,0.031541332602500916,0.13719503581523895,0.9611393410537968,0.1304293309045726,11.691883087158203 -SEM1,9,0.016587449237704277,0.09476665407419205,0.9794442696256356,0.03336390404813546,8.47879695892334 -SENP6,10,0.012266688048839569,0.08510453999042511,0.9844981077740871,0.1261209110105147,7.29135274887085 -SEPHS2,16,0.008277579210698605,0.06965071707963943,0.9891381570808457,-0.08010277586102496,5.989576816558838 -SEPSECS,8,0.01461070217192173,0.09327027946710587,0.9812970478919782,0.021053547883891837,7.957561492919922 -SERBP1,11,0.013352910988032818,0.08715160936117172,0.9824648681360443,-0.13133684265903325,7.607333183288574 -SET,1,0.06741299480199814,0.20840300619602203,0.9251982536061293,-0.046246687368057114,17.092920303344727 -SETD1A,7,0.03927898406982422,0.14835026860237122,0.9468474538738403,0.2908791596390159,13.047417640686035 -SETD2,11,0.011453819461166859,0.0824689194560051,0.9850044368418128,-0.03637686204061295,7.045626640319824 -SETDB1,13,0.011787746101617813,0.08189085870981216,0.9850536809412701,0.07905043708874653,7.1475934982299805 -SF1,4,0.03940770402550697,0.15337233245372772,0.9506094241521368,0.11293293251934348,13.068778991699219 -SF3A1,33,0.004405063111335039,0.05072519928216934,0.9943095026782997,0.06727809288039392,4.369387149810791 -SF3A2,32,0.009683387354016304,0.07086305320262909,0.9870177544394807,0.32080264583713564,6.478255748748779 -SF3A3,3,0.07093189656734467,0.20443502068519592,0.9104595197798159,0.2049134099448259,17.533363342285156 -SF3B1,17,0.006935217417776585,0.0641954243183136,0.990800780777704,0.05601988475128555,5.482447624206543 -SF3B2,50,0.006244901567697525,0.057547327131032944,0.992365032295203,0.2601324432189278,5.202442169189453 -SF3B3,38,0.012035875581204891,0.07954079657793045,0.9853776299631305,0.2433213213354639,7.222429275512695 -SF3B4,24,0.0075290752574801445,0.06528463214635849,0.9900463132158317,0.25616503406860613,5.712356090545654 -SF3B5,9,0.015567168593406677,0.09545990824699402,0.9795249651432296,0.06307756918646198,8.213897705078125 -SFPQ,31,0.020535383373498917,0.0993407741189003,0.9755262273154868,0.2449288879471458,9.433999061584473 -SFSWAP,17,0.009538635611534119,0.07421417534351349,0.9876535352653905,0.19620395215021647,6.429653644561768 -SGPP1,16,0.008115840144455433,0.06952925026416779,0.9896866272426841,0.07868809187337426,5.930771350860596 -SHQ1,10,0.011495118960738182,0.08298572897911072,0.9852045804775302,-0.04342112802996456,7.0583176612854 -SIGLEC14,9,0.015099876560270786,0.0938384011387825,0.9809851631130913,-0.0015965996605485301,8.089675903320312 -SIN3A,30,0.0054461052641272545,0.055356040596961975,0.9926722779232681,0.09482959151736337,4.858335018157959 -SKA1,7,0.020334092900156975,0.10749272257089615,0.9736098342627676,0.018312577953571704,9.387649536132812 -SKA3,11,0.014414451085031033,0.09011627733707428,0.981072508802941,0.1567292207232378,7.903937816619873 -SKP1,6,0.03229455277323723,0.13481897115707397,0.9567963906746947,0.09991969818171699,11.830662727355957 -SKP2,20,0.006999149918556213,0.06388606131076813,0.9906997161226886,0.20302448988291574,5.507659912109375 -SLBP,9,0.019251324236392975,0.10109035670757294,0.9742913466285539,0.10966010405606738,9.134289741516113 -SLC16A5,18,0.008310722187161446,0.06985228508710861,0.9888498325269379,0.06023284606675591,6.001555442810059 -SLC1A5,49,0.0030668354593217373,0.041334230452775955,0.9959688005561471,-0.027938086373546632,3.6457736492156982 -SLC25A10,31,0.005509120877832174,0.0554472915828228,0.9925688506715636,0.20122016466403458,4.886361598968506 -SLC25A26,3,0.0361054353415966,0.14853861927986145,0.955564848175255,0.02542572281863694,12.509235382080078 -SLC25A3,12,0.011445971205830574,0.080375075340271,0.9851969787071923,-0.05602319934595502,7.043211936950684 -SLC35B1,7,0.018401749432086945,0.10532519966363907,0.9765464574230063,-0.01992178220824837,8.930463790893555 -SLC35G6,17,0.008068693801760674,0.06897951662540436,0.9894169083086097,-0.048861273940135334,5.913519859313965 -SLC39A10,16,0.00917034037411213,0.07269054651260376,0.9878669356734386,-0.07531796572412024,6.304304599761963 -SLC39A7,13,0.012650937773287296,0.08497889339923859,0.983037343867184,0.12954389818954581,7.404672145843506 -SLC39A9,75,0.00266893208026886,0.037946052849292755,0.9964456822831365,0.1441843330521765,3.4010515213012695 -SLC4A5,22,0.00619491608813405,0.060379330068826675,0.991670124640641,0.06376995565136495,5.18157958984375 -SLC7A5,38,0.00531197153031826,0.05402158573269844,0.9929178985402485,0.19319927124042408,4.798133373260498 -SLC7A6OS,4,0.06025425344705582,0.18620406091213226,0.9239402237811086,0.11488255893324063,16.159887313842773 -SLC9B1,8,0.015989435836672783,0.09771380573511124,0.9792856170517257,-0.022461861626339487,8.324555397033691 -SLU7,23,0.03312898054718971,0.13016758859157562,0.9593440179051655,0.32456222239772126,11.982529640197754 -SMAGP,11,0.012356721796095371,0.08620066940784454,0.9835067140866495,0.1401334467800757,7.318061828613281 -SMARCB1,18,0.007456584367901087,0.06593027710914612,0.9901040912323525,0.09258595503814622,5.684790134429932 -SMARCE1,3,0.05052812770009041,0.17280156910419464,0.9386409399142281,-0.12988737884542387,14.798274040222168 -SMC1A,18,0.02378709428012371,0.10981515794992447,0.9680620276899902,0.2682877462553463,10.153486251831055 -SMC2,14,0.010986976325511932,0.07900519669055939,0.9851092756922416,0.29313456969844665,6.900547504425049 -SMC3,6,0.04020458832383156,0.1499834656715393,0.9471899041198405,0.23628735776069315,13.2002534866333 -SMC4,34,0.004981956910341978,0.05180033668875694,0.9935852706776098,0.2350239140206714,4.646697521209717 -SMC5,12,0.010694527998566628,0.07797032594680786,0.9869261660674212,-0.18520996440258283,6.808089256286621 -SMC6,13,0.011241447180509567,0.08106786012649536,0.9850667526286925,0.09981603344992206,6.980002403259277 -SMG1,3,0.05173817649483681,0.1764957755804062,0.9351776525933948,0.13102920448571592,14.974419593811035 -SMG5,9,0.021297644823789597,0.10275685042142868,0.9720356707977313,0.027172837490022025,9.60749626159668 -SMG6,6,0.022533239796757698,0.11391134560108185,0.9712347792707107,0.05720477316571567,9.8822603225708 -SMN2,32,0.020085249096155167,0.09862776845693588,0.9750253634432207,0.2717211087805846,9.33003044128418 -SMNDC1,1,0.1286877989768982,0.28521794080734253,0.8600671068894284,0.08177935190671606,23.616369247436523 -SMU1,15,0.014886735938489437,0.08987615257501602,0.9799152685682644,0.31012329127586846,8.032379150390625 -SMUG1,7,0.021777570247650146,0.11198211461305618,0.9721292905049557,-0.006260115801787595,9.715142250061035 -SNAPC1,9,0.0219533983618021,0.10947166383266449,0.9710153839801683,0.08239688889813995,9.75428295135498 -SNAPC2,7,0.023388512432575226,0.11574879288673401,0.9689688602483231,0.09366328656878442,10.068058967590332 -SNAPC3,21,0.008178641088306904,0.0669843927025795,0.9890106262311662,-0.013528168275631355,5.953673839569092 -SNAPC4,11,0.013402853161096573,0.0880269706249237,0.9820314012401465,0.12180160039902507,7.621546268463135 -SNAPC5,10,0.0136335464194417,0.08973485976457596,0.9823471725769435,-0.03498194126611238,7.686858654022217 -SNF8,13,0.010554486885666847,0.0785985067486763,0.9859545348732973,-0.048427604506268415,6.763367652893066 -SNIP1,5,0.03084326535463333,0.1338367462158203,0.9597895757664272,0.21593100657511163,11.56177806854248 -SNRNP200,17,0.016023175790905952,0.09310808032751083,0.9787945093299963,0.2573298734581141,8.333333015441895 -SNRNP25,13,0.010905127972364426,0.07885874807834625,0.985551009887954,0.04903291428965256,6.874796390533447 -SNRNP27,4,0.04348452389240265,0.16130441427230835,0.9452683769226069,0.07554807690338311,13.728144645690918 -SNRNP35,17,0.012612840160727501,0.08354690670967102,0.9833375748040757,0.19286111775285927,7.393514156341553 -SNRNP40,21,0.006755898240953684,0.06283928453922272,0.9912333234405962,-0.003119475259944515,5.411105632781982 -SNRNP48,19,0.010278896428644657,0.0754765197634697,0.9863054653637161,0.15125092767645015,6.674483776092529 -SNRNP70,13,0.020905571058392525,0.10716819018125534,0.9716227924820922,0.29492076506220255,9.518651962280273 -SNRPA1,16,0.025260569527745247,0.11503517627716064,0.9706537988515677,0.2460537667785891,10.463235855102539 -SNRPB,27,0.011556390672922134,0.07863719761371613,0.9860306150601164,0.23482457789881236,7.077103614807129 -SNRPC,4,0.056135717779397964,0.17897430062294006,0.9246095884523557,0.22424350559251116,15.597826957702637 -SNRPD1,14,0.014614928513765335,0.09072078764438629,0.981543557694127,0.20591768877594463,7.958712100982666 -SNRPD2,2,0.09711749106645584,0.23920613527297974,0.8887008765176573,0.23413275130943798,20.516023635864258 -SNRPD3,18,0.027832400053739548,0.11947157233953476,0.9648567400947822,0.279638743249117,10.982969284057617 -SNRPE,19,0.022969253361225128,0.10965033620595932,0.9713645249906661,0.27726477055184556,9.977411270141602 -SNRPF,13,0.03517434746026993,0.13473139703273773,0.9574491141542288,0.27750869149461155,12.346887588500977 -SNRPG,40,0.011497988365590572,0.07651504874229431,0.9844753942337113,0.2930560557419422,7.059198379516602 -SNUPN,14,0.012369949370622635,0.08528609573841095,0.9837697050047527,0.03960859786963941,7.321977615356445 -SNW1,22,0.007154072169214487,0.06458891183137894,0.9903755200392527,0.0029199034196148955,5.5682806968688965 -SNX15,63,0.002760139759629965,0.039335545152425766,0.9963209251823957,0.0922041606926899,3.458677053451538 -SOD1,17,0.008562442846596241,0.06995602697134018,0.9884524838432807,0.12039933232935199,6.091766834259033 -SOD2,20,0.006797391921281815,0.06329241394996643,0.9910534246949209,0.09120072720675024,5.42769718170166 -SON,51,0.009099695831537247,0.06723900139331818,0.9885923855269442,0.35552598599481794,6.279974460601807 -SP2,12,0.012668791227042675,0.08495563268661499,0.9833646831431584,-0.06292819313487027,7.4098944664001465 -SPATA5,7,0.01706240326166153,0.10072111338376999,0.9779562227644133,0.044968860831589894,8.599328994750977 -SPATA5L1,15,0.01121166069060564,0.08072331547737122,0.9848416706143458,0.15160678778351497,6.9707489013671875 -SPC24,10,0.016503266990184784,0.09596416354179382,0.977608546719653,0.19169075112492848,8.457254409790039 -SPC25,20,0.022967802360653877,0.10853311419487,0.9699579782765895,0.3319437289378961,9.977096557617188 -SPCS2,12,0.009902655147016048,0.07501078397035599,0.9872724405240255,0.036007769601392146,6.551191329956055 -SPCS3,16,0.00870251189917326,0.07048670202493668,0.9887786733593577,-0.07697420220479405,6.141390800476074 -SPDL1,9,0.016348471865057945,0.09637309610843658,0.9779877065977901,0.09463763419114629,8.417498588562012 -SPDYE2,18,0.007314351852983236,0.06557092815637589,0.9905500015140567,-0.03420922709837882,5.630310535430908 -SPRTN,14,0.00982755422592163,0.07554323226213455,0.9870275688075588,-0.05610455991123354,6.526302337646484 -SPTLC1,3,0.036742620170116425,0.14898256957530975,0.9537438810220233,0.05601890651792369,12.619132995605469 -SPTSSA,13,0.010039846412837505,0.07715463638305664,0.9867490661302785,0.04573541937769693,6.596415042877197 -SRBD1,15,0.007598338183015585,0.06736928969621658,0.9902887592563357,0.011656711542600109,5.738571643829346 -SRCAP,20,0.00825016014277935,0.06919866800308228,0.989216106493592,0.09184547976010313,5.979648113250732 -SRF,13,0.010350161232054234,0.07853934913873672,0.9869813169010061,-0.05674549967573008,6.6975812911987305 -SRFBP1,7,0.025717806071043015,0.12237067520618439,0.9658955851267679,-0.006704874590347822,10.557507514953613 -SRP14,13,0.012323436327278614,0.08336826413869858,0.9842074928255244,0.038502313539855795,7.308198928833008 -SRP19,13,0.009789574891328812,0.07574687153100967,0.9872050257945418,0.10431219085251904,6.513679027557373 -SRP54,8,0.021069645881652832,0.11236559599637985,0.9731185646588705,0.19094188176071988,9.555931091308594 -SRP68,17,0.015162110328674316,0.09025035053491592,0.9805909186593242,0.3285167319057102,8.106329917907715 -SRP72,23,0.008016766980290413,0.06695542484521866,0.9893721730412224,0.2788666401174117,5.894460201263428 -SRP9,17,0.009492826648056507,0.0721764788031578,0.9873047806759769,0.1511244449504861,6.414196014404297 -SRPRB,17,0.012028662487864494,0.08313198387622833,0.9854812191979315,0.15574811137465439,7.220264434814453 -SRRM1,18,0.008917240425944328,0.07068195194005966,0.9882881501660998,0.12839699561922452,6.2166972160339355 -SRRM2,18,0.010818329639732838,0.07702505588531494,0.9853616952499389,0.11220992814803873,6.847382068634033 -SRRT,30,0.019238872453570366,0.09666823595762253,0.9744874256299567,0.17248146494783367,9.13133430480957 -SRSF1,16,0.014326308853924274,0.0870317593216896,0.9805858036893029,0.1432298464447866,7.879734992980957 -SRSF10,14,0.015005729161202908,0.0905454233288765,0.9800999003692639,0.15015824400796332,8.064417839050293 -SRSF11,43,0.003509331261739135,0.045369263738393784,0.99537138581946,-0.06672767821570563,3.899928569793701 -SRSF2,7,0.022195495665073395,0.11210376769304276,0.9711943664732908,0.09914776736620112,9.8079195022583 -SRSF3,3,0.061733052134513855,0.1894056648015976,0.9206847057803983,0.16255139951129852,16.35698699951172 -SRSF6,23,0.005782647989690304,0.058105576783418655,0.9924386330822773,0.009323283063453161,5.006196022033691 -SRSF7,12,0.014277403242886066,0.08887866884469986,0.9818115599915741,0.026599338547710936,7.866273880004883 -SS18L2,15,0.012236897833645344,0.08261249214410782,0.983571397173799,0.11844146009139322,7.282493591308594 -SSBP1,12,0.015414622612297535,0.09098152071237564,0.9812059540641125,0.01973034004417734,8.173553466796875 -SSBP3,10,0.013797182589769363,0.08970925956964493,0.9817864871443377,-0.013412706750905173,7.732851505279541 -SSRP1,56,0.008900325745344162,0.06620325893163681,0.9887724396333805,0.2756856598006243,6.2107977867126465 -SSU72,9,0.021043825894594193,0.10899648070335388,0.9720241746958816,0.05934102665817734,9.55007553100586 -SSX4,10,0.013470126315951347,0.08942974358797073,0.9827043530325947,-0.025112376425558858,7.640649795532227 -STARD7,11,0.013473734259605408,0.08721250295639038,0.9833125604892946,0.07212733069475859,7.641672611236572 -STAT5A,21,0.006841474678367376,0.06299271434545517,0.9907578014045251,0.23362990436621006,5.4452691078186035 -STIL,17,0.008570080623030663,0.07034610211849213,0.9885772392372502,-0.06735994695700577,6.094483852386475 -STRAP,10,0.015336167067289352,0.09308364242315292,0.9804068953002416,-0.04089011141418665,8.152726173400879 -STRIP1,10,0.01254638284444809,0.08534439653158188,0.984158021231013,-0.07448469836786381,7.37401008605957 -STX18,8,0.019729148596525192,0.10607235878705978,0.9733924782683226,0.12145656487048992,9.246953010559082 -STX5,5,0.02919463999569416,0.13209624588489532,0.961236138100366,-0.01824394865005184,11.248536109924316 -SUGP1,15,0.009500124491751194,0.07477430254220963,0.9876403780083992,0.056640976559991854,6.416661262512207 -SUGT1,10,0.014701352454721928,0.09196778386831284,0.9808099631212963,-0.10480464123353504,7.982208728790283 -SUMO2,7,0.018772395327687263,0.10400228947401047,0.9759695745988101,0.008718630994629597,9.019953727722168 -SUPT16H,10,0.036952752619981766,0.13971519470214844,0.9518209828866445,0.2719805027809966,12.655165672302246 -SUPT4H1,8,0.022980064153671265,0.11460811644792557,0.9689318491147327,0.030878029635126147,9.979759216308594 -SUPT5H,40,0.035524990409612656,0.1295575201511383,0.9586871266423787,0.2858337133720726,12.408275604248047 -SUPT6H,29,0.03979916498064995,0.13886794447898865,0.9512842156206136,0.38593670065082053,13.133527755737305 -SUPV3L1,8,0.021899346262216568,0.10949425399303436,0.971723123117644,0.19213836470965745,9.742266654968262 -SURF6,13,0.011358649469912052,0.08100627362728119,0.985178890366991,0.09876101414066805,7.016294479370117 -SYF2,8,0.017137886956334114,0.09966724365949631,0.9782343737420219,0.16443641033119702,8.618330001831055 -SYMPK,9,0.03406753018498421,0.1346379816532135,0.9545684117790769,0.08990589721926112,12.151077270507812 -SYS1,6,0.02399194799363613,0.11770620197057724,0.9682406229697718,0.03234767338752477,10.197112083435059 -TACC3,13,0.010075587779283524,0.07612176984548569,0.9868674907345035,0.060782028417828815,6.6081461906433105 -TADA2A,4,0.031726300716400146,0.13682393729686737,0.961079858512775,0.12171324013151179,11.726115226745605 -TADA3,15,0.009538202546536922,0.07451363652944565,0.9874702505314124,0.06737529417399288,6.429507732391357 -TAF1,9,0.034124426543712616,0.1369607001543045,0.9554611323307749,0.130932320714122,12.161219596862793 -TAF10,4,0.05823231488466263,0.1847822666168213,0.9243667662274523,0.1558112474676128,15.886436462402344 -TAF12,6,0.04938677325844765,0.16467958688735962,0.931982649432529,0.21183636495346247,14.630184173583984 -TAF1B,10,0.015044620260596275,0.09462389349937439,0.9802674542857245,0.019590645035769665,8.074861526489258 -TAF1C,14,0.01192800235003233,0.08300404250621796,0.9844019338152202,0.08752210434474192,7.189990520477295 -TAF1D,14,0.010479791089892387,0.07710307836532593,0.9861149223261068,0.1570195659320855,6.7393927574157715 -TAF2,15,0.026224739849567413,0.11498991400003433,0.9656683103260681,0.14149419129560492,10.661051750183105 -TAF5,3,0.098456971347332,0.23611770570278168,0.8701034691430017,0.15707006563651926,20.657020568847656 -TAF6,12,0.022016242146492004,0.10716698318719864,0.9699516098218128,0.11567825308957332,9.768234252929688 -TAF7,10,0.022659210488200188,0.11265575885772705,0.969685172715574,0.0755128261564834,9.909844398498535 -TAF8,11,0.02632131241261959,0.11804064363241196,0.9644320210970316,0.04345328860326593,10.680663108825684 -TAMM41,12,0.010199621319770813,0.07845601439476013,0.9872362237757184,0.025962830355280384,6.648695468902588 -TANGO6,9,0.023536769673228264,0.11602547019720078,0.9695837331194129,0.1511540319129739,10.099919319152832 -TARDBP,17,0.009365558624267578,0.07292260229587555,0.9873523199058507,0.0882435228358757,6.371053695678711 -TARS,13,0.015230704098939896,0.0924651250243187,0.9798793740441288,0.23775643862639756,8.124645233154297 -TARS2,8,0.019829614087939262,0.10792344808578491,0.9749987366834557,0.02940508851030937,9.270466804504395 -TBC1D28,8,0.013356219045817852,0.08916506916284561,0.9831419091684718,-0.018216156264208673,7.608275413513184 -TBC1D3,3,0.05435856431722641,0.1819421797990799,0.9323809107939751,0.06525221776237448,15.3489408493042 -TBCA,18,0.008860310539603233,0.07105788588523865,0.9885364680647755,0.12453933814582742,6.1968207359313965 -TBCB,10,0.015016824007034302,0.09328445792198181,0.980345581927297,0.027470275864326536,8.067398071289062 -TBCC,14,0.009162673726677895,0.07295142114162445,0.9886097845402018,-0.07677370339969067,6.301668643951416 -TBCD,3,0.039217159152030945,0.1533403992652893,0.9533279825195751,-0.021635175999687797,13.037145614624023 -TBCE,13,0.010336107574403286,0.07822328805923462,0.9865934931631616,-0.02009726429002665,6.693033218383789 -TBL3,7,0.02324211224913597,0.11489470303058624,0.9702022515684593,0.06126572160393309,10.0364990234375 -TBP,15,0.011651032604277134,0.08061148226261139,0.9845431411159421,0.0011785304602093425,7.106023788452148 -TCERG1,11,0.013380688615143299,0.08600874245166779,0.9823572198269936,0.2739351678787561,7.615242004394531 -TCOF1,9,0.019448386505246162,0.10683995485305786,0.9748209753714592,0.16323039476149998,9.180920600891113 -TCP1,6,0.03517813980579376,0.1420094221830368,0.9566164937023036,0.14399769991256298,12.347552299499512 -TDGF1,10,0.012480784207582474,0.08541850745677948,0.9835535780403484,0.09835371390597342,7.35470724105835 -TEAD3,26,0.005017274059355259,0.054046381264925,0.9935092974208073,-0.003981309888040886,4.6631388664245605 -TEFM,21,0.00818528514355421,0.06630906462669373,0.9902823662116808,0.06526042224938686,5.956091403961182 -TELO2,8,0.01859649270772934,0.10421077162027359,0.9753453433108132,0.20374817473936654,8.977594375610352 -TEN1,8,0.015652284026145935,0.09634855389595032,0.9791784313369821,0.04694911048567536,8.236321449279785 -TERF1,9,0.013802863657474518,0.09047702699899673,0.9824698607708658,-0.09297506586650628,7.734442710876465 -TERF2,5,0.028115488588809967,0.12739567458629608,0.964480521861633,0.08654623737792531,11.03868293762207 -TEX10,16,0.010634446516633034,0.07718363404273987,0.986741382934971,0.09642184982848793,6.788938999176025 -TFAM,8,0.022286739200353622,0.11218473315238953,0.9708347592294698,0.11800191053145126,9.828059196472168 -TFB2M,16,0.007682491559535265,0.06707291305065155,0.9898336657171116,0.16566961171051076,5.770261287689209 -TFDP1,17,0.009114674292504787,0.07220016419887543,0.9877240745573377,0.0720532191143387,6.2851409912109375 -TFIP11,6,0.038384027779102325,0.1483449935913086,0.9549257191793726,0.1974390587353808,12.897921562194824 -TFRC,5,0.024847103282809258,0.121381014585495,0.9674537852718872,0.076549067616815,10.377250671386719 -TGS1,16,0.008396421559154987,0.07014468312263489,0.9889614906970091,-0.033471254167363684,6.032419681549072 -THAP1,11,0.015018736012279987,0.09248911589384079,0.9797725483454066,0.0975988019057435,8.067911148071289 -THAP11,7,0.018415063619613647,0.10283853858709335,0.9761212497242956,-0.014386602653393038,8.933693885803223 -THG1L,5,0.0338447168469429,0.13998056948184967,0.9588843919215315,0.18735894651636434,12.111276626586914 -THOC1,4,0.046021029353141785,0.16242150962352753,0.9387322082183169,0.07326312326586923,14.122859001159668 -THOC2,19,0.016211969777941704,0.09207577258348465,0.979077493356529,0.23919466703222023,8.382284164428711 -THOC3,14,0.01230599358677864,0.08240264654159546,0.9836721417012141,-0.016258439063284977,7.303025245666504 -THOC5,3,0.05294668674468994,0.17658933997154236,0.9350014265291002,0.10928604591644174,15.148299217224121 -THOC6,11,0.011888108216226101,0.08383578807115555,0.9841090670381831,-0.005020125931050473,7.177956581115723 -THOC7,13,0.011110411956906319,0.07980760931968689,0.9854669022187978,0.137084278210854,6.939202308654785 -TICRR,15,0.012889578007161617,0.08241119235754013,0.9825203624183608,0.09613025647063328,7.474184513092041 -TIGD1,6,0.0208931565284729,0.11030775308609009,0.9734073379567543,-0.07143240838350073,9.515825271606445 -TIMELESS,7,0.01494091097265482,0.09388884156942368,0.9808726698708763,0.03272757725575616,8.046981811523438 -TIMM10,6,0.01962330937385559,0.10753218829631805,0.975541830516476,0.025034859656691305,9.222116470336914 -TIMM13,18,0.012668745592236519,0.08129490911960602,0.9830381103073683,0.22133419204666577,7.409881114959717 -TIMM22,8,0.01551008876413107,0.09544887393712997,0.9804063106072598,-0.015809932353017895,8.198824882507324 -TIMM23B,50,0.018753215670585632,0.09191711246967316,0.9754923168575034,0.46508439913811933,9.01534366607666 -TIMM44,5,0.02486584708094597,0.12018238753080368,0.9675376835999036,0.11455839246181487,10.38116455078125 -TIMM8A,16,0.008048553951084614,0.06863123178482056,0.9894730016664282,0.03461157529837683,5.906135082244873 -TIMM9,20,0.007463525515049696,0.06571266055107117,0.9900230768217786,0.24979100915731575,5.687435150146484 -TINF2,7,0.02204984799027443,0.11340168118476868,0.9701808454804306,0.11274645310331352,9.775686264038086 -TIPIN,11,0.010908347554504871,0.07928115874528885,0.9861122854520057,-0.105521591644015,6.8758111000061035 -TKT,20,0.007537727244198322,0.06616870313882828,0.9899295855109086,-0.004132512202717021,5.71563720703125 -TLCD1,15,0.00809748936444521,0.06972857564687729,0.9895756989398334,0.0548104785032802,5.924062728881836 -TLN1,8,0.016110919415950775,0.09754308313131332,0.979136367201503,-0.030786561957249502,8.356119155883789 -TM7SF2,10,0.013350561261177063,0.08774162828922272,0.9824310696728078,0.04313725317730343,7.606663703918457 -TMA16,17,0.008241137489676476,0.06844097375869751,0.9898250411699504,0.15091089515719894,5.976377487182617 -TMED2,13,0.009701704606413841,0.07559020817279816,0.9875140933151216,0.03772085837640573,6.48438024520874 -TMEM127,13,0.010299726389348507,0.07838763296604156,0.9864950742648105,0.038868178323880016,6.681243896484375 -TMEM199,10,0.012120858766138554,0.08456116914749146,0.9843673844661887,-0.028004670698161788,7.247882843017578 -TMEM240,17,0.007852276787161827,0.0689024105668068,0.9899527388429775,-0.038031035879458976,5.833674907684326 -TMEM258,20,0.00746738538146019,0.06586667895317078,0.9901202560180666,0.08918805503248405,5.688905715942383 -TMSB10,23,0.006159526761621237,0.05871237441897392,0.9917491353746669,-0.014723970707046447,5.166758060455322 -TMX2,11,0.017790764570236206,0.10131402313709259,0.9761725695901592,0.19995171142762191,8.780954360961914 -TNNT2,3,0.03288890793919563,0.1407247632741928,0.959382732230255,-0.06281110569071711,11.939034461975098 -TNPO3,10,0.014750934205949306,0.0929034873843193,0.9809365912988576,-0.029854138665257976,7.9956583976745605 -TOE1,14,0.010870348662137985,0.08024521172046661,0.9857137196804866,-0.026840786854579236,6.86382532119751 -TOMM20,11,0.014950251206755638,0.09202894568443298,0.9802415400127342,0.02227947606119782,8.049495697021484 -TOMM22,11,0.016317859292030334,0.09634526073932648,0.9795398524978163,0.1937397157878036,8.409613609313965 -TOMM40,11,0.014024500735104084,0.08957663178443909,0.9822510547202271,0.12969736824805073,7.796292781829834 -TONSL,3,0.03430034965276718,0.1435852199792862,0.9578221571316119,-0.017991944216851127,12.192527770996094 -TOP1,13,0.013276730664074421,0.08801484853029251,0.9822015460729295,0.03079381916078406,7.585601806640625 -TOP2A,4,0.03464583680033684,0.14027492702007294,0.956030509935925,0.09068655517278212,12.253777503967285 -TOP3A,16,0.00835816003382206,0.07094550877809525,0.9889132730600464,-0.015019340826544799,6.018659591674805 -TOPBP1,5,0.026933584362268448,0.1260879933834076,0.9646704388190078,0.13474034886587669,10.804173469543457 -TOR2A,5,0.03105766326189041,0.13510870933532715,0.9603074833762563,-0.02173271765485475,11.60189151763916 -TP53RK,9,0.013142474927008152,0.0886603519320488,0.9833022232349686,-0.007396435292451184,7.5471510887146 -TPI1,4,0.033726975321769714,0.14095239341259003,0.9572337654334165,0.03217482083457221,12.090190887451172 -TPR,7,0.023602358996868134,0.11540485173463821,0.9684464930252629,0.14689676670315654,10.113981246948242 -TPRKB,4,0.03775327280163765,0.15092621743679047,0.951749827230581,0.142246391388899,12.791507720947266 -TPT1,5,0.030767204239964485,0.13347309827804565,0.9609220096875946,-0.05421550241898566,11.547513008117676 -TPX2,36,0.008620833978056908,0.06784369051456451,0.9896569796763055,0.27378723751428974,6.112503528594971 -TRA2B,22,0.008627399802207947,0.0683460384607315,0.988783103431152,0.10813741648900536,6.114830017089844 -TRAIP,8,0.01643761247396469,0.09699124097824097,0.9793495045313901,-0.023534883259953152,8.440415382385254 -TRAPPC1,10,0.013826764188706875,0.09093591570854187,0.9816676160158404,0.09637533960792506,7.7411370277404785 -TRAPPC11,11,0.012692770920693874,0.08700153976678848,0.9831829071816498,0.05936901091156656,7.416903972625732 -TRAPPC3,17,0.007437985390424728,0.06541051715612411,0.9900874863421033,0.04975886918204964,5.6776957511901855 -TRAPPC4,19,0.007486509624868631,0.06577065587043762,0.9901759833041467,0.007887385057305003,5.696186065673828 -TRAPPC5,13,0.009962175972759724,0.07534556090831757,0.9868545106321889,0.021984866350460767,6.570850372314453 -TRAPPC8,15,0.009015901945531368,0.0720924511551857,0.9880206819655525,0.06680800530724904,6.250993728637695 -TREML2,6,0.024381930008530617,0.1203332170844078,0.9688480497938123,0.011632270886808321,10.279653549194336 -TREX2,27,0.007668784353882074,0.06538352370262146,0.9899819940544866,0.23424644868257066,5.765111923217773 -TRIAP1,6,0.021730637177824974,0.11442975699901581,0.9735656820097939,0.07093687077602466,9.704668045043945 -TRIM49C,9,0.015041681937873363,0.09423195570707321,0.9804772539251243,-0.020297850038464592,8.07407283782959 -TRMT10C,9,0.016762511804699898,0.09712708741426468,0.9787583356721992,0.06097232577598001,8.523423194885254 -TRMT112,20,0.011008908040821552,0.07777443528175354,0.9858397343273623,0.14499512860649447,6.9074320793151855 -TRMT5,10,0.018786752596497536,0.10350476205348969,0.975498368312065,0.1767307633287587,9.023401260375977 -TRMT6,12,0.01072128489613533,0.07887145131826401,0.9858556066054649,-0.005488304271305062,6.816600799560547 -TRNP1,15,0.009187739342451096,0.07336295396089554,0.9879160477488602,-0.09863657939003535,6.3102827072143555 -TRNT1,21,0.006576916668564081,0.06261616945266724,0.9912318605695991,0.10535347963252951,5.338947296142578 -TRPM7,13,0.011219861917197704,0.0810215100646019,0.9851614687863363,0.02232856644091486,6.973297595977783 -TRRAP,42,0.011768748052418232,0.07567120343446732,0.9848630095509376,0.327616995815097,7.141831398010254 -TSEN2,17,0.010134199634194374,0.07478682696819305,0.9870638376389934,0.09275183324225603,6.627338886260986 -TSEN34,27,0.004629550967365503,0.05264841020107269,0.9939420483958272,0.0018029418067192315,4.479338645935059 -TSEN54,5,0.026699602603912354,0.12465839087963104,0.9652192617583452,0.010478357025521428,10.75714111328125 -TSFM,10,0.014058527536690235,0.09159446507692337,0.9814249574965295,0.058700437780627265,7.805745601654053 -TSG101,8,0.027256349101662636,0.12258194386959076,0.9639281407767324,0.2657416752200871,10.868717193603516 -TSPYL5,10,0.01442828867584467,0.09305572509765625,0.9812779200587215,0.008944053575884816,7.907730579376221 -TSR1,19,0.009872809052467346,0.07431036978960037,0.9870033739123648,0.18670737533535306,6.541311264038086 -TSR2,53,0.013327297754585743,0.08162534981966019,0.9841186826000938,0.31965942419254895,7.600033283233643 -TSSK3,21,0.006431893911212683,0.06161322444677353,0.9914678095138886,-0.020163011788010575,5.279756546020508 -TTC1,7,0.019287018105387688,0.10759162902832031,0.9755389021689005,0.11795626985193527,9.142753601074219 -TTC27,9,0.02125190570950508,0.10971786081790924,0.9735577213553229,-0.036591132536621856,9.597174644470215 -TTC4,11,0.013315939344465733,0.089176706969738,0.9827898676753212,-0.019729672322091187,7.596794128417969 -TTF2,7,0.017148546874523163,0.10169309377670288,0.9783357767877342,0.06544638025911119,8.621009826660156 -TTI1,15,0.013823295943439007,0.08872437477111816,0.9812393479123358,0.16440830254957284,7.7401652336120605 -TTI2,5,0.02987249754369259,0.13241660594940186,0.9608658005810528,0.01540537100383863,11.378374099731445 -TTK,20,0.010873720049858093,0.0771217867732048,0.9858418786773536,0.2904995673930781,6.864889144897461 -TUBA1B,8,0.021342232823371887,0.11003876477479935,0.9714321308035632,0.07762256937649717,9.617547988891602 -TUBA1C,5,0.021192286163568497,0.1108362227678299,0.9757692945322636,-0.05264585912595249,9.58370304107666 -TUBA3D,7,0.016050593927502632,0.096902996301651,0.9799796177461577,-0.08336762289040153,8.340459823608398 -TUBB,10,0.01255918201059103,0.08556179702281952,0.9834338207396712,0.07499339477721137,7.37777042388916 -TUBD1,19,0.007449361030012369,0.06569134443998337,0.9901490335827807,-0.0037573148808177645,5.68203592300415 -TUBE1,14,0.009696315973997116,0.07554226368665695,0.9872464187326209,0.033326252177119195,6.482579231262207 -TUBG1,2,0.08408243209123611,0.21381358802318573,0.9052149134677635,-0.0389828595639641,19.089611053466797 -TUBGCP2,24,0.0062115732580423355,0.05947518348693848,0.9916248121574904,0.12438000502233841,5.188540935516357 -TUBGCP3,3,0.04024548828601837,0.15238741040229797,0.9494104202544722,0.05997573507416476,13.206966400146484 -TUBGCP4,4,0.03935150429606438,0.15395338833332062,0.9481795093158449,0.11737558026607216,13.059456825256348 -TUBGCP5,7,0.016803458333015442,0.10055316984653473,0.9784281003366354,0.13050297316290255,8.533825874328613 -TUBGCP6,16,0.008719190955162048,0.07126621901988983,0.9884609679197425,-0.029238834297604318,6.147273540496826 -TULP1,13,0.010471414774656296,0.0784706324338913,0.9862865006250978,-0.044614081246693374,6.736699104309082 -TUT1,21,0.006451500579714775,0.06143244728446007,0.9915631583384569,-0.020420220665082332,5.287797451019287 -TVP23C,12,0.01294789183884859,0.08691847324371338,0.9830064794293067,-0.052841799300999845,7.491072654724121 -TWF1,14,0.009706814773380756,0.07592250406742096,0.9870570640257029,-0.004814704032188506,6.486087799072266 -TWISTNB,17,0.021977923810482025,0.10655717551708221,0.9720868367035201,0.31722227184666907,9.759729385375977 -TXN,10,0.013435909524559975,0.08795357495546341,0.9820108277905151,0.0962435654782248,7.63093900680542 -TXNL4A,9,0.01565621793270111,0.09549020975828171,0.9798301015706226,0.11827696599497534,8.237356185913086 -TXNL4B,13,0.009617997333407402,0.07544570416212082,0.9873854239726938,0.05258100756285158,6.456345081329346 -U2AF1,14,0.03126611188054085,0.127249076962471,0.9588447199602127,0.2784922591272039,11.640761375427246 -U2AF2,10,0.015083394013345242,0.0949847549200058,0.9808751789360542,0.09408023214013797,8.085260391235352 -U2SURP,29,0.01248722430318594,0.07885122299194336,0.9841215997161865,0.18847938043249293,7.35660457611084 -UBA1,28,0.017568835988640785,0.09291937947273254,0.9760876625310975,0.1926377174362875,8.726015090942383 -UBA2,17,0.011692570522427559,0.07771874964237213,0.9846205901411934,0.04584140759157464,7.118680000305176 -UBA3,9,0.018394920974969864,0.10391957312822342,0.9761058748468688,0.02316351454995223,8.92880630493164 -UBA5,10,0.011745085008442402,0.08238638937473297,0.9848900510927185,0.07756084762913588,7.134647369384766 -UBA52,6,0.02650960348546505,0.12421872466802597,0.9660400356569699,0.10994338310285784,10.71879768371582 -UBAP1,11,0.016302864998579025,0.0951235368847847,0.9781068757595135,0.1783696833300742,8.405749320983887 -UBC,10,0.013935118913650513,0.09007572382688522,0.9818730399600804,-0.011714798973622043,7.77140998840332 -UBE2D3,14,0.009206146001815796,0.07207635045051575,0.9878893746977129,0.11139515775553771,6.316600322723389 -UBE2H,20,0.007029626984149218,0.06327679753303528,0.990731316240566,-0.00451505955389369,5.5196380615234375 -UBE2I,7,0.023203153163194656,0.11377889662981033,0.9689938808771068,0.15347657297633266,10.028082847595215 -UBE2L3,20,0.00776681350544095,0.06372398138046265,0.9896069228620453,0.04607191843160069,5.801841735839844 -UBE2M,17,0.007998482324182987,0.06859748065471649,0.9894718256777036,0.025709064436243514,5.887734889984131 -UBE2N,18,0.0073645762167871,0.06519316881895065,0.9904945713964887,0.06620916302914276,5.649608135223389 -UBE2Z,7,0.01991027593612671,0.10983244329690933,0.9736217657839462,0.02757416817418122,9.289302825927734 -UBL5,23,0.011831063777208328,0.07926981151103973,0.9850977407759676,0.24599273096242696,7.160714149475098 -UBQLN4,15,0.01223623938858509,0.08362521976232529,0.9835825393794073,0.10110921002535304,7.282297611236572 -UBR4,2,0.049668218940496445,0.17784090340137482,0.940422464405994,0.02573082330502636,14.671812057495117 -UBR5,4,0.02549114264547825,0.1224096268415451,0.9686871249906763,-0.0432634104723901,10.510880470275879 -UBTF,20,0.006782772485166788,0.062997967004776,0.9911012262540575,0.02479876242946447,5.421856880187988 -UFL1,16,0.009083032608032227,0.07203225046396255,0.9879068555580796,0.17882871640752482,6.274221897125244 -UFM1,13,0.011604630388319492,0.08165352046489716,0.9847825539736325,0.13581294856144052,7.091859340667725 -UNC45A,18,0.007158499676734209,0.06455008685588837,0.9905451165623418,-2.925895898151077e-05,5.570003509521484 -UNC50,9,0.013478937558829784,0.08947227150201797,0.9821422560362364,0.06214662598247626,7.643147945404053 -UPF1,13,0.018127985298633575,0.09498205035924911,0.9765470616484667,0.14733833377413652,8.863784790039062 -UPF2,16,0.012268215417861938,0.08262991905212402,0.9838015729761583,-0.04490251080930452,7.291806697845459 -UPF3A,7,0.01790279895067215,0.10201767086982727,0.9772663957960747,-0.027868739287569704,8.808560371398926 -UQCC2,10,0.015638981014490128,0.09497564285993576,0.9793536109064689,0.03013353305956671,8.232820510864258 -UQCRB,9,0.016962045803666115,0.09933848679065704,0.9773814538368863,0.06217722983380016,8.574002265930176 -UQCRC1,31,0.00450828718021512,0.05138244479894638,0.994171296249179,-0.010423672384910918,4.420284748077393 -UQCRC2,7,0.020369447767734528,0.11020372062921524,0.9742386956407212,-0.03186094615605723,9.395807266235352 -UQCRFS1,12,0.01333332434296608,0.08657625317573547,0.9828311087349192,0.05739798528157985,7.601751804351807 -UQCRH,7,0.02044317126274109,0.10759661346673965,0.9730434346466105,0.1081717224778938,9.412795066833496 -UQCRQ,16,0.009118473157286644,0.07185205072164536,0.9885350772547704,0.048139246094016604,6.2864508628845215 -URB1,4,0.032573070377111435,0.14093516767024994,0.9591043161828713,0.0018035978341733527,11.881568908691406 -URB2,20,0.007123855408281088,0.06390733271837234,0.990848251789609,0.09785888585428905,5.556509017944336 -URI1,8,0.017603592947125435,0.10287755727767944,0.9771629118905935,0.15455862665351092,8.734642028808594 -URM1,13,0.009342716075479984,0.07479215413331985,0.9879149831639471,0.022395435761203085,6.363279819488525 -UROD,20,0.006533269304782152,0.061622876673936844,0.9918685712120856,-0.1260244373252854,5.321201801300049 -USF2,19,0.007807577960193157,0.067109115421772,0.989549884133385,-0.08109905044563409,5.817047595977783 -USP10,9,0.012844651006162167,0.08659473806619644,0.98332476259819,0.009318441074255017,7.461147308349609 -USP19,7,0.0169192086905241,0.09962154924869537,0.9788345181939744,-0.022240873707853367,8.563167572021484 -USP36,18,0.007871794514358044,0.06774821132421494,0.9895943030269201,-0.02425252611289244,5.840920448303223 -USP37,7,0.021610118448734283,0.11290775239467621,0.9711341240587745,0.023925559400619422,9.677720069885254 -USP39,22,0.013166236691176891,0.08370961993932724,0.982681376572361,0.23256506282593326,7.5539703369140625 -USP5,20,0.006852899212390184,0.0628996267914772,0.9908894187933589,-0.0044889567499956365,5.449813365936279 -USP7,27,0.0046434118412435055,0.0521312840282917,0.9939741033229483,0.03281678578544766,4.486039161682129 -USP8,8,0.015711065381765366,0.09594503790140152,0.9803346777253531,-0.027186334305890538,8.2517728805542 -USP9X,11,0.011661180295050144,0.08365869522094727,0.98454548394814,0.008949152627612086,7.10911750793457 -USPL1,10,0.012909683398902416,0.08576798439025879,0.9831704201785092,0.08025351500142364,7.480011463165283 -UTP15,15,0.02481803670525551,0.11273638159036636,0.9676437770094588,0.3201696000484993,10.371179580688477 -UTP18,7,0.023088881745934486,0.11602503061294556,0.9693921494529044,0.059234681143121,10.003360748291016 -UTP20,11,0.019174763932824135,0.10393348336219788,0.9743183121968513,0.145534619890879,9.116107940673828 -UTP23,17,0.012250717729330063,0.08249974250793457,0.9837131804844214,0.10059129590689801,7.286604881286621 -UTP3,2,0.08406317234039307,0.22859621047973633,0.9005616903798074,0.08383384965454406,19.087425231933594 -UTP6,7,0.03317917883396149,0.1362009197473526,0.9550889124608161,0.11900789282522482,11.991602897644043 -UXS1,13,0.011918218806385994,0.08291909843683243,0.9840872776921437,-0.028480953362550205,7.187041282653809 -UXT,21,0.009855121374130249,0.07475174963474274,0.9867171198726986,0.05137180017663427,6.535449504852295 -VARS,22,0.013574518263339996,0.08527723699808121,0.9829210405548335,0.2928845745635554,7.670199871063232 -VARS2,12,0.010514083318412304,0.07950419932603836,0.9867161407097298,0.08684882621000076,6.750410079956055 -VBP1,11,0.011222111992537975,0.08148898929357529,0.9852002338090711,0.003409763483788206,6.973996639251709 -VCP,24,0.0055458154529333115,0.05693952739238739,0.9926663442562458,0.0675950246562494,4.9026079177856445 -VEZT,21,0.006909748539328575,0.06324786692857742,0.9908613589820379,-0.0034180397260109153,5.472372055053711 -VHL,16,0.007688660640269518,0.06689188629388809,0.9901898858084569,-0.0827426634222192,5.772577285766602 -VMP1,10,0.012765257619321346,0.08761514723300934,0.9831294112622724,0.014548583376096866,7.438053131103516 -VPS13D,14,0.010132324881851673,0.07739933580160141,0.9865871197243324,-0.01579075313399325,6.626726150512695 -VPS18,15,0.008433825336396694,0.07056204229593277,0.9888308392096894,0.16938149433499908,6.045841693878174 -VPS25,9,0.016797982156276703,0.10004111379384995,0.9780863831163789,0.06976787452001201,8.532435417175293 -VPS28,6,0.05665629357099533,0.17022815346717834,0.9241575380779455,0.27042639398028123,15.66998291015625 -VPS29,17,0.010786599479615688,0.07762555032968521,0.9859547353896008,0.025794777021227067,6.8373332023620605 -VPS33A,18,0.0070136189460754395,0.06517685949802399,0.9910112187490974,-0.009066433692735552,5.513349533081055 -VPS35,5,0.03152933344244957,0.1371648907661438,0.9593544942575416,0.18097860756684295,11.689659118652344 -VPS37A,7,0.018498476594686508,0.10545818507671356,0.9771115315685962,-0.013806574524385136,8.953904151916504 -VPS37C,17,0.007522643078118563,0.06634379178285599,0.9899897473504052,0.11839134703233478,5.7099151611328125 -VPS41,33,0.0041013662703335285,0.04846574738621712,0.9946410038054475,0.008001267876193771,4.216078758239746 -VPS51,10,0.011931996792554855,0.08445528149604797,0.9846581398368667,0.03272284979384387,7.1911940574646 -VPS54,32,0.0046401540748775005,0.05163970962166786,0.9937492334667744,0.06430182005706649,4.484465599060059 -VPS72,10,0.015069183893501759,0.09295293688774109,0.9814253404742215,-0.01737530573434966,8.081450462341309 -WAC,16,0.009129725396633148,0.0728883147239685,0.9878423987895075,-0.0204634499830288,6.290328502655029 -WARS,7,0.018131136894226074,0.10389471054077148,0.9768174462828404,0.05806141063725109,8.864555358886719 -WBP1,10,0.015132891945540905,0.09464768320322037,0.9805837626062408,-0.054253560511943556,8.098515510559082 -WBP11,31,0.006131516303867102,0.05835292115807533,0.9917575593319127,0.3234804206787635,5.154996871948242 -WBP1L,20,0.006224761251360178,0.06151698902249336,0.9919123614926773,-0.03556819247570107,5.194046497344971 -WDHD1,18,0.007956018671393394,0.06799398362636566,0.9896581751265513,-0.02872221279335949,5.872085094451904 -WDR1,10,0.012165451422333717,0.08483761548995972,0.9843171781267279,-0.025283250250929,7.261202812194824 -WDR12,22,0.00960536114871502,0.07195264846086502,0.987744290566428,0.2585556499077037,6.452103137969971 -WDR18,11,0.013278167694807053,0.08737614005804062,0.9826605251495606,0.09984904508882328,7.58601188659668 -WDR24,8,0.023832380771636963,0.11716336011886597,0.9674816903092323,0.2488067131866668,10.163146018981934 -WDR25,16,0.008732471615076065,0.07228671759366989,0.9887038135692884,0.0819635707129656,6.151953220367432 -WDR26,22,0.007178477942943573,0.06422661244869232,0.9905067949669332,0.13789801490037526,5.577770233154297 -WDR3,37,0.009662947617471218,0.06957168877124786,0.9872377765763883,0.1560805236079453,6.471415042877197 -WDR33,11,0.01876368746161461,0.1041916236281395,0.9744948744492872,0.17715840472781685,9.017861366271973 -WDR36,29,0.010679449886083603,0.07393033057451248,0.9869824777711103,0.18376333990488597,6.80328893661499 -WDR4,34,0.003972639329731464,0.04811836779117584,0.9948487570870308,-0.03462581046535711,4.149387836456299 -WDR43,8,0.022412844002246857,0.11294344812631607,0.9701410479956365,0.14328549526172055,9.855823516845703 -WDR46,7,0.022164644673466682,0.11453381180763245,0.970463008318945,0.06838123519733968,9.801100730895996 -WDR5,8,0.02715321257710457,0.12045390903949738,0.9636000646151736,0.0017188314258590276,10.848134994506836 -WDR54,21,0.005875281523913145,0.058872997760772705,0.9923351549505104,0.045644401772523296,5.046133995056152 -WDR55,9,0.014095093123614788,0.09135328978300095,0.9822883171097087,0.05638922392871973,7.815890312194824 -WDR61,11,0.016614610329270363,0.09696739912033081,0.9789430959734324,0.05031225501868828,8.485735893249512 -WDR7,11,0.013230548240244389,0.08865290135145187,0.9825759816580405,-0.0005696491566472836,7.572397232055664 -WDR70,3,0.041421569883823395,0.16000057756900787,0.9510574529025919,0.045478781451775085,13.398548126220703 -WDR74,18,0.007979415357112885,0.06874582171440125,0.989496313659963,0.07535078774765037,5.880712985992432 -WDR75,16,0.013718661852180958,0.08689312636852264,0.9815107678128269,0.24399075267172315,7.710815906524658 -WDR77,4,0.03232190012931824,0.1387779712677002,0.959841378553375,0.04033225029970954,11.835672378540039 -WDR82,10,0.028924327343702316,0.12684452533721924,0.9609740522338507,0.25676429678543283,11.19633960723877 -WDR83OS,9,0.014562467113137245,0.09294258803129196,0.9807511062517911,-0.01916589765393291,7.944415092468262 -WDR92,6,0.021090026944875717,0.11104250699281693,0.9725484059874385,0.06761397820103676,9.560552597045898 -WDTC1,11,0.02004208229482174,0.10541123896837234,0.9731905398667884,0.09269605494057538,9.319999694824219 -WEE1,4,0.04830833151936531,0.17124398052692413,0.9393728124428806,0.16067693274891337,14.469565391540527 -WNK1,13,0.012057959102094173,0.08411452919244766,0.983679442208615,0.017264936901720956,7.22905158996582 -WTAP,17,0.007746560499072075,0.06725285947322845,0.9897968237744252,0.06552689762828896,5.794272422790527 -XAB2,16,0.022208858281373978,0.11009428650140762,0.9743391586388304,0.20236833185631933,9.810871124267578 -XPO1,20,0.01774921454489231,0.09380783885717392,0.9764205316613407,0.39334234364719073,8.770694732666016 -XPO5,22,0.00685111153870821,0.06258224695920944,0.9908237945640116,-0.01567170528506097,5.449102401733398 -XRCC1,26,0.005324552766978741,0.05566614121198654,0.992915908849541,0.04278449333256217,4.803812026977539 -XRCC2,15,0.00793400127440691,0.06775704026222229,0.9898196683732104,0.019111928166262004,5.863954544067383 -XRCC3,6,0.023790519684553146,0.11743269115686417,0.9684287551413495,0.12438623144193882,10.154216766357422 -XRCC5,9,0.018220029771327972,0.09953654557466507,0.9760676622150366,0.2138273463673781,8.886259078979492 -XRCC6,3,0.0529765784740448,0.17458084225654602,0.9357635174993728,0.271025713873248,15.152573585510254 -XRN1,1,0.186936154961586,0.33555662631988525,0.7781302895891473,0.17803112520692319,28.46368408203125 -XRN2,6,0.02408907376229763,0.11916915327310562,0.9687634412772228,0.05611763751025053,10.217731475830078 -YARS,19,0.0077106282114982605,0.06657260656356812,0.989748566994435,0.11921316647268,5.780818462371826 -YARS2,19,0.007306521758437157,0.06557030230760574,0.9902625293459433,0.13342747787993783,5.627296447753906 -YBX1,8,0.019101396203041077,0.10571128875017166,0.9743776562294473,-0.025947874083801215,9.098650932312012 -YEATS2,24,0.008979925885796547,0.06836532801389694,0.9881277149287316,-0.054697100450298346,6.238509178161621 -YEATS4,23,0.008837542496621609,0.06972155719995499,0.9883244448723186,0.04224571700429925,6.188853740692139 -YJEFN3,18,0.007152683567255735,0.06490316987037659,0.9911866088662703,-0.016383875992769276,5.567740440368652 -YKT6,7,0.02016528882086277,0.1078748106956482,0.9732284824557662,0.09031183001760869,9.348601341247559 -YPEL1,7,0.01495580468326807,0.09466224908828735,0.981243120411057,-0.0010244684636610245,8.050992012023926 -YPEL5,7,0.025784028694033623,0.12261635810136795,0.9656156788743745,0.13531267094677574,10.571091651916504 -YRDC,10,0.015624739229679108,0.09603076428174973,0.9794658768397901,-0.12200588055517218,8.229071617126465 -YTHDC1,23,0.007967416197061539,0.06689368933439255,0.9892473137598439,0.166128541605678,5.876289367675781 -YY1,8,0.01775502786040306,0.1016906201839447,0.977599642812145,-0.1640286370252709,8.772130966186523 -ZBTB11,14,0.010264248587191105,0.07774794846773148,0.986136974124878,0.1439724639034703,6.669726848602295 -ZBTB17,16,0.01281066332012415,0.08488525450229645,0.9829531164532535,0.06189185449866524,7.451269626617432 -ZC3H18,3,0.043576303869485855,0.16016517579555511,0.946467411081024,-0.0305647552030819,13.742624282836914 -ZC3H8,2,0.07030873000621796,0.20585960149765015,0.9175074381484783,-0.074863175482001,17.456174850463867 -ZCCHC9,6,0.030641471967101097,0.13471105694770813,0.9593732260093839,0.06297691586661588,11.523895263671875 -ZCRB1,14,0.012299534864723682,0.08454559743404388,0.9849937836369638,0.18912352917815967,7.301108360290527 -ZFC3H1,11,0.021338293328881264,0.09692390263080597,0.9714857982364727,0.014265422235406259,9.616661071777344 -ZFP69B,15,0.010961049236357212,0.07906293123960495,0.9856143340524973,0.03964274213313391,6.892401218414307 -ZMAT2,43,0.007583309896290302,0.06299998611211777,0.9903541571556703,0.1595041317549591,5.732893466949463 -ZMAT5,10,0.014941653236746788,0.09431391209363937,0.9801336317705975,0.07488275286887285,8.047181129455566 -ZNF100,9,0.013869541697204113,0.09055182337760925,0.9822657274338088,-0.04409238429705779,7.7531023025512695 -ZNF131,3,0.05070309713482857,0.17483468353748322,0.9343660277264348,0.19499555147162373,14.823873519897461 -ZNF133,10,0.013482339680194855,0.08858515322208405,0.9826142002094436,0.031491673391819144,7.644112586975098 -ZNF207,29,0.005051829386502504,0.054242342710494995,0.9932556752892321,0.003895997903904285,4.679169654846191 -ZNF236,9,0.020105041563510895,0.1055569052696228,0.9728610956353694,0.08875042208122962,9.334626197814941 -ZNF24,6,0.0177188478410244,0.10264896601438522,0.9775985274078902,0.0042413764229991425,8.763189315795898 -ZNF253,12,0.010444746352732182,0.07794404029846191,0.9867155381495916,-0.07614442121314127,6.728114604949951 -ZNF284,22,0.0055121383629739285,0.05695102736353874,0.9926859023366246,0.05392320922778682,4.887699604034424 -ZNF317,13,0.012642270885407925,0.08458991348743439,0.9831976639241075,0.05799227141978983,7.402134418487549 -ZNF335,14,0.00947117991745472,0.07410623133182526,0.98766639414511,-0.038582788632065206,6.40687894821167 -ZNF407,11,0.012698821723461151,0.08656458556652069,0.9837691306274099,-0.05402129072998636,7.41867208480835 -ZNF468,20,0.007066772319376469,0.06405884027481079,0.9908207779796779,-0.06788972242809263,5.5342020988464355 -ZNF492,15,0.009347354993224144,0.07343199849128723,0.9877048986930793,-0.08895612603196025,6.364859580993652 -ZNF506,17,0.008729121647775173,0.07130194455385208,0.9886412545834613,-0.0075227027585313645,6.150773525238037 -ZNF559,11,0.011050750501453876,0.08028276264667511,0.9862945762056574,-0.06874440676834002,6.9205451011657715 -ZNF574,32,0.004198749084025621,0.049662280827760696,0.9944419149443146,0.04646961973403059,4.265838623046875 -ZNF658,11,0.011913131922483444,0.08381322771310806,0.984605820350822,-0.07070054828413001,7.185506820678711 -ZNF674,12,0.010384442284703255,0.0782390683889389,0.9863967370214348,-0.06929939095360042,6.708663463592529 -ZNF676,9,0.012175018899142742,0.08471096307039261,0.9844851896865747,-0.015387380601260445,7.264057636260986 -ZNF687,21,0.0074616409838199615,0.06435292214155197,0.9902805239926482,-0.022243845416167897,5.6867170333862305 -ZNF718,33,0.004060612991452217,0.04895775020122528,0.9946062108172068,-0.0012528299135509513,4.195079803466797 -ZNF720,8,0.01640053279697895,0.09894020110368729,0.9789147412263132,-0.02868034586906224,8.430890083312988 -ZNF763,15,0.008531314320862293,0.07063953578472137,0.9889819589874855,-0.03794499487916171,6.080683708190918 -ZNF830,1,0.08709241449832916,0.23445840179920197,0.9007415875739937,0.07968902529999665,19.428293228149414 -ZNHIT2,9,0.013454452157020569,0.0899963229894638,0.9825798949688466,-0.016975249236162532,7.636202812194824 -ZNHIT3,7,0.02055770717561245,0.11029960960149765,0.973491650484061,-0.08111415525723525,9.439126014709473 -ZNHIT6,11,0.014712718315422535,0.09244530647993088,0.9803233444477387,0.014130266506275534,7.985294342041016 -ZNRD1,8,0.020984143018722534,0.11179748922586441,0.9718756940822243,0.1738481495386927,9.53652286529541 -ZRANB2,4,0.033375710248947144,0.14046815037727356,0.9590267847576464,-0.04503803771808171,12.027066230773926 -ZRSR2,12,0.011060850694775581,0.0812184140086174,0.9859046366568035,0.013843397950517108,6.923707962036133 -ZW10,20,0.00727293873205781,0.06518132239580154,0.9904377022834406,0.13121852667695552,5.614348888397217 -ZWINT,4,0.03870357200503349,0.14750000834465027,0.9493910435186644,0.10037423685637788,12.951497077941895 +AAAS,21,0.007128496654331684,0.06409785896539688,0.9908975556050067,-0.08412781266371759,5.558318614959717 +AAMP,10,0.01875360496342182,0.1031767874956131,0.9748458605459952,0.31992305266141974,9.015438079833984 +AARS,4,0.043422624468803406,0.1586150825023651,0.9451732835339303,0.38464385463317247,13.718369483947754 +AARS2,18,0.007762049790471792,0.06780442595481873,0.9896573190928062,-0.031809335833800224,5.80006217956543 +AASDHPPT,5,0.029958361759781837,0.13361647725105286,0.9644449003625863,0.05261099996977247,11.39471435546875 +AATF,30,0.010757389478385448,0.07492978125810623,0.9856489919760861,0.16478026912312918,6.828068733215332 +ABCB10,5,0.02784634381532669,0.12738963961601257,0.9640863659810471,0.16673745580112204,10.98572063446045 +ABCB7,20,0.006757416762411594,0.0629572719335556,0.9907824427215817,0.14813352552045284,5.411713600158691 +ABCE1,25,0.015523304231464863,0.08934441208839417,0.9786560470577926,0.433538394131402,8.202316284179688 +ABCF1,33,0.005278406199067831,0.05474933609366417,0.9929652399895035,0.1881206142109484,4.782949924468994 +ABCG1,15,0.009791839867830276,0.07591027766466141,0.9874597929256332,-0.08060829028200353,6.51443338394165 +ABHD11,5,0.026137232780456543,0.12652786076068878,0.9667884667381353,0.031819359420828,10.643250465393066 +ABHD17A,9,0.016242578625679016,0.0979544073343277,0.9789250015913621,0.07769438211226153,8.390192985534668 +ABT1,10,0.017588337883353233,0.10044572502374649,0.9767777207555348,-0.017164765328465864,8.730856895446777 +ACD,7,0.017464134842157364,0.10193204879760742,0.9788051562649722,-0.12319302608572805,8.69997501373291 +ACIN1,19,0.007794989738613367,0.06677686423063278,0.989952897493914,-0.10763002783033034,5.812356472015381 +ACOT12,18,0.007995796389877796,0.06879372149705887,0.9894597359439217,0.03082844803218322,5.886746406555176 +ACTB,17,0.008189328014850616,0.0676315575838089,0.9891443743930648,0.04244916475233133,5.95756196975708 +ACTL6A,5,0.03113120049238205,0.13372266292572021,0.9601777199537155,0.03122980744826084,11.615618705749512 +ACTR10,11,0.012855787761509418,0.08774030953645706,0.9826619005747775,0.1222662180655573,7.464381217956543 +ACTR1A,15,0.00953635387122631,0.0752796158194542,0.9879099714055272,-0.060688163015731496,6.428884506225586 +ACTR1B,20,0.007557112257927656,0.06684702634811401,0.9898652289132192,-0.0510625295163948,5.722982406616211 +ACTR2,23,0.0063506425358355045,0.05940631031990051,0.9918014993965816,0.03676506892401757,5.246302127838135 +ACTR3,12,0.011479003354907036,0.08210543543100357,0.9850776964340053,0.07673222477649931,7.053367614746094 +ACTR6,11,0.014373648911714554,0.0929567888379097,0.9830489076493064,-0.03435243867678789,7.892743110656738 +ACTR8,25,0.008904100395739079,0.07024776190519333,0.9878902670026845,0.05945957984538465,6.2121148109436035 +ADAT2,9,0.015848582610487938,0.09723002463579178,0.9797940791080137,-0.05084543395123365,8.287808418273926 +ADAT3,13,0.010591324418783188,0.07866224646568298,0.9856948169178886,0.05365355367125814,6.775160789489746 +ADRM1,11,0.01543673686683178,0.0960545465350151,0.9801202400748821,-0.0038985675543898746,8.179413795471191 +ADSL,14,0.00966042559593916,0.07457978278398514,0.9875810868763326,0.02771931922318336,6.4705705642700195 +AFG3L2,1,0.12308494001626968,0.2809484899044037,0.8723726057691267,0.11025309854731853,23.096538543701172 +AHCTF1,29,0.0050647808238863945,0.05406622588634491,0.9931790195078829,0.18580238732581653,4.685163497924805 +AHCY,14,0.012287267483770847,0.0839652344584465,0.9842948745016724,-0.18449052714646508,7.297466278076172 +AIFM1,5,0.02619108185172081,0.12610651552677155,0.9668703681068422,-0.03812032069200016,10.654207229614258 +AKIRIN2,17,0.011824766173958778,0.08198411762714386,0.9840239969067177,0.23672815374585493,7.158808708190918 +ALDOA,8,0.020296700298786163,0.1089637354016304,0.9728668228936728,0.3000994816396055,9.379014015197754 +ALG1,7,0.020469702780246735,0.1111297532916069,0.9740331265195046,-0.039302637753586285,9.418900489807129 +ALG11,9,0.014336146414279938,0.09165781736373901,0.9810849143359716,0.12263233913512665,7.882440090179443 +ALG13,17,0.007841257378458977,0.06827333569526672,0.9900234034113221,0.08501139699175574,5.829580307006836 +ALG14,19,0.008182553574442863,0.0686853900551796,0.9890996591419062,-0.04417994547485744,5.955097675323486 +ALG1L,21,0.006938644219189882,0.06330644339323044,0.991102607887868,-0.14539561798197384,5.48380184173584 +ALG2,24,0.005987366195768118,0.05826641619205475,0.9920551001675049,0.11339457545410407,5.094040393829346 +ALYREF,32,0.0044334582053124905,0.050199780613183975,0.9941832001852156,0.11603647270938919,4.383447170257568 +ANAPC1,13,0.010837022215127945,0.0792791023850441,0.985399548444497,0.08980715583232093,6.85329532623291 +ANAPC10,3,0.054175734519958496,0.1803981214761734,0.9324588043158188,0.04704405629827443,15.323107719421387 +ANAPC11,7,0.02033473365008831,0.10788098722696304,0.9747285360398954,-0.09549318879737649,9.387797355651855 +ANAPC13,11,0.013961324468255043,0.09111633896827698,0.9816352041147537,-0.035588996697258325,7.778713226318359 +ANAPC15,43,0.0038629984483122826,0.04700247943401337,0.9950179497970504,-0.095065194790729,4.0917277336120605 +ANAPC2,23,0.006564669776707888,0.061700284481048584,0.9910792723171167,0.09047795192950137,5.3339738845825195 +ANAPC4,7,0.022335320711135864,0.11502024531364441,0.970421202123736,0.18119630896124703,9.838763236999512 +ANAPC5,4,0.033739153295755386,0.14191214740276337,0.9563363487321599,0.09017612395789867,12.092373847961426 +ANKLE2,12,0.01157174352556467,0.08194737136363983,0.9846494393292662,-0.010505035251939805,7.081803321838379 +ANKRD11,26,0.005636780988425016,0.05586089938879013,0.9932380307662221,0.007103659294390629,4.942651748657227 +ANKRD49,15,0.008833199739456177,0.07297623157501221,0.9889664587314885,-0.02614737667603081,6.187333106994629 +ANKS6,9,0.013158312067389488,0.08814582228660583,0.9831943940839131,-0.07921127872588381,7.55169677734375 +ANLN,9,0.014320566318929195,0.09198550134897232,0.9821269047156265,-0.11420233542178171,7.878155708312988 +AP2M1,21,0.006334280129522085,0.060699835419654846,0.9917106794469723,0.03873787285025099,5.23953914642334 +AP2S1,13,0.009964138269424438,0.07533567398786545,0.9868315813676746,0.23048957565638198,6.571496963500977 +AQP7,11,0.013847720809280872,0.0897940918803215,0.9815318084366746,-0.031766730350629026,7.747000694274902 +AQR,9,0.05864939093589783,0.1714717447757721,0.9226080710035298,0.3615645927658826,15.943225860595703 +ARCN1,34,0.004745080601423979,0.05279397964477539,0.9935002978602911,0.29989792262107323,4.534884929656982 +ARF4,9,0.015057992190122604,0.09359431266784668,0.9804191157850435,-0.027674464422117535,8.078449249267578 +ARFRP1,14,0.00876233633607626,0.07131146639585495,0.9892209918497672,-0.03004379449320928,6.162464618682861 +ARGFX,9,0.016281507909297943,0.09790187329053879,0.978615915242485,0.05753271676001183,8.40024185180664 +ARGLU1,50,0.003800898790359497,0.04594767838716507,0.9952608630537122,0.07638434585054515,4.058706283569336 +ARHGAP11B,21,0.006938018836081028,0.06361910700798035,0.9909599763210865,-0.06680411930719952,5.483554840087891 +ARID3A,3,0.04687276482582092,0.17057864367961884,0.941399066081841,-0.10584246228605503,14.252948760986328 +ARIH1,6,0.022306878119707108,0.11425647139549255,0.9708975749808713,0.020846834793923203,9.832497596740723 +ARL2,8,0.02102584019303322,0.1096438392996788,0.9723161653983158,0.03312891956733669,9.545992851257324 +ARL4D,12,0.011805938556790352,0.0832768902182579,0.9847788664706775,-0.0426782492053536,7.153106689453125 +ARMC6,8,0.01713375374674797,0.10077443718910217,0.9782706093391577,-0.06524806422059369,8.617290496826172 +ARMC7,11,0.015670286491513252,0.09708176553249359,0.9790874758654747,0.29018845215232525,8.241056442260742 +ARPC2,22,0.006847826763987541,0.06287677586078644,0.9911187689356403,-0.08257853635940081,5.447795867919922 +ARPC3,14,0.009921357035636902,0.07631650567054749,0.9872708302069193,-0.05342276926640706,6.557374477386475 +ARPC4,22,0.006635815836489201,0.06264334917068481,0.991063551516713,0.10441834406938065,5.362800121307373 +ASF1B,29,0.004508113954216242,0.05184686556458473,0.9941304632046949,0.034642763600275864,4.420199871063232 +ATF4,7,0.01894420199096203,0.10518351942300797,0.975873726494999,0.031130525320349798,9.061134338378906 +ATF5,14,0.0226991456001997,0.10782971233129501,0.9688619892695327,0.45929907922627744,9.918573379516602 +ATL2,12,0.012646248564124107,0.08644168078899384,0.9833245085152174,0.07310766637218707,7.403299331665039 +ATP1A1,9,0.0198360662907362,0.10558926314115524,0.9736529562219788,0.10389100989943356,9.271974563598633 +ATP2A2,16,0.00817564595490694,0.06897268444299698,0.9898192318490608,-0.002179707191092162,5.9525837898254395 +ATP6AP1,12,0.011529148556292057,0.08272464573383331,0.9848763119280185,0.0916466375857078,7.0687575340271 +ATP6AP2,17,0.007609053514897823,0.0669984295964241,0.9898915706097138,0.12807278620207602,5.742616176605225 +ATP6V0B,14,0.009800946339964867,0.07330752909183502,0.9872004916045621,0.09070547282783892,6.51746129989624 +ATP6V0C,6,0.023505795747041702,0.11752928048372269,0.969264406765025,0.21770312113806192,10.093271255493164 +ATP6V0D1,10,0.01426150742918253,0.09155573695898056,0.9812482500149556,0.09844720011176132,7.861893653869629 +ATP6V1A,9,0.013574925251305103,0.08802710473537445,0.9817067117675363,0.16764407760823508,7.670314311981201 +ATP6V1B2,44,0.005753742065280676,0.05413704738020897,0.9920854588326893,0.3848635853213743,4.993668079376221 +ATP6V1C1,6,0.020798729732632637,0.11125951260328293,0.9735372985901121,-0.017421609797793076,9.494297981262207 +ATP6V1D,14,0.009406395256519318,0.0746854841709137,0.9875289131169424,0.048919399488021854,6.3849287033081055 +ATP6V1E1,19,0.007307623978704214,0.06549777090549469,0.990218866441801,0.023531797495827553,5.627720832824707 +ATP6V1F,7,0.01984557695686817,0.10805398225784302,0.9747858907854685,-0.012042548651182292,9.274197578430176 +ATP6V1G1,6,0.024099215865135193,0.11859390139579773,0.9676008077408023,0.20106697698148185,10.21988296508789 +ATP6V1H,9,0.017569048330187798,0.10256188362836838,0.9776085877979748,0.016851587178879385,8.726067543029785 +ATR,14,0.012382948771119118,0.08500949293375015,0.9838029820286358,0.09611508389515337,7.32582426071167 +ATRIP,19,0.008958079852163792,0.07246626168489456,0.9886759972088732,0.13718660671460534,6.230916500091553 +AURKA,6,0.025799540802836418,0.11941273510456085,0.964849126576682,0.2934449207381427,10.574271202087402 +AURKAIP1,7,0.019375504925847054,0.10694348812103271,0.9750769586014353,0.013917851662006374,9.163702964782715 +AURKB,4,0.04595900699496269,0.16650846600532532,0.9430242048543448,0.1215071970735421,14.1133394241333 +BANF1,16,0.009410608559846878,0.07401272654533386,0.9873852122055703,-0.06621391140513137,6.386358737945557 +BANP,11,0.012566357851028442,0.08647292852401733,0.9835077026529797,0.03897223951021271,7.379877090454102 +BAP1,20,0.010998697951436043,0.07755865901708603,0.9863939468575195,-0.06678153148150873,6.9042277336120605 +BARD1,14,0.009345474652945995,0.07368806004524231,0.9883446531201001,-0.10968742422710469,6.3642191886901855 +BCAR1,24,0.006070566363632679,0.05873184651136398,0.9919433688098489,-0.039176626492196005,5.1293110847473145 +BCAS2,22,0.013226245529949665,0.0852266401052475,0.9826136583595256,0.28538354608440525,7.571165084838867 +BCL2L1,18,0.007982962764799595,0.06870917975902557,0.989524241624828,-0.033175077261366266,5.882019996643066 +BCLAF1,5,0.027741141617298126,0.12964756786823273,0.9647837796690898,0.011036736471908634,10.964949607849121 +BCR,56,0.019298817962408066,0.09151862561702728,0.9759795428553284,0.423521050664847,9.145549774169922 +BCS1L,7,0.017430853098630905,0.10277153551578522,0.9775897122060727,0.0649909683915262,8.691681861877441 +BDP1,33,0.006581231951713562,0.060375817120075226,0.9909926845045569,0.35583183287351355,5.340698719024658 +BET1,13,0.010189637541770935,0.0775526687502861,0.9862962204334377,0.04189178532560323,6.645441055297852 +BGLAP,23,0.00681046349927783,0.06268509477376938,0.990626834094648,0.27766628920006753,5.432913780212402 +BIRC5,4,0.030291900038719177,0.13374295830726624,0.9618160653510526,0.10101513555953756,11.45797061920166 +BMS1,3,0.051853589713573456,0.17387603223323822,0.9357746572582636,-0.029649656964347477,14.99111270904541 +BNIP1,11,0.012590274214744568,0.08490905910730362,0.9832957938924278,0.19886649616897656,7.386897087097168 +BOP1,26,0.005547574255615473,0.057278960943222046,0.9927950105380843,0.03150480855998972,4.903385162353516 +BORA,5,0.022023048251867294,0.1144050806760788,0.9713997058351685,0.10988593544760533,9.769742965698242 +BPTF,5,0.024792736396193504,0.12264467030763626,0.97002778084613,0.009001111711985865,10.365891456604004 +BRCA1,11,0.01178513839840889,0.08243289589881897,0.9848980133783055,0.028016067424035115,7.14680290222168 +BRCA2,14,0.00953587330877781,0.07567936927080154,0.9871927427483531,0.10857509877599474,6.428722858428955 +BRD4,11,0.011332708410918713,0.08115735650062561,0.9856783498510313,-0.0715047076965817,7.008277893066406 +BRD8,7,0.0210091695189476,0.11012360453605652,0.9730937517973268,0.20537236369888268,9.542208671569824 +BRF1,16,0.01016654260456562,0.07655996829271317,0.9866963012339809,0.06700117270325064,6.637906074523926 +BRF2,18,0.01019774004817009,0.07649070769548416,0.98648224956283,-0.02769309738250595,6.648083209991455 +BRIP1,13,0.012193471193313599,0.08501717448234558,0.984014696525131,-0.10116228114369208,7.269559860229492 +BRIX1,13,0.01415400579571724,0.09081391990184784,0.9805967609592903,0.28829277682269944,7.832206726074219 +BRK1,18,0.006892213597893715,0.06291589885950089,0.9911241010606869,0.025929572953887543,5.465423583984375 +BTF3,8,0.01873985305428505,0.10389088094234467,0.9754338735231604,0.11449076379582328,9.01213264465332 +BTF3L4,14,0.00946121383458376,0.07462728023529053,0.9877033340576681,-0.019006068532714568,6.403506755828857 +BUB1,65,0.0024745629634708166,0.03787804767489433,0.9967541229896074,0.03705979142318459,3.274867296218872 +BUB1B,14,0.009954561479389668,0.07610761374235153,0.987073384918413,0.045587204232141514,6.568338394165039 +BUB3,41,0.005002930760383606,0.05244073271751404,0.9932566212755518,0.2622626463746304,4.656468868255615 +BUD13,18,0.015877215191721916,0.09493321180343628,0.9808456497801836,0.22337077032092478,8.29529094696045 +BUD31,3,0.04380623996257782,0.16160590946674347,0.9463067144429177,0.018515896628079113,13.77883243560791 +BYSL,10,0.020391836762428284,0.10866609960794449,0.9723894982573864,0.154082590447935,9.400969505310059 +C12orf45,10,0.01659652404487133,0.09770245850086212,0.9778926835928846,0.029772181176305465,8.48111629486084 +C12orf60,11,0.015470003709197044,0.09287692606449127,0.9798780943218192,0.11624733236136452,8.188222885131836 +C14orf178,15,0.016777753829956055,0.09594974666833878,0.9774117715018515,0.29198049519372493,8.52729606628418 +C16orf86,19,0.007385218050330877,0.06688179820775986,0.9905880883978267,0.002133394629013319,5.657519817352295 +C17orf58,6,0.02106274478137493,0.11251097172498703,0.9734908820602381,-0.09289287759691235,9.554367065429688 +C19orf25,19,0.008957192301750183,0.07127827405929565,0.9881757919180532,-0.06388756083969227,6.2306084632873535 +C19orf53,17,0.008913266472518444,0.07179629057645798,0.9879819792908578,0.059854034903459995,6.215311527252197 +C1D,16,0.008473352529108524,0.07087798416614532,0.989162814394379,-0.05196460340685132,6.05999231338501 +C1QBP,21,0.008006257005035877,0.06596270948648453,0.988975265948687,0.1511648857388954,5.89059591293335 +C1QTNF4,20,0.006999356672167778,0.06401567161083221,0.9907034902834303,0.025546060351359163,5.5077409744262695 +C1orf109,24,0.005874726455658674,0.05886448174715042,0.9922994007215534,0.04996534511414463,5.045895576477051 +C6orf15,17,0.00804146844893694,0.06892522424459457,0.9890988531122863,0.17639994767084669,5.903534412384033 +C7orf26,8,0.019161725416779518,0.10417432337999344,0.974141785187182,0.18085952308104591,9.113008499145508 +C7orf50,10,0.013029811903834343,0.08807568997144699,0.9838339308442241,-0.05704651001996926,7.514732837677002 +C9orf16,21,0.007081174291670322,0.06319835782051086,0.9905944569670229,0.0031233639440790668,5.539838790893555 +C9orf78,20,0.007100037764757872,0.06344280391931534,0.9908738557381249,-0.06457976959195974,5.54721212387085 +CA5A,19,0.007906259037554264,0.06789058446884155,0.9895510533145233,0.07556807266654414,5.853693008422852 +CACNB3,31,0.004761483985930681,0.05208282545208931,0.9940905178946359,-0.11473856878915459,4.5427165031433105 +CACTIN,6,0.023376325145363808,0.11883782595396042,0.9698220932149311,0.014257359618219957,10.065435409545898 +CADM4,38,0.0041750152595341206,0.04856787994503975,0.9948367968521459,-0.1336712537497145,4.253765106201172 +CAMLG,10,0.012526457197964191,0.0858476385474205,0.9834361782153581,0.03734746881914633,7.368152141571045 +CAP1,2,0.05672432482242584,0.1842823177576065,0.9306933025983887,0.012089456117716484,15.679388999938965 +CAPZB,9,0.013778584077954292,0.08928315341472626,0.9817657759200474,-0.05161770808899986,7.72763729095459 +CARS,2,0.09550617635250092,0.2420976161956787,0.8823305994688788,0.273561384526275,20.345115661621094 +CASP8AP2,21,0.010724120773375034,0.07228165864944458,0.9855544947289001,0.22658035679460206,6.817502975463867 +CBLL1,10,0.017611689865589142,0.1008979007601738,0.9755676454372133,0.07279319928365913,8.736650466918945 +CBLN1,22,0.0065909274853765965,0.06159338727593422,0.9915432502888605,-0.10692751242196896,5.344630718231201 +CBX1,15,0.008995582349598408,0.0725891962647438,0.9880120605179692,0.02212248384405356,6.243945121765137 +CCAR1,5,0.027180183678865433,0.12760937213897705,0.964979093632283,0.09321084331745197,10.853520393371582 +CCDC115,21,0.00667144451290369,0.06273666769266129,0.9913156159556406,0.013428577745888006,5.377177715301514 +CCDC130,12,0.010807315818965435,0.0791931226849556,0.9859176943652449,0.04411045035678879,6.843895435333252 +CCDC137,20,0.007618506904691458,0.06699161976575851,0.9903125889116171,0.007322140494687613,5.746181488037109 +CCDC144A,11,0.013672241009771824,0.09040379524230957,0.9823141165100133,-0.02510707238499668,7.69775915145874 +CCDC144NL,8,0.01610064134001732,0.0975148156285286,0.9789807259486071,0.041002540851774016,8.353453636169434 +CCDC174,11,0.017335571348667145,0.09887966513633728,0.9769771302436857,0.31069064280265296,8.667892456054688 +CCDC59,13,0.011799631640315056,0.08235538005828857,0.9841627053969484,0.05158654592397597,7.151196002960205 +CCDC78,14,0.012207087129354477,0.08312880247831345,0.9834372495330244,0.15952176969464993,7.273617267608643 +CCDC84,8,0.018818248063325882,0.10381029546260834,0.9752267545789041,0.20721345780375697,9.030961990356445 +CCDC86,26,0.008787420578300953,0.07025125622749329,0.9885466072499817,0.2350276202997936,6.171278953552246 +CCNA2,3,0.04327977076172829,0.16358552873134613,0.9459958074010719,-0.003427954914610324,13.695785522460938 +CCNB1,16,0.009548383764922619,0.0753488764166832,0.9870898337612299,0.0785321317981201,6.432938098907471 +CCND1,14,0.0088085001334548,0.0727616623044014,0.9888094739975206,0.021030684553561062,6.178676605224609 +CCNH,8,0.025297174230217934,0.1211424469947815,0.9650087630283563,0.17076693226228737,10.47081470489502 +CCNK,32,0.005789158400148153,0.05633892863988876,0.9922357466833137,0.06170275232948853,5.0090131759643555 +CCNL1,17,0.008352790959179401,0.06971865147352219,0.9893121441149266,-0.019251861516765698,6.016726493835449 +CCP110,4,0.03308305889368057,0.14216741919517517,0.9589653490775716,-0.0618948030551794,11.974221229553223 +CCT2,20,0.013825993984937668,0.08391980826854706,0.9811555200630198,0.3488304301237635,7.7409210205078125 +CCT3,24,0.010751370340585709,0.0746133029460907,0.9856663641751826,0.21525204832182462,6.82615852355957 +CCT4,25,0.010873404331505299,0.076090008020401,0.9849849145949652,0.35715931074471363,6.8647894859313965 +CCT5,6,0.03130228817462921,0.13205206394195557,0.9587501876265678,0.12158979202128696,11.647493362426758 +CCT6A,13,0.01218393910676241,0.08136601746082306,0.9833857966351877,0.2309435882321528,7.266717910766602 +CCT7,15,0.015593644231557846,0.09067034721374512,0.9787142396944178,0.1264840068654294,8.220879554748535 +CCT8,11,0.014406608417630196,0.09018974006175995,0.9806572537131018,0.16171007160725195,7.901787281036377 +CD2BP2,13,0.010452264919877052,0.07805299758911133,0.9866521795841915,-0.05701293702212678,6.730535984039307 +CD3EAP,9,0.018893538042902946,0.10372931510210037,0.9743360428759226,0.1456204830865266,9.04901123046875 +CD8B,11,0.010692080482840538,0.07991167157888412,0.9862158469568,0.02010454010362187,6.807310581207275 +CDAN1,13,0.01235855184495449,0.08537011593580246,0.9831764005541253,0.05021878478781877,7.318603992462158 +CDC123,11,0.012155825272202492,0.08389241993427277,0.9834819387587476,0.21895775072430423,7.258329391479492 +CDC16,17,0.00942330714315176,0.07376927882432938,0.9877437381069127,0.13205761076536818,6.3906660079956055 +CDC20,11,0.013573789969086647,0.08761505782604218,0.9827076329797635,0.10409077082203408,7.669993877410889 +CDC23,17,0.008839596062898636,0.07055814564228058,0.9885792119828006,0.02215531786076317,6.189571857452393 +CDC26,5,0.02312890626490116,0.11756627261638641,0.9710269928905627,-0.020414763536873087,10.0120267868042 +CDC27,36,0.004004082642495632,0.047530289739370346,0.9946856283987289,0.06232775048403781,4.16577672958374 +CDC37,18,0.007611109875142574,0.06716065853834152,0.9900194240358434,0.005519190524344008,5.743391990661621 +CDC40,10,0.012414875440299511,0.08634649217128754,0.9837502687130062,0.05960601873882897,7.335261821746826 +CDC42,7,0.01866195723414421,0.10535541921854019,0.9753141617211811,0.14479837651956357,8.993382453918457 +CDC45,9,0.013453643769025803,0.08961942791938782,0.9827540906036802,0.01692702804381649,7.6359734535217285 +CDC5L,10,0.027276095002889633,0.12276417762041092,0.9649673248171554,0.2851763453027985,10.872653007507324 +CDC6,19,0.01013280265033245,0.07480350136756897,0.986024989319224,0.38102367836009,6.626882553100586 +CDC7,9,0.01501829270273447,0.09359657019376755,0.9802840720528688,0.017181513165443082,8.067791938781738 +CDC73,17,0.023331506177783012,0.11236488819122314,0.9710527405403045,0.23326775676540082,10.055782318115234 +CDCA5,6,0.023643050342798233,0.11886145919561386,0.9689937293127237,0.08273601904113075,10.122695922851562 +CDCA8,8,0.01869421824812889,0.10574491322040558,0.974538429238507,0.19937614141030205,9.001152038574219 +CDIPT,24,0.005840123165398836,0.059067048132419586,0.9923050114547809,0.27089842043439494,5.031013011932373 +CDK1,34,0.006396520417183638,0.058951787650585175,0.991297276223521,0.30599826225387833,5.2652177810668945 +CDK11A,27,0.004725160077214241,0.052902840077877045,0.9938185633699326,0.07840981758610052,4.525355339050293 +CDK2,11,0.01313017588108778,0.08815974742174149,0.9825015603077041,0.04455824444664833,7.543618679046631 +CDK7,2,0.07629810273647308,0.21794351935386658,0.9067097669653505,0.033376144563913406,18.184499740600586 +CDK9,14,0.014490039087831974,0.0906381607055664,0.9811985449105689,0.07223808884681587,7.9246344566345215 +CDT1,3,0.0729508027434349,0.20115354657173157,0.9003171170106143,0.3768745652001574,17.78113555908203 +CEBPZ,12,0.011175642721354961,0.08019044995307922,0.9853749458964021,0.19850375433620418,6.959542751312256 +CENPA,7,0.01771029271185398,0.10266582667827606,0.9781477419331425,0.027286046094657117,8.761073112487793 +CENPC,13,0.012503525242209435,0.08602559566497803,0.9837799984558139,0.15186059055030077,7.3614044189453125 +CENPE,17,0.010488288477063179,0.07659628242254257,0.9855875161515087,0.15991992884844872,6.742124557495117 +CENPH,7,0.02035794034600258,0.10872220993041992,0.9744714242511642,-0.07968084680308986,9.393152236938477 +CENPI,4,0.042539335787296295,0.160370334982872,0.9460141545802366,0.14409640119998046,13.578125 +CENPJ,28,0.008848119527101517,0.06679626554250717,0.9881600705624164,0.21557106690133088,6.192556381225586 +CENPK,9,0.01812046393752098,0.10354802757501602,0.9759211824170456,0.15674087479850568,8.861946105957031 +CENPN,14,0.00978882797062397,0.07621864229440689,0.9872662838027679,-0.008610598655210782,6.513430595397949 +CENPT,7,0.02172761969268322,0.11319386214017868,0.9710586163726307,0.04869413856839377,9.703993797302246 +CENPW,7,0.019123075529932976,0.10454418510198593,0.976454572303861,-0.09473433961126704,9.103813171386719 +CEP152,16,0.008902153931558132,0.07231788337230682,0.9880437094474269,0.08331046850099094,6.211435794830322 +CEP192,20,0.008408209308981895,0.0698361024260521,0.9885792175258573,0.1928203374945601,6.0366530418396 +CEP68,7,0.018533995375037193,0.1040898934006691,0.9763427643831655,0.10834136427468523,8.962496757507324 +CEP85,13,0.010904965922236443,0.0791434571146965,0.9853939901149198,0.07988866523311829,6.8747453689575195 +CEP97,13,0.010594654828310013,0.07972890883684158,0.9864528557007118,0.02469755027117655,6.776225566864014 +CFDP1,15,0.009332636371254921,0.0737508162856102,0.9884750746806685,-0.1282984405527257,6.359846591949463 +CFL1,3,0.03609921783208847,0.14615951478481293,0.957405801964774,-0.04972727421106949,12.508158683776855 +CHAF1A,4,0.040193092077970505,0.1521623581647873,0.9494926851731169,0.1377497996399709,13.198366165161133 +CHAF1B,9,0.028165608644485474,0.12307588756084442,0.9630297171337324,0.2681229961887667,11.048518180847168 +CHCHD1,9,0.014816784299910069,0.09402291476726532,0.9809147276731496,0.10179457953587902,8.013484954833984 +CHCHD2,9,0.013847216963768005,0.08911827206611633,0.9824736505559711,-0.027494340138132584,7.746860027313232 +CHCHD4,14,0.01117146760225296,0.08068163692951202,0.9857721091658289,0.24760636197744704,6.958242893218994 +CHD4,5,0.02955789305269718,0.13289505243301392,0.9605631433772693,0.2211231784145919,11.318299293518066 +CHEK1,29,0.008563201874494553,0.07003078609704971,0.9901313688487556,0.09190783103334127,6.092037200927734 +CHERP,6,0.024397267028689384,0.11984299123287201,0.9691085978245512,0.17765104369338974,10.282886505126953 +CHMP1A,8,0.014751527458429337,0.09419482201337814,0.9811570454021957,0.012226058706220344,7.995819091796875 +CHMP2A,10,0.035369355231523514,0.13215823471546173,0.9514069504118099,0.32951990431778405,12.381065368652344 +CHMP3,13,0.04334375262260437,0.13951338827610016,0.9396950323310739,0.39983466890407404,13.705904960632324 +CHMP4B,7,0.020220361649990082,0.11115362495183945,0.9724471089366038,0.18123402786175957,9.361359596252441 +CHMP5,11,0.016407331451773643,0.09847413748502731,0.9786744768908782,0.038432018795365404,8.432637214660645 +CHMP6,63,0.005738493520766497,0.05279867351055145,0.9923908460397552,0.36529732963983197,4.987046241760254 +CHMP7,24,0.005745814647525549,0.05800599977374077,0.9924229402610765,0.01666188499054627,4.990226745605469 +CHORDC1,26,0.005274494644254446,0.055573467165231705,0.993332128522397,-0.03785399284772702,4.781177520751953 +CHTF18,9,0.0157838873565197,0.09586348384618759,0.9794917543025653,0.10495890427187306,8.270874977111816 +CHTF8,5,0.026591062545776367,0.12442483007907867,0.9658499301706913,-0.003784265176431049,10.73525333404541 +CHTOP,15,0.010043242946267128,0.07642288506031036,0.9872828685584805,-0.05896943845202914,6.597530841827393 +CIAO1,25,0.005997756961733103,0.05878888815641403,0.9922326254540431,0.013241732339427187,5.098458290100098 +CIAPIN1,15,0.011513644829392433,0.08020000904798508,0.9843789653716212,0.253898108696863,7.064002990722656 +CINP,8,0.018713319674134254,0.10535193234682083,0.9753916477993204,0.010642045874853758,9.005749702453613 +CKAP5,31,0.004886130336672068,0.05334364250302315,0.9936936526172853,-0.07429908288897011,4.601791858673096 +CLASRP,12,0.012152873910963535,0.0854717418551445,0.9848615114446228,0.18778674768960116,7.257448196411133 +CLCC1,4,0.043176330626010895,0.16204889118671417,0.9445067303759872,-0.0035694185502417163,13.679408073425293 +CLK2,9,0.014771162532269955,0.093662329018116,0.981377476163477,-0.08557765912717161,8.001138687133789 +CLNS1A,6,0.023404087871313095,0.11758933216333389,0.9706328881563042,0.21480930908896895,10.0714111328125 +CLP1,5,0.028240809217095375,0.12762820720672607,0.9635891131890726,0.14408703700782524,11.063257217407227 +CLSPN,14,0.010270325466990471,0.07636728137731552,0.9858705072446649,0.17839773528295108,6.671700477600098 +CLTC,15,0.009826624765992165,0.0756857618689537,0.9867174474321779,0.13892577121716715,6.525993824005127 +CMPK1,18,0.007774416357278824,0.06735764443874359,0.9895233397359835,0.12975539057358967,5.804680824279785 +CMTR1,15,0.021515902131795883,0.10156458616256714,0.9711542074948484,-0.10730716496518161,9.656599998474121 +CNIH4,6,0.02288377657532692,0.11696359515190125,0.9698560908705887,0.08008285641516681,9.958829879760742 +CNN2,19,0.006920092273503542,0.06328117847442627,0.9909997788450242,-0.00809273163440744,5.476465702056885 +CNOT1,27,0.024196576327085495,0.10863137990236282,0.9687434015263319,0.3220297541368618,10.240506172180176 +CNOT2,9,0.018732398748397827,0.10442425310611725,0.9742588079953403,0.2161737654452537,9.010339736938477 +CNOT3,33,0.009472801350057125,0.0707705095410347,0.9870834706377533,0.22387475164215023,6.407426834106445 +COA5,17,0.008338593877851963,0.07028716802597046,0.9896160690192325,0.012123009217845104,6.011610984802246 +COASY,24,0.006280815228819847,0.060147568583488464,0.991795497691573,-0.10514085253339103,5.217380046844482 +COG1,4,0.02965390682220459,0.13313908874988556,0.9627831228810959,-0.07757259101697946,11.33666706085205 +COG2,35,0.004556277766823769,0.050972338765859604,0.9940330753742204,0.024455392016769235,4.44374942779541 +COG3,5,0.03371231630444527,0.14096981287002563,0.9555017740468676,0.15835806131055366,12.087562561035156 +COG4,13,0.010205304250121117,0.07731705158948898,0.986964066993074,0.01820659320559906,6.650547981262207 +COG6,19,0.008920739404857159,0.07232757657766342,0.9881272541482823,0.03740812246997387,6.217916011810303 +COG8,4,0.048533059656620026,0.17129261791706085,0.9375978847353282,0.24189627485416573,14.503181457519531 +COPA,25,0.0059105996042490005,0.058360449969768524,0.992078901932424,0.30963710757001306,5.061278343200684 +COPB1,9,0.02692984789609909,0.12126544862985611,0.9634811880142422,0.4738681636116937,10.803422927856445 +COPB2,13,0.010697636753320694,0.07986599951982498,0.9856441076065795,0.19071186517426542,6.809079170227051 +COPE,15,0.0097627192735672,0.07466036081314087,0.987382673223467,-0.019538630698383498,6.504738807678223 +COPG1,26,0.0052983807399868965,0.05614886060357094,0.9931332448572558,0.04649799828810417,4.791991710662842 +COPS2,4,0.04399922862648964,0.16294851899147034,0.9459163690917259,0.14420896639748085,13.809151649475098 +COPS3,8,0.021939555183053017,0.1143587976694107,0.9708169440744819,0.2017709175043747,9.751206398010254 +COPS4,26,0.006673944182693958,0.06238432973623276,0.9912913741601903,0.22871054024229379,5.378184795379639 +COPS5,10,0.014861210249364376,0.09354567527770996,0.9812532977386622,-0.04114601770359132,8.025490760803223 +COPS6,23,0.007688106968998909,0.06581969559192657,0.9899520901752572,0.12273104262178829,5.772369861602783 +COPS8,23,0.0077224066480994225,0.06672842055559158,0.9900580615725955,0.03949594563572402,5.785232067108154 +COPZ1,20,0.007912139408290386,0.06828692555427551,0.9891937985057251,0.17746233399261352,5.855870246887207 +COQ2,6,0.020084742456674576,0.10976701974868774,0.9747646529186612,-0.05745634354361342,9.329912185668945 +COQ4,16,0.00944883469492197,0.0736500546336174,0.9880576681054654,-0.1069885158497602,6.399316310882568 +COQ5,12,0.010829709470272064,0.07967490702867508,0.9859813265332191,-0.05222085724937456,6.850982189178467 +COX10,10,0.014854209497570992,0.09369612485170364,0.9806822982781283,0.17577674218436445,8.023599624633789 +COX11,11,0.014502781443297863,0.09157716482877731,0.98072575574245,0.11103142009843792,7.928117752075195 +COX15,11,0.016611170023679733,0.09872053563594818,0.9789418727831988,0.2374313197461043,8.484857559204102 +COX17,15,0.009209275245666504,0.07283848524093628,0.9878784518167367,0.05160640064821629,6.317673683166504 +COX5A,9,0.014483707956969738,0.09108179062604904,0.980538025541899,0.19077597056973866,7.922903060913086 +COX5B,12,0.013294004835188389,0.0870092362165451,0.9821277616094177,0.14790653305348694,7.590534687042236 +COX6C,11,0.012266267091035843,0.08404324948787689,0.9841089829313018,0.1392063072606249,7.2912278175354 +COX7B,12,0.010687073692679405,0.07885732501745224,0.9863486255893069,0.08160826578160109,6.805716514587402 +COX7C,17,0.00969485193490982,0.07330609858036041,0.9872310323102255,0.11002427406930171,6.482089519500732 +CPAMD8,24,0.006082791835069656,0.059453658759593964,0.9919658316794907,-0.03457432446294512,5.1344733238220215 +CPNE7,10,0.012969398871064186,0.08706972748041153,0.9840147785619707,0.0018032094593260241,7.497291088104248 +CPOX,10,0.012682096101343632,0.08641903847455978,0.9837342582221812,-0.030641161320618057,7.413784503936768 +CPSF1,5,0.02785971574485302,0.1274346113204956,0.9630231323782797,0.027322550887741413,10.988358497619629 +CPSF2,2,0.11401885747909546,0.2634263336658478,0.8615954099177617,0.13851550661427608,22.229658126831055 +CPSF3,62,0.0075644319877028465,0.06016023829579353,0.9898964261470098,0.2818142541445264,5.725752830505371 +CPSF4,6,0.030211880803108215,0.1318381279706955,0.9599669531420908,0.1773794095949964,11.442826271057129 +CPSF6,5,0.04002736136317253,0.15205170214176178,0.9455222537627064,0.11897098750535015,13.171126365661621 +CRCP,16,0.009070890955626965,0.07309330254793167,0.987832312368407,0.0687700864742703,6.270027160644531 +CRKL,12,0.010781385004520416,0.07979877293109894,0.9860466746137024,0.06865353402270818,6.83568000793457 +CRLS1,11,0.012135657481849194,0.08598902821540833,0.9851125701537539,-0.015959144359058954,7.252305507659912 +CRNKL1,26,0.006969988811761141,0.06306437402963638,0.9913451823877037,0.16568369563337812,5.496174335479736 +CS,22,0.006780983880162239,0.061304379254579544,0.9917985962537541,-0.051160493066001574,5.421142578125 +CSE1L,43,0.019604282453656197,0.09445767104625702,0.9731124517531892,0.359458826447435,9.217643737792969 +CSH2,18,0.007654344197362661,0.06640931218862534,0.9901198380494639,-0.03079986324413902,5.759681224822998 +CSNK1A1,17,0.010599249973893166,0.07832644134759903,0.9855162057593754,0.2757809990093154,6.777695178985596 +CSNK2B,13,0.014829191379249096,0.09055079519748688,0.9801452406377059,0.06985109981264792,8.016839027404785 +CSTF1,14,0.011842800304293633,0.08214573562145233,0.9837557130308039,0.17101546172433246,7.164265155792236 +CSTF3,6,0.031234558671712875,0.13561096787452698,0.9605393260125606,0.06603080228484655,11.634886741638184 +CTBP2,11,0.012545817531645298,0.08569277077913284,0.9838770699562732,-0.045915925064226606,7.373843669891357 +CTCF,7,0.023043980821967125,0.11585932970046997,0.9700924352339039,0.007735666372259205,9.99362850189209 +CTDP1,11,0.015545251779258251,0.09501615166664124,0.9790257662688349,0.07532149281626904,8.208113670349121 +CTNNBL1,7,0.0219681765884161,0.1121930256485939,0.9718256581919377,0.08009422127946232,9.75756549835205 +CTPS1,2,0.055204104632139206,0.18453240394592285,0.9367056505324218,0.01006595032718005,15.467856407165527 +CTR9,21,0.022985460236668587,0.10888146609067917,0.9741697173121475,0.1841044497626935,9.980931282043457 +CTU2,12,0.010902800597250462,0.0808345153927803,0.986012515744278,0.07507251522936516,6.874062538146973 +CUL1,23,0.006963520776480436,0.06318464875221252,0.9907735799018587,-0.10398912419019554,5.493623733520508 +CUL2,10,0.016615817323327065,0.09905248135328293,0.9783637245962813,0.10854789639251718,8.486044883728027 +CUL3,12,0.010965660214424133,0.07991520315408707,0.9862583005857086,-0.01471935807680835,6.893850326538086 +CUL7,4,0.04019489511847496,0.15635310113430023,0.9485316748664506,0.177636083643348,13.198661804199219 +CWC15,19,0.01607442833483219,0.09359310567378998,0.9795237082761546,0.23985764256980152,8.346650123596191 +CWC22,4,0.030963337048888206,0.13691751658916473,0.9618429236248353,-0.020614696174871418,11.584260940551758 +CWC25,15,0.008324690163135529,0.07090707868337631,0.9892312200260042,0.02950414011525427,6.006597518920898 +CWF19L2,20,0.011738321743905544,0.08060954511165619,0.984881616982529,0.1466219361432857,7.132593154907227 +CXXC1,12,0.015579253435134888,0.09311731159687042,0.9787499488750854,0.09463192105492557,8.217084884643555 +CYC1,10,0.011876622214913368,0.08324163407087326,0.984359777725668,0.21890550554704544,7.174488067626953 +CYCS,17,0.008240814320743084,0.06742562353610992,0.9893515203567221,0.04054373116759624,5.976260662078857 +CYFIP1,11,0.015190842561423779,0.09465963393449783,0.9798520214985038,0.20894392491445427,8.114006996154785 +CYP2A13,14,0.01126854494214058,0.08145982027053833,0.9854944703989006,0.1454599717507703,6.988409996032715 +CYP4F11,12,0.011021116748452187,0.08084071427583694,0.9857359682540965,-0.04093093800776788,6.91126012802124 +CYS1,17,0.00811015535145998,0.06861616671085358,0.9893701540526381,0.041285814119964216,5.928694248199463 +DAD1,8,0.022667860612273216,0.11130659282207489,0.9700509768610506,0.07267757082183704,9.911735534667969 +DAP3,16,0.008071564137935638,0.06790114939212799,0.9900632541405874,0.06913327234279758,5.914572238922119 +DARS,7,0.030786624178290367,0.13371197879314423,0.9622009553382735,0.363150399996903,11.551156997680664 +DBF4,13,0.010604682378470898,0.0789080485701561,0.9868198942806918,-0.09214916952593837,6.7794318199157715 +DBR1,10,0.019157953560352325,0.1070312038064003,0.9776002959454032,0.14069445438821768,9.11211109161377 +DCAF6,13,0.009409556165337563,0.07420926541090012,0.9878774514950764,-0.041320773295482865,6.3860015869140625 +DCLRE1B,11,0.011489808559417725,0.08244534581899643,0.9853064493419825,-0.04396696437541725,7.056686878204346 +DCTN1,19,0.007521549705415964,0.06614664942026138,0.9904112779402804,0.0036291570659169503,5.709500312805176 +DCTN2,12,0.013142427429556847,0.08774692565202713,0.9822110546706361,0.09166050928976971,7.547137260437012 +DCTN3,20,0.006466243881732225,0.06114673614501953,0.9918371907182176,-0.023487497400940974,5.2938361167907715 +DCTN4,7,0.01959371194243431,0.10822895169258118,0.9738240652946942,0.07936096416825537,9.215158462524414 +DCTN5,14,0.01113790925592184,0.08070620894432068,0.9847927158368027,0.17945769261561575,6.947783470153809 +DCTN6,7,0.018001582473516464,0.10237956792116165,0.9772276275674664,-0.0833721506328711,8.832828521728516 +DCUN1D5,19,0.008965381421148777,0.07144846767187119,0.9879505690108752,0.1717357636861363,6.233455181121826 +DDB1,17,0.008202305063605309,0.06939295679330826,0.9890409789034472,0.10006063275369083,5.9622802734375 +DDN,25,0.005601956974714994,0.05755434185266495,0.9925259638011881,0.1478436113265472,4.927360534667969 +DDOST,8,0.017878996208310127,0.10236894339323044,0.9769138567238288,0.16668065337166496,8.802702903747559 +DDX1,10,0.012041198089718819,0.08319046348333359,0.9847649005277196,-0.010204635500403523,7.224026203155518 +DDX10,11,0.0171599593013525,0.09909489750862122,0.9766967633222868,0.10383951263308415,8.62387752532959 +DDX11,13,0.011072823777794838,0.08124122023582458,0.9857268417678564,0.024073834947820293,6.927453994750977 +DDX18,13,0.019611146301031113,0.10386188328266144,0.9732399656216323,0.30661321941709807,9.219256401062012 +DDX19A,11,0.01490130927413702,0.09309443086385727,0.9800717126047802,0.011126355877286203,8.036309242248535 +DDX19B,29,0.004983145743608475,0.05414245277643204,0.9934232002200045,0.021624483247309097,4.647252559661865 +DDX20,12,0.01393137127161026,0.08960186690092087,0.9820391594000784,0.23169836856227327,7.770364284515381 +DDX21,9,0.01754748448729515,0.09957078844308853,0.9776019440391519,-0.032189615208440145,8.720710754394531 +DDX23,16,0.012949180789291859,0.08633614331483841,0.9830088018731263,0.17221760296136676,7.491445064544678 +DDX24,28,0.00919429399073124,0.06934396177530289,0.988610880504034,0.29502156508787625,6.312532424926758 +DDX27,5,0.03067423216998577,0.1358024924993515,0.9612670496769251,0.12059120662203407,11.530054092407227 +DDX3X,3,0.03712727501988411,0.1494331955909729,0.9533843344713765,0.13288985616271745,12.685014724731445 +DDX41,29,0.015105077996850014,0.09063953161239624,0.9821119495462565,0.184609708249087,8.091070175170898 +DDX42,5,0.02999158948659897,0.13249096274375916,0.961343068649215,-0.03554647869216008,11.401033401489258 +DDX46,22,0.00781735498458147,0.0664353147149086,0.989890698351146,0.1845574138398041,5.820688724517822 +DDX47,32,0.016884205862879753,0.09295645356178284,0.9788018630690528,0.21051382388528536,8.554305076599121 +DDX49,5,0.03395777568221092,0.142480731010437,0.9553190028090656,0.0017393970180977008,12.131487846374512 +DDX5,15,0.009601486846804619,0.07317660003900528,0.9884077908994217,-0.05998843199896681,6.450801372528076 +DDX51,13,0.012522079050540924,0.08597184717655182,0.9836425358928527,0.1520697002361309,7.366864204406738 +DDX52,11,0.015190557576715946,0.09311700612306595,0.9793605209854103,0.043566836478732766,8.113929748535156 +DDX54,9,0.018271200358867645,0.10396546870470047,0.9757090704923038,0.21027697313195862,8.89872932434082 +DDX55,17,0.008344528265297413,0.06946501135826111,0.9891472430676138,0.09593734222992743,6.013749599456787 +DDX56,18,0.015622956678271294,0.08939702063798904,0.9799552329000651,0.26654861627903625,8.228602409362793 +DDX59,16,0.009094771929085255,0.07277549058198929,0.9879410598543897,0.05860042094825048,6.278275489807129 +DDX6,7,0.028290430083870888,0.12676271796226501,0.9627546737378995,0.34826553351985134,11.072972297668457 +DENR,25,0.006707112304866314,0.06220926716923714,0.9909431887476736,0.15235288907625658,5.391532897949219 +DERL2,13,0.010284392163157463,0.07699897140264511,0.9861802613707302,0.14961488782991633,6.676268100738525 +DHDDS,9,0.023589879274368286,0.11195264756679535,0.9695829723229141,0.2628329292783204,10.111307144165039 +DHFR,13,0.009958701208233833,0.0766112431883812,0.9873164187325124,-0.06881529855898094,6.569704055786133 +DHODH,23,0.006334434729069471,0.06037026271224022,0.9916015825656319,-0.029738227591268,5.239603042602539 +DHPS,6,0.02108401618897915,0.11125317215919495,0.9741274253462454,-0.07821399278832931,9.55919075012207 +DHX15,36,0.020477214828133583,0.10015862435102463,0.9735517592376821,0.5205165380858264,9.420628547668457 +DHX16,33,0.017135504633188248,0.09606847912073135,0.9799312693845501,0.2527100210139655,8.617731094360352 +DHX33,14,0.013974220491945744,0.0896746888756752,0.9807732652735702,0.19237873248203816,7.782304763793945 +DHX36,27,0.0075371526181697845,0.06491917371749878,0.9896913767121631,0.1568548905013503,5.715419292449951 +DHX37,21,0.00955179799348116,0.07327210158109665,0.9869930724275743,0.06507564496367799,6.434088230133057 +DHX8,11,0.03164439648389816,0.13372176885604858,0.962363477608001,0.31355375557229404,11.710968971252441 +DHX9,9,0.014826749451458454,0.09308803826570511,0.980723791259948,0.16224510279986104,8.016179084777832 +DICER1,16,0.009861244820058346,0.07571356743574142,0.9866165971986388,0.03896892881895326,6.537479400634766 +DIMT1,10,0.022504635155200958,0.11253291368484497,0.9686410518465908,0.13500339396481975,9.875985145568848 +DIS3,5,0.0352376289665699,0.14188498258590698,0.9523848379715218,0.09989575069778198,12.357987403869629 +DKC1,6,0.0283296350389719,0.12887005507946014,0.963589627954291,-0.05023889506336683,11.080642700195312 +DLD,31,0.011958357878029346,0.07635369151830673,0.9834686870123786,0.3730841526851819,7.199133396148682 +DMAP1,8,0.030916066840291023,0.12487587332725525,0.959876833410109,0.2422938761537773,11.575414657592773 +DMRTA2,10,0.012729604728519917,0.0859551727771759,0.9833166895545907,-0.00457687764551304,7.4276580810546875 +DNAJA1,29,0.005149058066308498,0.054344646632671356,0.99299627553212,0.11345301134846472,4.723982810974121 +DNAJA3,12,0.016178835183382034,0.09603910893201828,0.9789998750700497,0.32877407550986104,8.373713493347168 +DNAJC11,4,0.03597979247570038,0.14789408445358276,0.9518951278575914,0.15313238895074974,12.48745059967041 +DNAJC17,3,0.05701911821961403,0.18506862223148346,0.9326878762458989,0.09159587341722462,15.720078468322754 +DNAJC19,40,0.01985320821404457,0.09566332399845123,0.9734682099643466,0.5900672677024968,9.275979995727539 +DNAJC8,27,0.005479902494698763,0.05519060790538788,0.9928160122802735,0.07820844695568409,4.873386383056641 +DNAJC9,9,0.01493031345307827,0.0932052880525589,0.9818556601455977,-0.13612688720656935,8.044126510620117 +DNLZ,14,0.009429091587662697,0.0745694562792778,0.987698684157549,0.036090125136949716,6.392627239227295 +DNM1,14,0.010592534206807613,0.07833421230316162,0.9859405674453801,0.03161031611521654,6.775547504425049 +DNM1L,17,0.011159584857523441,0.08072025328874588,0.9865550056097916,0.21051365232182756,6.954541206359863 +DNM2,1,0.12467099726200104,0.28211450576782227,0.875267459312935,0.06041276991187753,23.244873046875 +DNMT1,17,0.010477142408490181,0.07676810026168823,0.9865132207768695,-0.16852034038708613,6.738541126251221 +DNTTIP2,27,0.014174305833876133,0.08538255840539932,0.9813140549130478,0.2512589387449468,7.837821960449219 +DOHH,15,0.011830687522888184,0.08157170563936234,0.983924864064896,0.05636343956425786,7.160600662231445 +DOLK,14,0.009829362854361534,0.076058030128479,0.9869623854936929,0.016731585557121124,6.526902675628662 +DONSON,75,0.004403507336974144,0.04682037979364395,0.9940218148130632,0.4269629039062182,4.368615627288818 +DPAGT1,13,0.009831368923187256,0.0763210579752922,0.9871997581927435,-0.011818949988912805,6.527568817138672 +DPH1,33,0.0046234130859375,0.05229463800787926,0.9940461703075182,0.07751089114873629,4.476368427276611 +DPH2,17,0.008066271431744099,0.06922058761119843,0.9892023507910438,0.2078575851968629,5.912632465362549 +DPH3,6,0.020742256194353104,0.11085093021392822,0.9746823773586065,-0.10732476190706713,9.481399536132812 +DPH6,10,0.012821421027183533,0.08656691014766693,0.9832150710492455,0.04755649882974031,7.454397678375244 +DPM2,16,0.008155723102390766,0.06969964504241943,0.9892500624745757,0.004648850803326243,5.945326328277588 +DPPA2,6,0.02872960828244686,0.13156220316886902,0.9644460678543074,0.06410280827846851,11.158590316772461 +DPY19L2,5,0.027094393968582153,0.12700872123241425,0.9663581450118208,-0.05519450416564672,10.83637809753418 +DR1,12,0.012183429673314095,0.08541025966405869,0.9850472777278387,-0.09029617310469383,7.266566276550293 +DRAP1,22,0.006613471079617739,0.06134888157248497,0.9913691181273402,0.14509025048076865,5.353763580322266 +DRG1,19,0.007619166746735573,0.06692066788673401,0.9903308637215367,0.010565791077960848,5.746431350708008 +DSN1,25,0.010807017795741558,0.07680254429578781,0.9862256159515863,0.4313591535668317,6.843801498413086 +DTL,13,0.011995846405625343,0.08276258409023285,0.9834948999060299,0.2497039762625773,7.210409164428711 +DTYMK,12,0.010150291956961155,0.07721003890037537,0.9866411594972391,0.05399534024330763,6.632598876953125 +DUT,5,0.026856679469347,0.12506192922592163,0.9642200771512602,0.22267098885631764,10.788737297058105 +DYNC1H1,29,0.00802352000027895,0.06680120527744293,0.9895510860164529,0.40154358014182856,5.896942615509033 +DYNC1I2,30,0.004958403296768665,0.053721342235803604,0.9932086301096444,0.16981330354639695,4.635700225830078 +DYNLL1,7,0.021370884031057358,0.11097227782011032,0.9721332440492777,0.05901245854363564,9.624002456665039 +DYNLRB1,28,0.004794621374458075,0.05221918225288391,0.993476104913906,0.13947530422124643,4.558495998382568 +E2F6,15,0.009508057497441769,0.07513606548309326,0.9871108263590812,0.02373859125121717,6.419339656829834 +E4F1,14,0.01149966660887003,0.08067285269498825,0.9847756409078209,0.1303948300305711,7.059713363647461 +EARS2,19,0.007376690860837698,0.0648663118481636,0.9907067777455522,-0.08894749117291538,5.654253005981445 +EBNA1BP2,8,0.022670762613415718,0.1150774359703064,0.9693926966345663,0.19734609775922526,9.912369728088379 +ECD,7,0.021771486848592758,0.11309890449047089,0.9719936722958703,0.17174106388604993,9.713785171508789 +ECT2,5,0.017474398016929626,0.10129483044147491,0.9787522209832021,-0.06114391397904289,8.702529907226562 +EDC4,16,0.0101075554266572,0.07657238095998764,0.9868214129333506,-0.07982821545925001,6.618621349334717 +EEF1A1,3,0.06941187381744385,0.2007695585489273,0.9170492414162732,0.05013721809117663,17.344480514526367 +EEF1B2,3,0.03506007418036461,0.14550825953483582,0.9581598832823877,0.008255090746908444,12.326814651489258 +EEF1G,17,0.01249100361019373,0.07950147986412048,0.9838371123078542,0.12959020819415382,7.357717990875244 +EEF2,32,0.006411488633602858,0.05916226655244827,0.9912868071500277,0.13092185627212782,5.271374702453613 +EFTUD2,12,0.019726231694221497,0.10612944513559341,0.9753295282061905,0.2259199903903281,9.246269226074219 +EGLN2,36,0.0040118154138326645,0.04827454686164856,0.9947972379509014,0.03859028602659414,4.169797420501709 +EIF1,9,0.014171569608151913,0.08913350850343704,0.9825726521671306,-0.07336710134597868,7.837064743041992 +EIF1AD,4,0.028217054903507233,0.13101156055927277,0.9641589220380659,-0.06546822530734521,11.058603286743164 +EIF1AX,14,0.01244332455098629,0.08371052891016006,0.984214751650445,0.2939254011063934,7.343661785125732 +EIF2B1,18,0.010086041875183582,0.07523316890001297,0.9869137083587605,0.22282137364931046,6.611573696136475 +EIF2B2,32,0.014081045053899288,0.08248591423034668,0.9811512417857678,0.5482153304566895,7.811993598937988 +EIF2B3,11,0.01854785531759262,0.10115019232034683,0.9758429558224408,0.4360801426306383,8.96584701538086 +EIF2B4,33,0.01565600372850895,0.0873333141207695,0.9794681597587951,0.5681823704105001,8.237300872802734 +EIF2B5,46,0.012812118977308273,0.07813894003629684,0.9831362272341914,0.5982324436079364,7.451692581176758 +EIF2S1,25,0.025248898193240166,0.10788753628730774,0.9648017066137021,0.5162689805964992,10.46081829071045 +EIF2S2,12,0.013031266629695892,0.08579221367835999,0.9822969664013297,0.27727000533224555,7.515151500701904 +EIF2S3,5,0.0732586607336998,0.20408348739147186,0.9041163799484149,0.433185457923632,17.818614959716797 +EIF3A,38,0.0095644136890769,0.06951235234737396,0.9872026571221166,0.4588418951844001,6.43833589553833 +EIF3B,4,0.03333216533064842,0.14131735265254974,0.9576766312584213,0.1312167899626809,12.019219398498535 +EIF3CL,5,0.03210647776722908,0.1361665427684784,0.9604595955956858,0.06725931165813863,11.796163558959961 +EIF3D,18,0.01400177925825119,0.08621246367692947,0.982188601219573,0.3526842121886867,7.789975166320801 +EIF3E,23,0.01096309069544077,0.07594235241413116,0.9859184745210865,0.15084566085985068,6.893042087554932 +EIF3F,6,0.022530848160386086,0.11432013660669327,0.9698529301842059,0.2286999625881097,9.881735801696777 +EIF3G,6,0.03169462829828262,0.1337207406759262,0.9586937098445443,0.24141341941422573,11.720261573791504 +EIF3H,37,0.007011726964265108,0.06018940359354019,0.9911853292595123,0.33936808425670206,5.512606143951416 +EIF3I,6,0.024273548275232315,0.11941666156053543,0.9695516290704255,0.014046373814627443,10.256780624389648 +EIF3J,8,0.01941615715622902,0.10693611204624176,0.9763617209901128,-0.009320116482173573,9.173310279846191 +EIF3L,5,0.024557042866945267,0.11899151653051376,0.9692449096973288,-0.006566402886288969,10.316502571105957 +EIF3M,24,0.009202875196933746,0.0685865581035614,0.9879578258672413,0.24375479049800888,6.315478324890137 +EIF4A1,4,0.031312406063079834,0.13698431849479675,0.9607043237126057,-0.027875083731887663,11.649375915527344 +EIF4A3,1,0.12886400520801544,0.287149041891098,0.8621781673733062,-0.07453199163852152,23.632532119750977 +EIF4B,4,0.03258473798632622,0.1394353061914444,0.957925123779779,0.01039068691207504,11.883697509765625 +EIF4E,35,0.008688535541296005,0.06707505136728287,0.9885191733916439,0.33702701895674864,6.136457443237305 +EIF4G1,37,0.007301923353224993,0.061241429299116135,0.9903233979325627,0.3598476372101106,5.62552547454834 +EIF4G2,34,0.00465225288644433,0.051837008446455,0.9937947878687282,0.21541964946450895,4.490307807922363 +EIF5,17,0.010299385525286198,0.07697773724794388,0.985878344578853,0.24130502362138906,6.6811323165893555 +EIF5A,14,0.01298563089221716,0.08486536145210266,0.9828496751568964,0.09295681229967626,7.501981258392334 +EIF6,13,0.012847449630498886,0.08519792556762695,0.9833001268693389,0.15443069383478258,7.4619598388671875 +ELAC2,9,0.01346092950552702,0.08877212554216385,0.9829822723796573,0.019800396054143094,7.638040542602539 +ELL,9,0.015504276379942894,0.09613783657550812,0.9798905509364294,0.0767670553574099,8.197288513183594 +ELOVL1,17,0.008286809548735619,0.06946435570716858,0.988817904980074,0.027057906490196664,5.992915153503418 +ELP2,19,0.008103842847049236,0.06828221678733826,0.9898748238648881,-0.13171400471337874,5.926385879516602 +ELP3,12,0.01249456312507391,0.08607834577560425,0.9840981101554396,0.06717743337871969,7.358766078948975 +ELP5,17,0.00846011657267809,0.07060052454471588,0.9890721089540238,0.02966580316035816,6.055257797241211 +ELP6,9,0.014224485494196415,0.09145330637693405,0.9834812265154278,-0.14972431452478957,7.851682662963867 +EMC1,9,0.01615319959819317,0.09848025441169739,0.979026635545667,0.15904044649468901,8.367076873779297 +EMC3,19,0.007431442849338055,0.06629708409309387,0.9901447304273037,0.057964198705458776,5.675197601318359 +EMC4,5,0.025326089933514595,0.12327217310667038,0.9677267655764166,-0.046723663368631305,10.476797103881836 +EMC7,20,0.0066323475912213326,0.062366437166929245,0.9914432586819173,0.009736353453306701,5.361398696899414 +ENO1,7,0.021161742508411407,0.10873395949602127,0.9715328260171402,0.07595967922538518,9.576794624328613 +EP400,8,0.02828850969672203,0.12218137085437775,0.9605851249484835,0.25914762319868406,11.072596549987793 +EPRS,8,0.02404121868312359,0.1166725903749466,0.9674310884026124,0.3188952227537946,10.2075777053833 +ERAL1,11,0.013517378829419613,0.08947518467903137,0.9822817659661784,0.1061320627174737,7.65403938293457 +ERCC2,11,0.015484706498682499,0.09434623271226883,0.9787301044689241,0.019350769464707655,8.192112922668457 +ERCC3,17,0.009409484453499317,0.07416628301143646,0.9872153189201621,0.0724132104880752,6.385976791381836 +ERH,8,0.013948224484920502,0.09059011191129684,0.981609786782737,0.17478253922104411,7.7750630378723145 +ERVW-1,17,0.009030543267726898,0.07261613011360168,0.9886699477829225,-0.06995205410525167,6.256066799163818 +ESF1,15,0.011383365839719772,0.0815468579530716,0.9846544325279056,0.048929359726059114,7.023923873901367 +ESPL1,12,0.011947080492973328,0.08374999463558197,0.9837452493920088,0.22180742612682683,7.195738315582275 +ESPN,42,0.003751330077648163,0.04647495597600937,0.9952369508947172,-0.14734849413207948,4.032154083251953 +ETF1,5,0.046383727341890335,0.1608801931142807,0.939669730595002,0.42391101564189837,14.178400993347168 +EWSR1,5,0.024061203002929688,0.11926127225160599,0.9699570853256818,-0.050335326517262066,10.21181869506836 +EXOC1,18,0.007286890875548124,0.0654599666595459,0.9903366504394859,0.06387313703836571,5.619731903076172 +EXOC2,9,0.014132492244243622,0.0923333466053009,0.9820833909252198,0.04135444041977549,7.826251983642578 +EXOC3,24,0.006363933905959129,0.06064379960298538,0.9916705767739259,-0.04293142151354326,5.251789093017578 +EXOC4,11,0.012411162257194519,0.08580264449119568,0.9842920887174188,0.021651298635706338,7.334165096282959 +EXOC5,11,0.015946723520755768,0.09668239206075668,0.9781452132341673,0.23607233310716977,8.31342887878418 +EXOC7,3,0.08749885112047195,0.23168179392814636,0.8872790868312503,0.12036001953832537,19.473573684692383 +EXOC8,2,0.06498320400714874,0.1995922476053238,0.9237118704325538,-0.10017944249014782,16.78204917907715 +EXOSC1,7,0.019630638882517815,0.10647664219141006,0.9737270421905909,0.10487572292338182,9.223837852478027 +EXOSC10,9,0.016696419566869736,0.09841462969779968,0.9772056336260607,0.06469231330869142,8.506601333618164 +EXOSC2,22,0.021810224279761314,0.09754392504692078,0.9695988160777994,0.2138558734435629,9.722423553466797 +EXOSC3,6,0.04405263066291809,0.15579751133918762,0.9387174947743269,0.18370558627966088,13.81752872467041 +EXOSC4,10,0.028762871399521828,0.12315382063388824,0.9598482456968176,0.31003037691817575,11.165047645568848 +EXOSC5,4,0.04782647266983986,0.16510307788848877,0.9362222307225692,0.2972521854707232,14.39721965789795 +EXOSC6,7,0.024158621206879616,0.12024863809347153,0.9667054770532197,0.24220332271811224,10.232470512390137 +EXOSC7,12,0.012770920060575008,0.08583014458417892,0.9829640971035625,0.04964338611866259,7.439702033996582 +EXOSC8,12,0.022588422521948814,0.10578330606222153,0.9688509253608389,0.20875326963058372,9.894352912902832 +EXOSC9,4,0.06120649725198746,0.1879885047674179,0.9165195127473169,0.15877607541731714,16.287078857421875 +F8A1,3,0.046077556908130646,0.1660042703151703,0.9412319825187276,0.07133131304524962,14.131529808044434 +FAF2,17,0.007970256730914116,0.06727486848831177,0.9899059335746044,0.002051821547165674,5.877337455749512 +FAM133B,3,0.04274905472993851,0.16306796669960022,0.9485657604992198,0.032432419633120976,13.611554145812988 +FAM136A,5,0.04643465206027031,0.1646784543991089,0.9363088483592806,0.20487771067560365,14.186182975769043 +FAM207A,11,0.013837129808962345,0.09059315919876099,0.9816805598045039,0.04176588733575672,7.744037628173828 +FAM229A,13,0.010150608606636524,0.07776325941085815,0.9873928312488933,-0.05581001942957304,6.632701873779297 +FAM32A,39,0.005216163583099842,0.054362740367650986,0.9930647994261326,0.28829758120222065,4.754666328430176 +FAM50A,4,0.06020369753241539,0.18607281148433685,0.9215954229337667,0.14126155835792842,16.153104782104492 +FAM72D,12,0.014708335511386395,0.09315967559814453,0.9800472366385203,0.052876716827852166,7.984104633331299 +FARS2,4,0.04294100031256676,0.15922138094902039,0.9438043365171933,0.3336934186261536,13.642078399658203 +FARSA,14,0.017142146825790405,0.09740205854177475,0.9780610695087322,0.4241041120400338,8.619400978088379 +FARSB,9,0.02146429382264614,0.11037799715995789,0.9718356810965817,0.30193058053763155,9.645011901855469 +FASTKD5,18,0.007891297340393066,0.067708820104599,0.9896784338250512,0.09915759292821555,5.848151683807373 +FAU,5,0.03610195592045784,0.1449223756790161,0.951722523802406,0.1693602077271432,12.50863265991211 +FBL,13,0.02648257464170456,0.11791659891605377,0.964813909876051,0.26101386016910016,10.713332176208496 +FBLIM1,14,0.010396375320851803,0.07815363258123398,0.9863292198976208,0.00962109680924217,6.712517738342285 +FBXO5,32,0.004504431039094925,0.050402574241161346,0.9942242664137677,0.01965616583563685,4.418394088745117 +FCF1,23,0.015970095992088318,0.09077351540327072,0.9798793658981875,0.13549230693298128,8.31951904296875 +FDPS,1,0.12005903571844101,0.2777152955532074,0.8804480899447131,-0.023487361034683883,22.810871124267578 +FDXR,14,0.010185561142861843,0.07695703208446503,0.9863569256507625,0.005394045298173635,6.6441121101379395 +FEN1,14,0.01022583432495594,0.0775613933801651,0.986980740161358,0.0008545288790427147,6.657233715057373 +FGFR1OP,12,0.012226062826812267,0.08452772349119186,0.9840558536344871,-0.02032133746519085,7.279268741607666 +FIP1L1,2,0.09620827436447144,0.24262727797031403,0.888700649678676,0.20381401073697755,20.419761657714844 +FNBP4,16,0.008357936516404152,0.06973010301589966,0.9898417579097526,0.01530902768532118,6.018579483032227 +FNTA,8,0.017712783068418503,0.10213975608348846,0.9764269577145893,0.10205317830646482,8.761690139770508 +FNTB,10,0.012641791254281998,0.08677321672439575,0.9838500818604285,-0.09404229965009292,7.401994705200195 +FOLR3,31,0.004578708671033382,0.05207595229148865,0.9937553837034457,0.2575846523057236,4.454674243927002 +FOXD4,10,0.013653639703989029,0.09094014018774033,0.9822251282038189,-0.018112143234799753,7.692520618438721 +FOXL2,17,0.00810183584690094,0.06873264163732529,0.9891989441583487,0.014797667008880576,5.925652027130127 +FRG2,5,0.02492351271212101,0.12261741608381271,0.9688191161620344,0.04332937861268151,10.393195152282715 +FTSJ3,20,0.010703302919864655,0.0764884427189827,0.9854541872343755,0.22172467848002136,6.810882091522217 +FXN,18,0.007608470972627401,0.0667664185166359,0.9899180642024062,0.244392819165879,5.742395877838135 +GAB2,68,0.012132493779063225,0.06562389433383942,0.983870233633478,0.21442324118093878,7.2513604164123535 +GABPA,11,0.01324284914880991,0.08813171833753586,0.983062146657726,-0.0023768493828031185,7.575916290283203 +GAK,16,0.008863724768161774,0.07250916212797165,0.9882787260984609,0.11446743085498275,6.198014259338379 +GAPDH,19,0.008800845593214035,0.06947852671146393,0.9884828176066374,0.17876359717067355,6.175990581512451 +GAR1,14,0.012862752191722393,0.08584454655647278,0.9828138070215274,0.007654197160860152,7.466402530670166 +GARS,25,0.006092376075685024,0.058805689215660095,0.9921060372284782,0.10656018598423656,5.138516902923584 +GATA1,11,0.08786454051733017,0.21051275730133057,0.8823880233005499,0.39955958986098034,19.514223098754883 +GBF1,16,0.02166481874883175,0.10624687373638153,0.9719128477002537,0.4879099927807953,9.689959526062012 +GCOM1,6,0.020213885232806206,0.10971333086490631,0.9739798439123519,0.004551724059126651,9.35986042022705 +GEMIN4,8,0.019370274618268013,0.10778819769620895,0.9745491951136049,0.2164708945473736,9.16246509552002 +GEMIN5,14,0.012141497805714607,0.08331248164176941,0.9845204051196543,0.21858086496935428,7.2540507316589355 +GEMIN6,9,0.01478920504450798,0.09298750758171082,0.9810789722599832,0.02435603036506027,8.006023406982422 +GEMIN7,13,0.00946118775755167,0.07495996356010437,0.9875825858478492,0.06847692511344516,6.40349817276001 +GEMIN8,8,0.014536363072693348,0.0935838371515274,0.9814626901629103,-0.00011369673152413821,7.937291622161865 +GET1,8,0.016929078847169876,0.09959402680397034,0.9783488945625932,-0.05255853041869011,8.565665245056152 +GET3,12,0.010858816094696522,0.07950638234615326,0.9858612673024677,0.049212124495187595,6.860182285308838 +GFER,22,0.006821612361818552,0.06288991123437881,0.9908702675277454,0.24529442037365254,5.437358379364014 +GFM1,24,0.005475916899740696,0.05623749643564224,0.9929407868826637,0.18450500517591903,4.8716139793396 +GGPS1,16,0.008326733484864235,0.0697193518280983,0.9893033967312627,-0.09948690008524753,6.007333755493164 +GINS1,18,0.01868126541376114,0.09545135498046875,0.9741610996655307,0.3672494251676741,8.99803352355957 +GINS2,15,0.010807923972606659,0.07851988077163696,0.9857960137973061,0.12783125612078938,6.844087600708008 +GINS3,10,0.013433012180030346,0.08844821900129318,0.9818953842412339,0.25446043944202157,7.6301164627075195 +GINS4,6,0.022881969809532166,0.11673424392938614,0.9687602127466157,0.2800375990488205,9.958436012268066 +GIT2,10,0.013419545255601406,0.08907068520784378,0.9818042791323791,0.10256352337304929,7.626290798187256 +GJA3,8,0.015712816268205643,0.09675649553537369,0.9796773071438725,-0.05130707095673716,8.252233505249023 +GLE1,5,0.03233133256435394,0.13551782071590424,0.9573917277781562,0.3162675083535274,11.837398529052734 +GLI4,15,0.009489132091403008,0.07380719482898712,0.9878225879419819,-0.005933916328343279,6.412948131561279 +GLMN,8,0.01748148910701275,0.10107072442770004,0.9776682382694077,-0.03537093854610888,8.704296112060547 +GLRX3,13,0.010393905453383923,0.07755454629659653,0.9866197028924665,-0.025511275524823056,6.711719989776611 +GLRX5,10,0.013744762167334557,0.0896090567111969,0.9825673177939652,-0.09455924675707815,7.718147277832031 +GMPPB,17,0.008272337727248669,0.06835336238145828,0.9890863278819636,0.030056379452692083,5.987679958343506 +GMPS,13,0.01018050778657198,0.07640617340803146,0.9864028236091744,0.14138512048074098,6.642463207244873 +GNB1L,7,0.018590405583381653,0.10499322414398193,0.9766372878490562,-0.048357391700106236,8.97612476348877 +GNL2,35,0.008513045497238636,0.06639548391103745,0.9885687537824178,0.40033787181761366,6.074169635772705 +GNL3,4,0.030065514147281647,0.1328626424074173,0.9630665709844932,0.04276131132737536,11.415075302124023 +GNL3L,7,0.024072151631116867,0.1185542419552803,0.9676763583992063,0.26081950386665,10.214142799377441 +GNPNAT1,11,0.014151474460959435,0.09160540252923965,0.9813436443348366,-0.07116783031641823,7.831506252288818 +GOLGA6L1,19,0.007520841900259256,0.06546960771083832,0.9902648381367022,-0.0068915429529964,5.709231376647949 +GOLT1B,28,0.005822858307510614,0.05823766440153122,0.9921514002763225,0.2668219854714281,5.023571014404297 +GON4L,23,0.005964263342320919,0.0594945028424263,0.9924449617352883,0.06953125768352786,5.084202766418457 +GOSR2,9,0.014340230263769627,0.09225369989871979,0.9822805855942066,-0.08784552821210764,7.8835625648498535 +GPKOW,10,0.014577655121684074,0.09247921407222748,0.9818956842746126,-0.01748971229980996,7.948556900024414 +GPN1,8,0.028580281883478165,0.12886637449264526,0.9620762331353434,0.2297528878738317,11.129552841186523 +GPN2,5,0.03463941067457199,0.14214172959327698,0.9520506981114368,0.1483994869920611,12.252640724182129 +GPN3,27,0.017480097711086273,0.09457788616418839,0.9766696941950218,0.34701731110677136,8.703949928283691 +GPR61,11,0.011221356689929962,0.0809495598077774,0.9862062125420488,-0.09745232121621118,6.973762512207031 +GPS1,43,0.0048375846818089485,0.051814913749694824,0.9936253668404909,0.25323663418996967,4.578874588012695 +GRB2,17,0.010718023404479027,0.07746291160583496,0.9856030346617539,-0.08764745984644716,6.815564155578613 +GRPEL1,13,0.018663164228200912,0.09984065592288971,0.9749101177620627,0.4092329577378942,8.993673324584961 +GRSF1,17,0.009728418663144112,0.07561388611793518,0.9874879149423881,0.2578063086443163,6.493301868438721 +GRWD1,19,0.007946678437292576,0.06774305552244186,0.9894840222615393,0.021007230322981102,5.8686370849609375 +GSDMA,11,0.011197289451956749,0.08142367005348206,0.9853076959102683,0.10600734360393892,6.96627950668335 +GSPT1,11,0.018794909119606018,0.10070279985666275,0.975309229705002,0.4373490272385496,9.025361061096191 +GTF2A1,12,0.016786832362413406,0.09742536395788193,0.9768983604783189,0.11376909482634143,8.529603004455566 +GTF2A2,18,0.01150423102080822,0.07804681360721588,0.9841567090175088,0.04146940464361797,7.06111478805542 +GTF2B,3,0.08252085745334625,0.21808655560016632,0.8953349235669951,0.21156021674637993,18.911516189575195 +GTF2E1,18,0.00986933521926403,0.07527513056993484,0.9868806966368037,0.2524029498721574,6.540159702301025 +GTF2E2,7,0.021799029782414436,0.11325047165155411,0.9714487733595099,0.06231882709088085,9.719927787780762 +GTF2F1,4,0.036532629281282425,0.1469697803258896,0.952738312679295,0.03092331731670935,12.58302116394043 +GTF2F2,6,0.027250021696090698,0.12546859681606293,0.9639800869548392,0.1172057804536674,10.86745548248291 +GTF2H1,44,0.01032573264092207,0.07440542429685593,0.9863666656082513,0.42872356015737934,6.689672946929932 +GTF2H2,15,0.010582312941551208,0.07825874537229538,0.9859872144137589,-0.042565525057863554,6.772277355194092 +GTF2H2C,12,0.014512741006910801,0.09218475967645645,0.9816166293527465,0.17456201537748142,7.930840015411377 +GTF2H3,7,0.02077769860625267,0.11119431257247925,0.9727333392116517,0.07569046345414501,9.489496231079102 +GTF2H4,11,0.015180875547230244,0.09245796501636505,0.9795405909751699,0.007887539363404198,8.111345291137695 +GTF3A,9,0.016238471493124962,0.09714417159557343,0.9781405553591898,0.16106434959744947,8.389131546020508 +GTF3C1,16,0.008896281011402607,0.07201412320137024,0.9883976056975585,-0.05331320526100333,6.209386348724365 +GTF3C2,12,0.012264249846339226,0.08392924070358276,0.9834099340389303,0.12981860110111773,7.290627956390381 +GTF3C3,13,0.01195709127932787,0.08304788917303085,0.9845367281236743,-0.04468571073548913,7.198752403259277 +GTF3C4,13,0.01031023170799017,0.07736112177371979,0.9869110758369178,-0.05321038543851825,6.68464994430542 +GTF3C5,16,0.008670019917190075,0.07110963016748428,0.9892913715513795,-0.10267176890267869,6.129915714263916 +GTF3C6,25,0.005851477384567261,0.05716957896947861,0.992718334365172,-0.11396741456016544,5.035901069641113 +GTPBP4,7,0.02064298465847969,0.10916712135076523,0.9751163947487071,0.08670519250645978,9.458683967590332 +GUCD1,28,0.006697201635688543,0.0618940144777298,0.9909806209554235,0.27996687836593775,5.387547969818115 +GUK1,12,0.013750377111136913,0.08957544714212418,0.9831163328931115,0.09921152480453714,7.719723701477051 +H2AFX,14,0.009931303560733795,0.07658284157514572,0.9870025609160779,0.03646332448028059,6.5606608390808105 +H2AFZ,10,0.015121282078325748,0.0917343720793724,0.9814123600412038,-0.1820246480041999,8.09540843963623 +H3F3A,23,0.00671818433329463,0.06143064424395561,0.9912269837098534,-0.10236722244423481,5.3959808349609375 +HAMP,4,0.04101436212658882,0.15748190879821777,0.9470713694612032,0.18457686257437197,13.332525253295898 +HAPLN2,12,0.01155345793813467,0.08225616812705994,0.9849589549207408,-0.0197807295929837,7.076205730438232 +HARS,29,0.020244169980287552,0.09908517450094223,0.9737512721114855,0.5448803805619555,9.36686897277832 +HARS2,6,0.02125529944896698,0.11329111456871033,0.973933849176004,0.05518179546732678,9.597940444946289 +HAUS1,6,0.022318515926599503,0.11406555026769638,0.9712565479139623,0.030941131852274147,9.835062026977539 +HAUS3,17,0.008335145190358162,0.06966656446456909,0.9889063398584995,-0.02174704835953689,6.010367393493652 +HAUS4,19,0.008189680054783821,0.06915950775146484,0.9889642694849292,0.21235564929501136,5.957690238952637 +HAUS5,16,0.008460110984742641,0.07151255756616592,0.988832473034221,0.07102229763113946,6.055255889892578 +HAUS6,4,0.03380788862705231,0.14162234961986542,0.9553455581125954,0.22973202191322478,12.104683876037598 +HAUS7,24,0.00552332354709506,0.05741521716117859,0.9926769014197716,0.1081743579947874,4.892655849456787 +HAUS8,12,0.013646726496517658,0.08922939747571945,0.9817305997832713,0.0700010522529047,7.690573215484619 +HBS1L,15,0.020974798128008842,0.10659321397542953,0.9719930950986785,0.5438484385682518,9.534399032592773 +HCFC1,5,0.03406720981001854,0.14244459569454193,0.9586467172430658,0.11589558856879718,12.151020050048828 +HCRTR1,11,0.01109137013554573,0.08073251694440842,0.9856763825757632,0.012319860110282924,6.933253288269043 +HDAC3,6,0.01993408054113388,0.1083049550652504,0.9752423227615946,0.10492737368763673,9.294853210449219 +HDAC7,17,0.009366000071167946,0.07239099591970444,0.9872381088093426,0.18439823374470446,6.371204853057861 +HEATR1,24,0.01712121069431305,0.09288685768842697,0.9774552730096324,0.25360895778520304,8.6141357421875 +HECTD1,13,0.009860536083579063,0.07461223751306534,0.9876256371301443,-0.05419933652606998,6.5372443199157715 +HGS,10,0.016192637383937836,0.09689932316541672,0.9784561015060532,0.3301967902337502,8.377284049987793 +HINFP,16,0.013455445878207684,0.08480831235647202,0.9816715610487045,0.3024525057449657,7.6364850997924805 +HIRA,34,0.00404377281665802,0.04890350252389908,0.9946822646438183,0.10676812728192002,4.18637228012085 +HIST1H2AB,9,0.015030794776976109,0.0945170521736145,0.9806356950617698,-0.012698717468956525,8.071149826049805 +HIST1H2AE,13,0.010350250639021397,0.07860191911458969,0.9863928447339546,0.02804930256213004,6.697610378265381 +HIST1H2AI,5,0.027892038226127625,0.12810319662094116,0.9640465063084176,-0.0588853570776015,10.994729995727539 +HIST1H2BB,13,0.010499895550310612,0.0779101550579071,0.9864450565209751,-0.027953240903542412,6.745854377746582 +HIST1H2BC,7,0.01942032016813755,0.10646741092205048,0.9750766606649033,-0.04156440511042015,9.174293518066406 +HIST1H2BE,5,0.025289403274655342,0.12283959239721298,0.9675339844004592,0.04875396335607655,10.469205856323242 +HIST1H2BJ,5,0.029505250975489616,0.13318923115730286,0.9632110674228507,0.041448770517167874,11.308216094970703 +HIST1H2BL,21,0.006669478956609964,0.06246672198176384,0.9915799682338804,-0.020568572112967438,5.37638521194458 +HIST1H2BM,15,0.008999219164252281,0.07327080518007278,0.9884250439806462,0.011404858635647807,6.245208263397217 +HIST1H2BN,10,0.014805818907916546,0.09351189434528351,0.9798176631692064,0.21095272380407734,8.010519027709961 +HIST2H2AA3,6,0.020360354334115982,0.11026784777641296,0.9739017958645787,-0.0558788027089786,9.393709182739258 +HIST2H2AC,8,0.019347578287124634,0.10752934217453003,0.9749410452822795,0.02664896725496668,9.157095909118652 +HIST2H2BE,3,0.04087281599640846,0.1558927297592163,0.9500593844689991,0.013265648023871146,13.309499740600586 +HIST2H2BF,12,0.01157826092094183,0.08187491446733475,0.9851465736831778,-0.12952351628491987,7.083797454833984 +HIST2H3A,8,0.01676037535071373,0.09940962493419647,0.9783620141908387,0.06891019597576023,8.522879600524902 +HIST2H3D,19,0.007930661551654339,0.06835893541574478,0.9898576585356855,-0.14368265363417518,5.862720489501953 +HJURP,11,0.011441903188824654,0.0829482451081276,0.9849572943411009,0.1462185641732374,7.0419602394104 +HLA-C,5,0.02522527426481247,0.1213654950261116,0.9661655746099189,0.20695117001219154,10.455924034118652 +HMGA1,12,0.012922510504722595,0.08629966527223587,0.9830761226766112,0.10520547423766,7.4837260246276855 +HMGB1,24,0.0059020621702075005,0.058984700590372086,0.9922042851173608,0.014456476642266772,5.057621479034424 +HMGCR,16,0.008230009116232395,0.06927049160003662,0.9892150186716198,0.026516146464248037,5.972341537475586 +HMGCS1,23,0.006904957816004753,0.061972834169864655,0.9906665521372207,0.24714148349023218,5.4704742431640625 +HMGN2,10,0.01372512150555849,0.0893019288778305,0.9818802636149221,0.034271272433897465,7.712630748748779 +HNRNPC,34,0.011938790790736675,0.07683951407670975,0.9837280154628791,0.457791048567351,7.193240642547607 +HNRNPH1,16,0.01055989321321249,0.07830017060041428,0.9861816339777679,0.09986071469804084,6.765100002288818 +HNRNPK,3,0.03918837383389473,0.15397612750530243,0.9527213298562003,0.06680826634022048,13.032361030578613 +HNRNPL,8,0.01611519418656826,0.09711834043264389,0.979715247682621,0.12271576536707102,8.357227325439453 +HNRNPM,26,0.005495166406035423,0.0566675066947937,0.9926432708504018,0.22200043040002693,4.88016939163208 +HNRNPR,15,0.009688113816082478,0.07481309771537781,0.9871877468585806,-0.04071527262308012,6.479836463928223 +HNRNPU,10,0.015404694713652134,0.09286569803953171,0.9803516229392257,0.08592440535641489,8.170921325683594 +HOXC10,9,0.014846360310912132,0.09304600954055786,0.9807547919778477,-0.02615770362881893,8.021478652954102 +HS6ST1,17,0.008029946126043797,0.06835414469242096,0.9897288421061065,-0.10030473178324287,5.899303913116455 +HSCB,8,0.017337530851364136,0.10194975882768631,0.9778171665175851,0.08158820700645653,8.66838264465332 +HSD17B10,2,0.06604056060314178,0.19976946711540222,0.919634886572042,0.03187255988108168,16.91802978515625 +HSD17B12,14,0.012913153506815434,0.08663049340248108,0.9824015075877639,0.32371419448349625,7.481016159057617 +HSF1,13,0.011354303918778896,0.0808333158493042,0.9852414312163134,0.02670102172852899,7.014952182769775 +HSP90B1,12,0.011532466858625412,0.0816318616271019,0.9853756248217654,0.005406344040163441,7.069774150848389 +HSPA14,12,0.011139924637973309,0.08101926743984222,0.9861318092790239,-0.07984077556894756,6.94841194152832 +HSPA5,44,0.02117292210459709,0.09539660066366196,0.9721850478061791,0.49083039838788545,9.579323768615723 +HSPA8,25,0.01478803250938654,0.0846744179725647,0.981664347536502,0.36314946614368326,8.005705833435059 +HSPA9,42,0.022545794025063515,0.10061029344797134,0.9699380694847752,0.6183576413686246,9.88501262664795 +HSPD1,15,0.01018892228603363,0.07608485966920853,0.9869254588210739,0.11203440590863568,6.64520788192749 +HSPE1,3,0.05265326425433159,0.17559020221233368,0.9373333714863502,0.2533910520792584,15.1062650680542 +HTATSF1,9,0.01400853507220745,0.09122493863105774,0.9814209276748672,0.19821387174589558,7.791853904724121 +HUS1,20,0.007147008553147316,0.06368701905012131,0.9911190552926425,-0.07342928230546426,5.565530776977539 +HUWE1,10,0.01557548251003027,0.09633530676364899,0.9799447574233429,0.06166256467573925,8.216090202331543 +HYOU1,3,0.04944773018360138,0.1660958081483841,0.9360781885056882,0.17797521068182012,14.639209747314453 +HYPK,12,0.011814135126769543,0.08386193960905075,0.984585011011929,-0.014648431690432258,7.155589580535889 +IARS,6,0.02664090320467949,0.1254262775182724,0.9655346138178355,0.32724027912365,10.745309829711914 +IARS2,21,0.011179530993103981,0.07505250722169876,0.98537540103223,0.3819970818286405,6.960752964019775 +ICE1,43,0.005620355252176523,0.055615365505218506,0.9923852004069783,0.30894455846047975,4.935445308685303 +IFITM2,15,0.009837191551923752,0.07587996870279312,0.9871310119000967,-0.0212004005231721,6.529501438140869 +IFITM3,2,0.07694606482982635,0.21962904930114746,0.9110188756080178,0.08882231367192503,18.261550903320312 +IGBP1,28,0.008256668224930763,0.06820695847272873,0.9889502498488938,0.24013491209983745,5.982006549835205 +IK,29,0.00553952157497406,0.056598469614982605,0.9925210467295299,0.15701749509215726,4.899825096130371 +ILF2,5,0.02639743685722351,0.12383162975311279,0.969025142477757,-0.17424605547193353,10.696097373962402 +ILF3,14,0.01223701424896717,0.08342432230710983,0.9841233712524488,0.03258788333073192,7.282528400421143 +IMP3,2,0.05707695335149765,0.18688103556632996,0.9288590137460933,0.04903567386980402,15.728048324584961 +IMP4,17,0.020897600799798965,0.10464248061180115,0.9732676725668136,0.20914204321180596,9.516838073730469 +IMPDH2,7,0.02259064093232155,0.11426783353090286,0.9696311498454334,0.16947381714030976,9.8948392868042 +INCENP,6,0.02100558578968048,0.11106778681278229,0.972417421755289,0.05819604283037252,9.541394233703613 +ING3,7,0.020405415445566177,0.10971373319625854,0.9738180736915246,-0.03402605724759835,9.404098510742188 +INO80,15,0.012801049277186394,0.08410357683897018,0.9825339236281726,0.17296285171581496,7.44847297668457 +INO80B,11,0.015485145151615143,0.09466573596000671,0.9801023182221509,0.048453384782924463,8.192229270935059 +INTS1,21,0.007601887919008732,0.06631013751029968,0.9905730021218716,-0.08003651425793955,5.7399115562438965 +INTS2,36,0.03276711329817772,0.12317381799221039,0.9547656057032172,0.4411465142566883,11.916906356811523 +INTS3,10,0.014984923414885998,0.09428588300943375,0.9806549124356391,0.20522162152883258,8.058825492858887 +INTS4,5,0.0400933176279068,0.15209881961345673,0.9455299468848591,0.2648422055822063,13.181973457336426 +INTS5,7,0.03475870192050934,0.13690416514873505,0.9547663473001063,0.23334365108744357,12.273720741271973 +INTS6,14,0.011554228141903877,0.08207967877388,0.9843240478596278,0.09983321903567739,7.076441287994385 +INTS7,8,0.02511218935251236,0.11903635412454605,0.9664805636097129,0.14093961164171045,10.432459831237793 +INTS8,33,0.021459702402353287,0.10150585323572159,0.9703887833427186,0.377143226431992,9.643980026245117 +INTS9,23,0.018033841624855995,0.09633423388004303,0.9760065062286909,0.3845738529389972,8.840739250183105 +IPO11,7,0.0185525082051754,0.10542578250169754,0.9751148681502453,0.054160082308150434,8.966971397399902 +IPO13,5,0.034705065190792084,0.14024627208709717,0.9556450010968844,0.09860516550406799,12.264246940612793 +IPO7,7,0.020152932032942772,0.10945214331150055,0.9746106937889474,0.04541046657986182,9.345736503601074 +IPO9,18,0.00819077156484127,0.0684279277920723,0.9889572384617228,0.13575798704410472,5.95808744430542 +ISCA2,23,0.00578110758215189,0.05720645934343338,0.9925952248702997,0.030065153587438727,5.005528926849365 +ISCU,6,0.020297421142458916,0.109512098133564,0.9732834855093684,0.15510968127020386,9.379180908203125 +ISG20L2,8,0.016957184299826622,0.09995195269584656,0.978187743914308,0.2583356131306153,8.572773933410645 +ISY1,3,0.042470574378967285,0.1590655893087387,0.9471114496197067,-0.017666467312255647,13.567147254943848 +JAZF1,21,0.007513015531003475,0.06703387945890427,0.9906238732863911,0.054777254015597084,5.706260681152344 +JMJD6,20,0.006254678126424551,0.06079357862472534,0.9919649972247067,0.1129018611429539,5.206512928009033 +KANSL1,5,0.041534896939992905,0.1576451063156128,0.9511974945597107,0.08763945347435102,13.416863441467285 +KANSL2,16,0.016695337370038033,0.09653540700674057,0.9792264295710689,0.16901100016098702,8.506326675415039 +KANSL3,18,0.011033166199922562,0.07890331745147705,0.9865893030079463,0.12598715124769988,6.915038108825684 +KARS,6,0.024130774661898613,0.12064886093139648,0.9689582413775905,0.11032983818233541,10.226572036743164 +KAT5,12,0.010471549816429615,0.07824245095252991,0.9858983536788659,0.16089149851269083,6.7367424964904785 +KAT8,28,0.0069407387636601925,0.0625803992152214,0.9919028641797925,0.021373806099993557,5.4846296310424805 +KATNB1,18,0.007673821412026882,0.06623130291700363,0.9900201856049163,-0.08533868397850243,5.767004489898682 +KCMF1,16,0.008819942362606525,0.07189037650823593,0.9883547132608813,0.11006159558376322,6.182688236236572 +KCNA10,14,0.00976624060422182,0.07502730190753937,0.9870060702633626,0.04771300463606433,6.505911827087402 +KCTD10,7,0.020351344719529152,0.10811616480350494,0.9735503432886493,0.0462561416553951,9.391631126403809 +KDM2A,11,0.011069792322814465,0.07987451553344727,0.9856245497924799,0.0687241517901111,6.926505088806152 +KDM8,5,0.0290662944316864,0.13094420731067657,0.9627039451617471,0.1381063226158865,11.223784446716309 +KIAA1143,20,0.007292959839105606,0.06560260057449341,0.9905078596251766,0.19602240584329883,5.622071266174316 +KIF11,98,0.001892490778118372,0.03231515735387802,0.9975468316457307,-0.0032722314738492285,2.8639228343963623 +KIF14,12,0.00999927893280983,0.0769321471452713,0.9869518713851813,0.08115813937194931,6.583075046539307 +KIF18A,4,0.030797922983765602,0.13606740534305573,0.9625921575683165,-0.05472170223277934,11.553276062011719 +KIF18B,12,0.010381392203271389,0.07897651940584183,0.9860666178362292,0.08061181941202841,6.707679271697998 +KIF20A,9,0.01807579956948757,0.0998140349984169,0.9762461397896459,0.19681727996986484,8.851017951965332 +KIF23,8,0.017243310809135437,0.10021384805440903,0.9774602454738089,0.033061684386858255,8.644796371459961 +KIF4A,13,0.010021662339568138,0.07670405507087708,0.9873292192336639,-0.0502123188720788,6.5904388427734375 +KIN,16,0.012034980580210686,0.08337749540805817,0.9842806732486257,0.2732222367274927,7.222160339355469 +KLC2,12,0.01292530819773674,0.08710319548845291,0.9825402275465349,0.091315715039374,7.484536170959473 +KLF7,18,0.007395468186587095,0.06632378697395325,0.9902344253719725,0.06911507622803259,5.661445140838623 +KLHL11,13,0.00996676180511713,0.0764341801404953,0.9872331359175556,0.02720582581029564,6.572361946105957 +KPNA6,8,0.019917918369174004,0.10840775817632675,0.9738365196448926,-0.05613818476353977,9.291085243225098 +KPNB1,36,0.005998900625854731,0.056604500859975815,0.9921055704190996,0.1139465575253142,5.098944187164307 +KRI1,5,0.033512890338897705,0.1394975483417511,0.9558487328406207,0.041123098273523714,12.0517578125 +KRR1,45,0.009839968755841255,0.06901775300502777,0.9869942410989854,0.1268554424837781,6.530423164367676 +KRT10,18,0.008058064617216587,0.06925678253173828,0.9897575066454267,-0.018526548258761464,5.909623622894287 +KRT17,9,0.01580159179866314,0.09627882391214371,0.978852957327589,0.1391920665787456,8.275511741638184 +KRT8,34,0.0044622584246098995,0.05017750710248947,0.9940691841914223,0.0746007301433432,4.397661209106445 +KRTAP4-2,13,0.010248514823615551,0.07810571789741516,0.986671292796244,0.11548792972310531,6.664612770080566 +KRTAP4-7,17,0.008892592042684555,0.07159404456615448,0.9883344855137937,0.02448830052364577,6.208098888397217 +KTI12,5,0.02504987083375454,0.12141882628202438,0.9681653226912267,0.21353561443049723,10.419507026672363 +KXD1,10,0.012819220311939716,0.08611461520195007,0.9834256794262295,0.04009572361235335,7.4537577629089355 +LAMTOR2,6,0.036917004734277725,0.14624984562397003,0.9504796329438304,0.2638556151014643,12.649043083190918 +LAMTOR3,4,0.027533184736967087,0.12952126562595367,0.9668154348299053,-0.0796471183552832,10.923772811889648 +LAMTOR4,13,0.011472105979919434,0.08103642612695694,0.9847773124084612,0.07616748479086978,7.051248550415039 +LARS,10,0.022954799234867096,0.11117424070835114,0.9686957338904832,0.45657871097183156,9.974271774291992 +LARS2,15,0.009045989252626896,0.07365062087774277,0.9880096180329913,0.03600284567320201,6.261415004730225 +LAS1L,9,0.014445742592215538,0.09271175414323807,0.9812374695667316,0.12784390603119528,7.912512302398682 +LCE1C,15,0.010403349995613098,0.07773938030004501,0.9861610897513325,0.03926408121510724,6.714768886566162 +LCE1E,13,0.00963620562106371,0.0751495435833931,0.9879293148774694,-0.030219876318351078,6.462453842163086 +LCN10,8,0.01688634231686592,0.10010015964508057,0.9791720564827996,-0.045304889529948454,8.55484676361084 +LENG8,12,0.011418885551393032,0.08254492282867432,0.9852697969163345,0.0977239601877169,7.034873962402344 +LETM1,13,0.012537266127765179,0.08538898825645447,0.9838183713071731,0.266368359150924,7.371330261230469 +LIMS1,28,0.0051738545298576355,0.055221181362867355,0.9932499953522271,0.07028768076668747,4.735344409942627 +LIN52,2,0.04781836271286011,0.17136269807815552,0.9445142360027985,0.10343525574262402,14.39599895477295 +LIN54,30,0.004490343388170004,0.05141890048980713,0.9939827362382492,0.1487507251623844,4.4114789962768555 +LONP1,13,0.01137382909655571,0.08061245828866959,0.9853318003691564,0.17246460808222414,7.0209808349609375 +LPA,12,0.010754909366369247,0.07920535653829575,0.9861843215438314,-0.06128733416134816,6.827281951904297 +LRPPRC,11,0.017411712557077408,0.09235648810863495,0.9789770741914106,0.15536679970029027,8.686907768249512 +LRR1,12,0.01530132070183754,0.09348466247320175,0.9795726545299271,0.08006860523521445,8.14345932006836 +LRRC37A3,12,0.010921373963356018,0.0807933509349823,0.9859476548262883,-0.09906870206669002,6.879915237426758 +LRRC37B,11,0.012805919162929058,0.08624805510044098,0.9829917957943837,0.027524577752521314,7.449889183044434 +LSG1,11,0.014017919078469276,0.08869966864585876,0.9817369023591859,0.16623778695648336,7.794463634490967 +LSM10,13,0.012292727828025818,0.08162626624107361,0.9841636100670549,0.035929367732149095,7.299088001251221 +LSM11,13,0.014791225083172321,0.08828195929527283,0.9797386530575743,0.19648149187675798,8.006570816040039 +LSM12,7,0.01926179975271225,0.10700049996376038,0.9755578959121775,-0.052179493155312055,9.136773109436035 +LSM2,35,0.01502703782171011,0.08767439424991608,0.9809689655921497,0.3601719795958177,8.070140838623047 +LSM3,8,0.026805954053997993,0.12262456119060516,0.966115983936322,0.27087143663392105,10.778543472290039 +LSM4,9,0.022020334377884865,0.11252232640981674,0.9734651946404063,0.1287738044720081,9.76914119720459 +LSM5,20,0.008203144185245037,0.06873655319213867,0.9893712589081665,0.09296064474316096,5.96258544921875 +LSM6,46,0.01363543700426817,0.08226358890533447,0.9820652018909125,0.38653245466616504,7.68739128112793 +LSM7,14,0.013951736502349377,0.08974774181842804,0.9818742562451144,0.1507029881813273,7.7760419845581055 +LST1,15,0.009173606522381306,0.07307266443967819,0.9880858118579529,-0.05904208252374743,6.305427074432373 +LTB4R2,19,0.007427150383591652,0.06601578742265701,0.9905807593451136,-0.1247291332437641,5.673558712005615 +LTBP3,15,0.008915121667087078,0.07266227900981903,0.9885387504017114,-0.03874952081313155,6.215958118438721 +LTV1,32,0.009828572161495686,0.07136053591966629,0.9870326721345758,0.22157020478037726,6.526639938354492 +LUC7L3,5,0.025060715153813362,0.1205303817987442,0.9682222388992223,0.018218073914549385,10.421762466430664 +LURAP1,7,0.018091773614287376,0.10417509824037552,0.9776344774731128,-0.0596338709710951,8.854928016662598 +LYRM4,7,0.01879260689020157,0.10581813007593155,0.9760628598292256,-0.005859926607295678,9.024807929992676 +MAD2L1,16,0.009132214821875095,0.07251457870006561,0.9878189564243073,0.23038635730538054,6.2911858558654785 +MAD2L2,20,0.006833931431174278,0.06330748647451401,0.9908794376051709,0.04506846790002365,5.44226598739624 +MAGOH,24,0.009203929454088211,0.06958109140396118,0.9874828635096369,0.31180869397496486,6.315839767456055 +MAK16,22,0.009227748028934002,0.07110608369112015,0.9878955560827002,0.24807658077011346,6.3240065574646 +MANF,14,0.01030069962143898,0.07748886197805405,0.9865588945237808,-0.0998247225939593,6.681559085845947 +MARS,9,0.019022386521100998,0.10472387820482254,0.9749577791679226,0.34952413483687317,9.079813957214355 +MARS2,7,0.01590651459991932,0.09816198796033859,0.9791940353152974,0.11081826380087838,8.30294132232666 +MASTL,11,0.01595633290708065,0.09536483883857727,0.9788768839093757,0.2194524277415036,8.315933227539062 +MAT2A,5,0.03459211811423302,0.14197483658790588,0.9561847560887589,0.10691239104650009,12.244274139404297 +MAU2,6,0.022046906873583794,0.114546038210392,0.9711883839397114,0.10734139322442822,9.77503490447998 +MAX,18,0.02572331577539444,0.11363658308982849,0.9679929774899305,0.3120964627088742,10.558638572692871 +MBTPS1,7,0.018394911661744118,0.10478615015745163,0.9765314240678359,0.16457816247150614,8.928804397583008 +MBTPS2,22,0.00775834359228611,0.06509274244308472,0.9898177016330695,0.075107371135592,5.798677444458008 +MCL1,8,0.015545634552836418,0.09717462956905365,0.9803988548802947,-0.03814247784351515,8.208213806152344 +MCM10,22,0.00898900255560875,0.06837040185928345,0.9876627910235645,0.2665299153297223,6.241661071777344 +MCM2,13,0.012252742424607277,0.08337622135877609,0.9833964913214351,0.22064560531275448,7.287207126617432 +MCM3,13,0.010767404921352863,0.0785067155957222,0.9856630111273271,0.14996573853017944,6.831246852874756 +MCM3AP,11,0.01451130211353302,0.09227349609136581,0.9800743071698025,0.13990658395222277,7.930446147918701 +MCM4,8,0.02352748066186905,0.11594324558973312,0.9677451579888064,0.10459748168052359,10.097926139831543 +MCM5,9,0.015144949778914452,0.09418216347694397,0.9803734331135501,0.1436797323577801,8.101741790771484 +MCM6,10,0.01339635532349348,0.08737397938966751,0.9831025150985241,0.006824740378883383,7.619698524475098 +MCM7,16,0.007983057759702206,0.06843750923871994,0.9894296172419528,0.029622156073731726,5.882055282592773 +MCMBP,15,0.0116815734654665,0.08106572926044464,0.9847308216376297,0.11908331054491293,7.115331649780273 +MCRS1,7,0.016003012657165527,0.09843700379133224,0.9806911650373032,0.009248285761371985,8.328088760375977 +MDN1,9,0.015235327184200287,0.09492332488298416,0.9807007894271709,0.1834125845424475,8.125879287719727 +MED1,26,0.029972901567816734,0.12471102178096771,0.9612920469705679,0.39902658437868066,11.397480010986328 +MED10,51,0.010214902460575104,0.07276884466409683,0.9862779220016832,0.3366268630590571,6.653675079345703 +MED11,23,0.01835942082107067,0.09915758669376373,0.9753709169256224,0.2670922508376708,8.920186996459961 +MED12,58,0.034063439816236496,0.1270913928747177,0.9552016104119483,0.48246575448645684,12.150347709655762 +MED14,11,0.01897374354302883,0.10438603907823563,0.9739512761463512,0.18347067405293296,9.068197250366211 +MED17,15,0.03170185536146164,0.13027778267860413,0.9568622959132072,0.34134219922289727,11.721597671508789 +MED18,2,0.09638049453496933,0.24330376088619232,0.8780954633823049,0.043424789435786745,20.438030242919922 +MED19,55,0.01911604031920433,0.09698490798473358,0.9760742845397914,0.42450524324698324,9.102137565612793 +MED20,8,0.038675256073474884,0.1453559696674347,0.9465300496240486,0.15286476850340938,12.946758270263672 +MED21,7,0.04996257647871971,0.16805720329284668,0.9324226415650133,0.36745150662687154,14.715224266052246 +MED22,29,0.01662961207330227,0.09124062955379486,0.9769805950838385,0.4197240123303349,8.4895658493042 +MED26,11,0.014195018447935581,0.09193340688943863,0.9811675349813906,0.05002188160344468,7.843545913696289 +MED27,2,0.12759943306446075,0.27961182594299316,0.8444316678927161,0.08130948040666265,23.51629066467285 +MED28,8,0.037692468613386154,0.14600738883018494,0.9482557336151133,0.2672363709277584,12.781203269958496 +MED29,10,0.02501119114458561,0.11956902593374252,0.9662478802378622,0.22135885283110848,10.411459922790527 +MED30,29,0.03401588276028633,0.1309356987476349,0.955460853238814,0.42025522792533604,12.141861915588379 +MED31,4,0.03639828413724899,0.14911721646785736,0.951380669801087,0.057084416762874055,12.559863090515137 +MED4,6,0.04267347604036331,0.15849320590496063,0.9421430802964755,0.2403819837674775,13.599515914916992 +MED6,24,0.02237989567220211,0.10919572412967682,0.9713860464086863,0.4577042169027081,9.848576545715332 +MED7,27,0.028169091790914536,0.12086549401283264,0.9638733246036845,0.43674589975737976,11.049201011657715 +MED8,11,0.019397800788283348,0.10635880380868912,0.973300943049093,0.14831960541868736,9.168973922729492 +MED9,12,0.06027212366461754,0.17719219624996185,0.9198795192192324,0.45801769007494086,16.162282943725586 +MEPCE,12,0.009679608978331089,0.07604091614484787,0.9876923939362519,-0.029131040197883415,6.476991653442383 +METAP1,10,0.013241248205304146,0.08874978125095367,0.9828070432499555,-0.07657395943986647,7.575458526611328 +METAP2,16,0.013060104101896286,0.08358932286500931,0.9826218729060202,-0.007754737711508668,7.523462772369385 +METTL1,21,0.006610690150409937,0.06182774156332016,0.9911432307843323,0.07241955473134375,5.35263729095459 +METTL14,12,0.014188219793140888,0.09103338420391083,0.9805712363939274,0.08509086151958017,7.841667652130127 +METTL16,4,0.044459182769060135,0.16426198184490204,0.945563640576408,0.0352165550349402,13.881141662597656 +METTL17,12,0.011881301179528236,0.08268190920352936,0.9845474634501858,-0.019917663795918684,7.175900936126709 +METTL23,12,0.012135935947299004,0.08470761775970459,0.9847107934041507,-0.009218895650388885,7.2523884773254395 +METTL3,14,0.01223427802324295,0.08371011912822723,0.9839228007308903,-0.07911838503317453,7.28171443939209 +MEX3A,17,0.00863012857735157,0.07041279226541519,0.989031990899354,-0.006587976396447987,6.115797519683838 +MFAP1,19,0.0152077442035079,0.09171224385499954,0.9793081291505747,0.3337993669296465,8.118518829345703 +MFN2,13,0.010661293752491474,0.07878796011209488,0.98591598384159,0.04303289750590292,6.7975029945373535 +MICOS10,8,0.017163705080747604,0.0999266505241394,0.9775058439556289,0.1988542862777343,8.624818801879883 +MIOS,17,0.00827751960605383,0.06911737471818924,0.9889552012163952,0.16246142962784957,5.989555358886719 +MIPEP,9,0.01685205101966858,0.10054682940244675,0.9778612389797424,0.19022663511277763,8.546156883239746 +MIS12,8,0.01764395460486412,0.10353396087884903,0.9767157445079546,0.0979029739186395,8.744649887084961 +MIS18A,13,0.01022421196103096,0.07698401808738708,0.9871536098385476,0.010359515223383978,6.656705379486084 +MIS18BP1,10,0.013954303227365017,0.09064505994319916,0.9819762145149433,-0.14079542605977854,7.77675724029541 +MKRN1,16,0.009220526553690434,0.07323890924453735,0.9874646548729116,0.1604430889147807,6.321531772613525 +MLST8,8,0.018266363069415092,0.10472392290830612,0.9753291052555634,0.12828413410448652,8.897551536560059 +MMGT1,19,0.006809407379478216,0.06331348419189453,0.9914125236746163,-0.06961029344117321,5.432492256164551 +MMS19,13,0.0109066441655159,0.08067352324724197,0.9859871027599105,0.009155057264508725,6.875274181365967 +MMS22L,5,0.022624248638749123,0.11660641431808472,0.9700512147449993,0.19535891120468235,9.902195930480957 +MNAT1,9,0.02009321004152298,0.10869564116001129,0.9730201811893808,0.15804739950014635,9.331879615783691 +MOB4,12,0.015038286335766315,0.09370537102222443,0.9798907687318685,0.2979738631016924,8.073161125183105 +MOCS3,12,0.010533546097576618,0.07865463197231293,0.9866979700616063,-0.03714382937211155,6.756655216217041 +MOG,13,0.010988868772983551,0.08038674294948578,0.9857817076148079,0.013808111770350651,6.90114164352417 +MOK,13,0.011272373609244823,0.08228267729282379,0.9853585471179377,0.05982847612475397,6.989596843719482 +MPHOSPH10,5,0.03954078257083893,0.15014603734016418,0.9488251000437324,0.0385669513111428,13.090826034545898 +MRGBP,8,0.02010534144937992,0.10735323280096054,0.9741766544197099,-0.02231997620402867,9.334695816040039 +MRPL1,17,0.007385685108602047,0.06556018441915512,0.9908132100592568,0.05727593413966035,5.657698631286621 +MRPL10,10,0.016398822888731956,0.09842013567686081,0.9783477190331278,0.21370051511259477,8.430450439453125 +MRPL13,8,0.02434147521853447,0.11925961077213287,0.9680751895837275,0.2923504601504712,10.271122932434082 +MRPL14,21,0.007146158721297979,0.06267153471708298,0.9906999484379173,0.1985005780476121,5.565200328826904 +MRPL15,4,0.03427392244338989,0.1414909064769745,0.9572844586375179,0.09780662244771264,12.18782901763916 +MRPL16,9,0.01532241702079773,0.09596595913171768,0.9803684671425686,0.2130663152135277,8.149070739746094 +MRPL17,10,0.015202865935862064,0.09500511735677719,0.9810330182850476,0.14238846704705396,8.117218017578125 +MRPL18,18,0.007698363624513149,0.06620750576257706,0.9899619160351848,0.23469555415928126,5.776219367980957 +MRPL19,19,0.006979506462812424,0.06324304640293121,0.9910969409377903,0.2918378236908635,5.49992561340332 +MRPL2,14,0.010654632933437824,0.07854000478982925,0.9856375916980115,0.23478636718589152,6.795379161834717 +MRPL20,12,0.013727974146604538,0.08815452456474304,0.9822224429226324,0.12124155623841353,7.713432788848877 +MRPL21,2,0.0736246183514595,0.21345016360282898,0.9141272805070196,-0.016262766311117876,17.863065719604492 +MRPL22,9,0.014689469709992409,0.09197799861431122,0.9803210034723716,0.15314956216640496,7.978982448577881 +MRPL24,20,0.007179735228419304,0.06411820650100708,0.990285594453073,0.16073754609022425,5.578258991241455 +MRPL27,10,0.01328505203127861,0.08706159144639969,0.9824819883876538,0.13875939020639588,7.587978839874268 +MRPL28,5,0.0292630847543478,0.1329692304134369,0.9638906081502431,0.08521472033387832,11.261713981628418 +MRPL3,17,0.009142840281128883,0.07278425991535187,0.9884048607862572,0.14444997311781999,6.294844627380371 +MRPL32,6,0.02059694193303585,0.11023865640163422,0.9750287491485088,0.16355855636796612,9.448128700256348 +MRPL33,15,0.009711754508316517,0.07484322041273117,0.9878850428488515,0.24885159765814271,6.487738132476807 +MRPL34,2,0.07611200958490372,0.21628183126449585,0.9080470247902911,-0.03825477304045435,18.162307739257812 +MRPL35,17,0.008987816981971264,0.0718107670545578,0.9884864845210926,0.2800760592959581,6.2412495613098145 +MRPL36,19,0.008366026915609837,0.06902505457401276,0.988495870661162,0.2574074308231378,6.021491527557373 +MRPL37,11,0.013707267120480537,0.09008707106113434,0.9826040016223605,0.2006410417147261,7.707612991333008 +MRPL38,15,0.009243834763765335,0.07335125654935837,0.987797061633534,0.08425368823416346,6.329516410827637 +MRPL39,16,0.009681636467576027,0.07363109290599823,0.9873939715176357,0.29819942088102575,6.477670669555664 +MRPL4,13,0.009320683777332306,0.07329989969730377,0.9880727205240316,0.13638003579166247,6.355772495269775 +MRPL41,7,0.019625594839453697,0.10760899633169174,0.9751471988874628,0.13369603796924165,9.222652435302734 +MRPL43,9,0.01528166513890028,0.09404696524143219,0.9797192961251564,0.14832478100351834,8.138226509094238 +MRPL44,8,0.015559724532067776,0.09664232283830643,0.9802867791037867,0.03613058021177161,8.211933135986328 +MRPL45,18,0.008223876357078552,0.06949871778488159,0.9893499955925896,0.032951222011989974,5.970115661621094 +MRPL49,4,0.043907128274440765,0.1615208238363266,0.9396850513258128,0.20694192631855088,13.794690132141113 +MRPL50,17,0.007329667918384075,0.06492210179567337,0.9906012330720674,0.04429485409034882,5.636202812194824 +MRPL51,11,0.015680020675063133,0.09634947776794434,0.9797584118857436,0.20496265329238267,8.243616104125977 +MRPL53,16,0.008913933299481869,0.07184092700481415,0.98804559125538,0.2937308553686152,6.2155442237854 +MRPL54,14,0.009287758730351925,0.07332417368888855,0.98819904207449,-0.023115776793984234,6.344536304473877 +MRPL55,17,0.008202415890991688,0.0679766908288002,0.9895515706035731,0.1670142985167782,5.962320804595947 +MRPL9,12,0.010943911969661713,0.08025944232940674,0.9857072521862195,0.16523658376136816,6.88701057434082 +MRPS10,12,0.012328377924859524,0.08514335751533508,0.9843165545852001,0.21087298691862344,7.309663772583008 +MRPS11,9,0.01608082465827465,0.09738572686910629,0.9798948486145811,0.10977970592692679,8.348310470581055 +MRPS12,5,0.028905965387821198,0.13265454769134521,0.9634379805119611,0.0953655770383844,11.192785263061523 +MRPS14,9,0.01875728741288185,0.10089771449565887,0.9773811356268279,0.19303844199692746,9.01632308959961 +MRPS16,4,0.033698033541440964,0.1421300321817398,0.9582330948029444,0.11637058729463692,12.085001945495605 +MRPS18A,12,0.011586030013859272,0.08214224874973297,0.9847378555184766,0.16123084763469897,7.0861735343933105 +MRPS18B,6,0.0279818307608366,0.12894664704799652,0.9640297416623083,0.02477388977311056,11.01241397857666 +MRPS2,11,0.011896549724042416,0.08433786034584045,0.9849762432920335,0.05399370951850631,7.180504322052002 +MRPS21,14,0.00961005873978138,0.07422832399606705,0.987630201188464,0.20749447065090193,6.453680992126465 +MRPS22,17,0.009461862966418266,0.07553018629550934,0.9878786432050898,0.15181413288523332,6.403726100921631 +MRPS23,12,0.01359582133591175,0.08873149752616882,0.9825083241156639,0.06867930225169742,7.676215648651123 +MRPS24,15,0.01030841562896967,0.07786720246076584,0.9867262127827077,0.16974456003332106,6.684061050415039 +MRPS25,10,0.014253550209105015,0.09136161208152771,0.9820775441288401,-0.043775309625045805,7.8597002029418945 +MRPS26,12,0.011524365283548832,0.08248192816972733,0.9851591179751712,0.22729874205794412,7.067290306091309 +MRPS27,8,0.01624728925526142,0.09797784686088562,0.97850697073896,0.16943613258882742,8.391409873962402 +MRPS28,16,0.008937749080359936,0.07171326875686646,0.9883091601003257,0.1770996416191512,6.223841667175293 +MRPS31,10,0.01356751099228859,0.08884388208389282,0.9827646171417034,0.05331011621523731,7.668219566345215 +MRPS34,14,0.01468406431376934,0.09168397635221481,0.9804108712533106,0.23487282572210566,7.977514266967773 +MRPS35,20,0.007112453691661358,0.06482820957899094,0.9907460580293905,0.30635228234913586,5.552060604095459 +MRPS5,8,0.016371026635169983,0.09734245389699936,0.9795980375032601,0.0014312983942619283,8.42330265045166 +MRPS6,15,0.008623701520264149,0.0710870772600174,0.9890233518147743,0.09405342838717583,6.113519668579102 +MRPS7,13,0.010201051831245422,0.07789528369903564,0.9870444808196449,0.1101301411143039,6.649162292480469 +MRPS9,7,0.0257341880351305,0.12283416837453842,0.9664905594369124,0.31112095652759436,10.560869216918945 +MRTO4,9,0.01421847753226757,0.09185422211885452,0.9823647995686084,0.05483442124014994,7.850024223327637 +MSRB1,21,0.007686338387429714,0.06619644165039062,0.989738836708746,-0.03384911321050534,5.7717061042785645 +MST1,14,0.00999054778367281,0.07605443894863129,0.9868422736303272,-0.0220373659586324,6.5802001953125 +MSTO1,10,0.014084101654589176,0.09142713993787766,0.9818445795170584,-0.09684272571500253,7.812841415405273 +MT2A,11,0.01335837785154581,0.08920414000749588,0.9817484004155758,0.11277244549799165,7.608890056610107 +MTBP,19,0.007601885590702295,0.06581626087427139,0.9899491015012535,0.1502537309396117,5.73991060256958 +MTG2,14,0.009741203859448433,0.07547922432422638,0.9872955113363444,0.003309596574126609,6.497567176818848 +MTOR,10,0.023290887475013733,0.11462382972240448,0.9706859190931458,0.3351094456955832,10.047024726867676 +MTPAP,9,0.020267946645617485,0.10253114998340607,0.9756502503170768,0.10699665646182775,9.372367858886719 +MTRNR2L1,5,0.035640716552734375,0.1442704051733017,0.9536174972349981,0.04514096435539733,12.42846965789795 +MVD,11,0.013952444307506084,0.08890728652477264,0.9811569846378945,0.27543500312146924,7.776238918304443 +MVK,13,0.011509086936712265,0.08154777437448502,0.9846140849114988,0.11352686336368463,7.062604904174805 +MYBBP1A,11,0.01673717238008976,0.09822538495063782,0.9773148629032656,0.3035937678549254,8.516977310180664 +MYBL2,10,0.01571665145456791,0.09605798870325089,0.9794256138864906,0.18596001428859402,8.253239631652832 +MYC,8,0.018576618283987045,0.10434643924236298,0.9765536805755783,-0.015190122102988646,8.972796440124512 +MYCBP,21,0.006440266966819763,0.060801971703767776,0.9914384009999604,0.06057574565075213,5.283192157745361 +MYO1H,13,0.011252662166953087,0.08155252784490585,0.9854352191699335,-0.028054266636970547,6.983482837677002 +MZF1,13,0.011602119542658329,0.08269316703081131,0.9844939660524605,-0.046389903421904875,7.091091632843018 +MZT1,45,0.004222697112709284,0.048088110983371735,0.9942369406873095,0.02549875661467498,4.277986526489258 +N6AMT1,14,0.011487196199595928,0.08159368485212326,0.9860263530072944,-0.1440157055775048,7.05588436126709 +NAA10,13,0.01190988626331091,0.08393044024705887,0.9837512759009528,0.03910119997154708,7.184528350830078 +NAA15,13,0.013320645317435265,0.08820692449808121,0.9819068700929073,0.10906701858021563,7.5981364250183105 +NAA20,18,0.008428064174950123,0.06884434819221497,0.9886388093870542,0.1083192758710501,6.043776035308838 +NAA25,17,0.011044220998883247,0.07969526946544647,0.9848437372928658,0.17413488592770673,6.9185004234313965 +NAA35,20,0.007530312519520521,0.06601398438215256,0.9898633635242711,0.13790814180082983,5.712825298309326 +NAA38,37,0.01239387784153223,0.07943154871463776,0.983767410237563,0.3439168504870265,7.329056262969971 +NAA50,11,0.015589469112455845,0.09391438215970993,0.9798373204152567,0.05389164945994777,8.219778060913086 +NACA,36,0.012303147464990616,0.07759325951337814,0.9832319381284851,0.4141101487153154,7.302181243896484 +NAE1,9,0.01680014468729496,0.09844750165939331,0.9784674879815866,-0.07795702688271385,8.532984733581543 +NAF1,10,0.014903881587088108,0.09313730895519257,0.9809115112144163,-0.06889274401045467,8.037002563476562 +NAGLU,27,0.005035752430558205,0.054847221821546555,0.9933468070894332,0.0790623971382157,4.671718120574951 +NAPA,1,0.13226284086704254,0.2891343832015991,0.8595081189023184,0.010925136207404856,23.942161560058594 +NAPG,6,0.02441546320915222,0.11921460181474686,0.9683273993782889,0.06373343131336173,10.286720275878906 +NARS,17,0.012839012779295444,0.08298629522323608,0.9822816373581373,0.44708938881180227,7.459509372711182 +NASP,19,0.007690734229981899,0.06574700027704239,0.9897466185216369,0.08821938523119055,5.7733564376831055 +NAT10,3,0.058152880519628525,0.18547244369983673,0.926856840821526,0.07174848897361574,15.87559700012207 +NBAS,10,0.01414565835148096,0.09136859327554703,0.9810496461363316,0.17136733295404355,7.829896450042725 +NBEAL1,10,0.0125093599781394,0.08510647714138031,0.9836436677843388,0.12184799686261863,7.363121509552002 +NBPF12,15,0.008460813201963902,0.07000339031219482,0.989558345103443,-0.057443552813256596,6.055506706237793 +NBPF15,6,0.01780930534005165,0.10050569474697113,0.978358589217772,-0.07107737215507272,8.785530090332031 +NBPF3,14,0.009413493797183037,0.07559888809919357,0.9874941154770391,0.03953424957594487,6.387337684631348 +NCAPD2,5,0.03056664578616619,0.13388164341449738,0.960302189134326,0.10437699975531399,11.509815216064453 +NCAPD3,8,0.01840999349951744,0.10530436038970947,0.9760408602234539,0.06943170142114605,8.932463645935059 +NCAPG,8,0.01678600162267685,0.09923260658979416,0.9771214767469326,0.15320229192574342,8.52939224243164 +NCAPG2,9,0.018379611894488335,0.1043354868888855,0.9754023772319292,0.015037830887551729,8.925089836120605 +NCAPH,7,0.019024604931473732,0.10590478032827377,0.9749347049322384,0.24834651897879412,9.080343246459961 +NCAPH2,15,0.00962549913674593,0.0744146928191185,0.9875685382617785,-0.0018964288267550099,6.458863258361816 +NCBP1,6,0.036935046315193176,0.14462867379188538,0.9510118237926689,-0.051883710009309324,12.652134895324707 +NCBP2,99,0.009729360230267048,0.06457478553056717,0.9866323064034688,0.05625197715752644,6.493615627288818 +NCKAP1,7,0.027193913236260414,0.1260858178138733,0.9633098172698935,0.15096194714474576,10.85626220703125 +NCL,30,0.005138306878507137,0.0538058839738369,0.9937501952557866,-0.07598742675998703,4.719048976898193 +NCOA4,14,0.01044030487537384,0.07837454974651337,0.9862657100557995,0.12142310684218686,6.726684093475342 +NDC1,4,0.04808560013771057,0.16646747291088104,0.9375875430545952,-0.036186529319651245,14.436169624328613 +NDC80,22,0.008330806158483028,0.0701291561126709,0.9885697255166057,0.3607065984597712,6.008802890777588 +NDOR1,4,0.062245409935712814,0.19233092665672302,0.9219123039523706,0.15988840016372594,16.424724578857422 +NDUFA11,8,0.01751535013318062,0.10188766568899155,0.9769794789291081,0.07602655656293654,8.712721824645996 +NDUFA2,18,0.013679527677595615,0.08712544292211533,0.9815219005915375,0.356839323047066,7.699810028076172 +NDUFA4,8,0.018392499536275864,0.10155096650123596,0.9770496691615213,0.027662727193135356,8.928218841552734 +NDUFA6,17,0.00739377411082387,0.06608310341835022,0.990270980995739,0.17841179932829132,5.660796642303467 +NDUFA8,11,0.011637238785624504,0.08250769972801208,0.9846442735847845,0.06876997477783817,7.101816177368164 +NDUFAB1,16,0.008963209576904774,0.07131404429674149,0.9882280530257556,0.12582665436109172,6.232699871063232 +NDUFAF3,8,0.01760237291455269,0.10308421403169632,0.9781098415425932,0.0963743429832039,8.734338760375977 +NDUFAF7,2,0.05755365267395973,0.1893901377916336,0.931823480352583,-0.031161023579860204,15.793590545654297 +NDUFB10,16,0.009597024880349636,0.07428848743438721,0.9874568704166361,0.1923875802641333,6.449303150177002 +NDUFB3,14,0.009074839763343334,0.07193045318126678,0.988492402122821,0.0307484757420486,6.271391868591309 +NDUFB4,16,0.008751939050853252,0.07170978933572769,0.9886447854157024,0.07822817546198049,6.158806800842285 +NDUFB6,19,0.007526406552642584,0.06619477272033691,0.9903840569001355,0.010145739865764848,5.711343765258789 +NDUFB7,8,0.014185433275997639,0.09123633801937103,0.9822872373717558,-0.050878649614094384,7.840897083282471 +NDUFB8,20,0.007309913169592619,0.06486491858959198,0.9905796011765826,0.04480758419110249,5.628602027893066 +NDUFS3,11,0.012362656183540821,0.08569154143333435,0.9831836877763326,0.10102235328633145,7.319819450378418 +NDUFS5,19,0.007229640614241362,0.0649433508515358,0.9903881832977295,0.035105299512095865,5.597611904144287 +NEDD1,14,0.009868640452623367,0.07550612092018127,0.9874940750101565,-0.11761138538624183,6.5399298667907715 +NEDD8,19,0.015672938898205757,0.0910838171839714,0.979131549656515,0.1640488402923116,8.241754531860352 +NELFA,9,0.02241036295890808,0.11303185671567917,0.9698929345808536,0.15144644825588804,9.855278015136719 +NELFB,14,0.012107877992093563,0.08379004895687103,0.9833079482644971,0.13881671475315607,7.2440009117126465 +NELFCD,20,0.00905668269842863,0.071004718542099,0.9876946452458496,-0.09884385889198766,6.265114784240723 +NELFE,8,0.016989169642329216,0.09986661374568939,0.9783093044025492,-0.011278057149387982,8.580854415893555 +NEMF,7,0.017698949202895164,0.10101456940174103,0.979047707919659,-0.033691245495043314,8.75826644897461 +NEPRO,4,0.04418681189417839,0.15878227353096008,0.9440665618250179,0.07881112178531588,13.838556289672852 +NFATC2IP,12,0.010499066673219204,0.07918643951416016,0.9872573004044689,-0.10187596698273581,6.7455878257751465 +NFKBIE,5,0.03246159851551056,0.1377064734697342,0.9589014142333226,-0.0681266492733019,11.861221313476562 +NFRKB,23,0.008130490779876709,0.06782524287700653,0.9890529395914143,-0.0005618113155594173,5.936122894287109 +NFS1,8,0.020402556285262108,0.10977965593338013,0.973246352349448,0.1145514730652483,9.40343952178955 +NFYB,3,0.036596257239580154,0.146284282207489,0.9574170235492387,-0.056884454552481485,12.593973159790039 +NFYC,3,0.05427411571145058,0.17711633443832397,0.9266518054327502,0.27519317992673004,15.337014198303223 +NHLRC2,9,0.01223082561045885,0.08480393886566162,0.9843536808745535,-0.030817296385731018,7.280686855316162 +NHP2,6,0.024295587092638016,0.12062340974807739,0.9680289361669145,0.09744485873786006,10.261436462402344 +NIFK,24,0.010131767950952053,0.07337195426225662,0.986342061539357,0.3464166764575228,6.626543998718262 +NIP7,28,0.006631919182837009,0.060854945331811905,0.9911245118632596,0.2893887879109167,5.361225128173828 +NISCH,15,0.008465513586997986,0.0703132301568985,0.9888474613297609,0.10531299483453083,6.057189464569092 +NKAP,13,0.015563743188977242,0.09333500266075134,0.9804327621830506,0.15258755457380727,8.212992668151855 +NLE1,42,0.007503546308726072,0.06257818639278412,0.990027096881862,0.36276987817772377,5.702663421630859 +NMD3,7,0.01954423263669014,0.106913723051548,0.9764798144284195,0.018034809739561012,9.20351505279541 +NMT1,3,0.039717234671115875,0.1549001932144165,0.9529836155087278,-0.01691380872495683,13.120002746582031 +NOB1,6,0.02891961857676506,0.13018473982810974,0.9613831987094763,-0.040117322489197305,11.195428848266602 +NOC2L,3,0.061232104897499084,0.1952817738056183,0.9248822686719902,0.057350653035639076,16.29048728942871 +NOC3L,9,0.017987702041864395,0.10256217420101166,0.9766455417090854,0.08503533790927491,8.829421997070312 +NOC4L,4,0.048925261944532394,0.16930022835731506,0.9361825316173122,0.10290836403792138,14.561664581298828 +NOL10,3,0.05096298083662987,0.17534227669239044,0.9334422542351452,0.09330791980722705,14.86181640625 +NOL11,27,0.013707212172448635,0.08300455659627914,0.9822427402199638,0.1519386672651124,7.707597732543945 +NOL12,10,0.016003787517547607,0.09586281329393387,0.9799695329393531,0.045376212726446236,8.328290939331055 +NOL6,12,0.023294605314731598,0.11331341415643692,0.9687998517279218,0.2575992593867205,10.047826766967773 +NOL7,8,0.017000144347548485,0.10158658027648926,0.9780106178035609,-0.013391858141323263,8.583625793457031 +NOL8,8,0.018536340445280075,0.10460803657770157,0.9758982580503476,0.2464855999372804,8.96306324005127 +NOL9,4,0.0249647069722414,0.12168288230895996,0.9708734552487277,-0.0010948256568418448,10.40178108215332 +NOLC1,12,0.01261979527771473,0.08506917208433151,0.9842056852495541,-0.0878762279304107,7.395552158355713 +NOM1,8,0.016318324953317642,0.09793341159820557,0.9778608718009403,0.1862664692426829,8.409732818603516 +NOMO3,19,0.007554677780717611,0.06655606627464294,0.9898233607938084,0.01889513291780111,5.722059726715088 +NOP10,14,0.012473076581954956,0.08330267667770386,0.9837204385182433,-0.04215967307805701,7.352436065673828 +NOP14,4,0.046323470771312714,0.16733938455581665,0.9396260490731031,0.03128167569949479,14.169189453125 +NOP16,15,0.014691844582557678,0.09101258963346481,0.9806637320335599,0.20957733363150763,7.9796271324157715 +NOP2,6,0.037668198347091675,0.1467798352241516,0.9499415149917529,0.2502802319289112,12.777087211608887 +NOP56,10,0.0125319454818964,0.08677810430526733,0.983593843747812,0.10077494152197237,7.369765758514404 +NOP58,5,0.03324959799647331,0.13647200167179108,0.956174411005384,0.06949651066191037,12.00432300567627 +NOP9,8,0.01711338758468628,0.09979871660470963,0.9770685057334062,0.03702669150554326,8.612166404724121 +NOS1AP,6,0.019202502444386482,0.10616288334131241,0.9757817312760823,-0.0024977778289591752,9.122699737548828 +NPAT,2,0.07980188727378845,0.21296757459640503,0.9014339815675794,0.16815745006953925,18.597349166870117 +NPB,19,0.007927377708256245,0.06818300485610962,0.9895924639230224,-0.05101629638013868,5.86150598526001 +NPEPPS,16,0.008685343898832798,0.07115509361028671,0.988777633777519,-0.031434605331823226,6.135330677032471 +NPLOC4,6,0.036010999232530594,0.14199666678905487,0.9516083665768725,0.3760106081398528,12.492864608764648 +NPM1,2,0.0842142179608345,0.22965800762176514,0.9016410039809915,-0.022910415584269,19.10456657409668 +NPM3,6,0.02484036423265934,0.12135949730873108,0.9684858678876053,-0.04976307211959405,10.37584400177002 +NR2C2AP,8,0.015925923362374306,0.09681399911642075,0.9795913527088195,-0.04565003171914368,8.308005332946777 +NRBP1,7,0.029656516388058662,0.1334778368473053,0.9602145635714857,0.12046672770894487,11.337166786193848 +NRDE2,51,0.008469578810036182,0.06435099989175797,0.9888345977891243,0.465166552941334,6.058642864227295 +NRF1,7,0.020393455401062965,0.10899143666028976,0.9732406747344361,0.10479994997803219,9.401342391967773 +NSA2,12,0.011105762794613838,0.08073970675468445,0.9858309446182352,0.024876009717689846,6.937750339508057 +NSF,17,0.007613009307533503,0.06763385981321335,0.9901973882081012,0.003830050197465087,5.744108200073242 +NSL1,7,0.021190088242292404,0.11146865785121918,0.9727159115509807,0.057495987169613114,9.583206176757812 +NSMCE1,6,0.025534827262163162,0.1222631186246872,0.9665889776196785,0.050689567633174816,10.519883155822754 +NSMCE2,12,0.010995296761393547,0.08032859116792679,0.9861853694957956,-0.07725256787423669,6.903160095214844 +NSRP1,23,0.008504725992679596,0.069620281457901,0.9889729958390716,0.21186793702859727,6.071200847625732 +NSUN4,15,0.008303025737404823,0.0690370574593544,0.98898883318487,0.0020744649708566523,5.998775959014893 +NUBP1,15,0.008569538593292236,0.07076508551836014,0.9886728806266487,0.09855002438250451,6.094290733337402 +NUBP2,16,0.008256785571575165,0.07000333070755005,0.9891385445475097,0.01786630851320211,5.982048988342285 +NUDC,11,0.014350592158734798,0.09103836119174957,0.980842881856939,0.06527142652440261,7.886410236358643 +NUDCD3,11,0.017251236364245415,0.09925361722707748,0.9769889293780506,0.22169904895800996,8.646783828735352 +NUDT21,6,0.03645434230566025,0.14128261804580688,0.9494528125774009,0.3357778805262569,12.569531440734863 +NUF2,51,0.003954438958317041,0.04741208255290985,0.9947475300374203,0.36264435867519607,4.139871597290039 +NUFIP1,7,0.024395037442445755,0.12026055157184601,0.9689459416661366,0.008695969735742937,10.282416343688965 +NUMA1,29,0.007409026846289635,0.06420963257551193,0.990063431789762,0.14288249838644457,5.666632175445557 +NUP107,31,0.008764219470322132,0.0682339072227478,0.9881735347576025,0.3382984991605783,6.163126468658447 +NUP133,8,0.018363066017627716,0.10354455560445786,0.9760251522095703,0.18304294931402174,8.921072006225586 +NUP153,8,0.017079731449484825,0.10074732452630997,0.9777525265636705,0.17440141365072837,8.603694915771484 +NUP155,1,0.10800661146640778,0.2643095850944519,0.8852112845621817,0.00037073008190117746,21.63563346862793 +NUP160,29,0.00733600277453661,0.06376205384731293,0.9898691563236605,0.4321076497634252,5.638637542724609 +NUP205,73,0.0023401130456477404,0.03697974979877472,0.996809471990741,0.2438839545312671,3.1846582889556885 +NUP214,13,0.010560372844338417,0.07804789394140244,0.9862477947895213,0.0247692221140686,6.76525354385376 +NUP35,7,0.02205071412026882,0.11470173299312592,0.972042805319385,0.036266636478357824,9.77587890625 +NUP43,3,0.05582086741924286,0.1837213933467865,0.9349222838462367,0.0008666462429350668,15.554022789001465 +NUP54,29,0.01093095913529396,0.07530194520950317,0.9856121034404742,0.3502204187761808,6.882934093475342 +NUP62,10,0.032804183661937714,0.13186191022396088,0.9550813746126475,0.4033002178360719,11.923645973205566 +NUP85,11,0.013572965748608112,0.08884274214506149,0.9816954031019218,0.1712131028929514,7.6697611808776855 +NUP88,3,0.060968585312366486,0.19013375043869019,0.9211155331539679,0.18536772923405784,16.255393981933594 +NUP93,16,0.008857712149620056,0.07191301137208939,0.9881494362465223,0.15319270972894636,6.195911884307861 +NUP98,51,0.008513850159943104,0.06560833007097244,0.9884188453953661,0.5390159530196782,6.074456691741943 +NUS1,13,0.010593988001346588,0.07811974734067917,0.9862191438783552,0.047880955521370734,6.776012897491455 +NUTF2,6,0.0319107286632061,0.13643521070480347,0.9604131019291953,0.24944404809975718,11.760148048400879 +NVL,11,0.011209733784198761,0.0811910405755043,0.9859450520406625,0.16215509445826853,6.970149517059326 +NXF1,25,0.005448271986097097,0.05729653686285019,0.9929065547435975,0.03668510424529814,4.8593010902404785 +OGFOD1,12,0.013815993443131447,0.08862341195344925,0.9815713340853828,0.10088655343738688,7.738121509552002 +OGT,3,0.03798000514507294,0.1497526317834854,0.9533326411771892,-0.058786624556577904,12.829861640930176 +OIP5,13,0.010476462543010712,0.0782756432890892,0.9862648616766246,-0.03205463900911549,6.7383222579956055 +OLFML3,18,0.0077833025716245174,0.06726568192243576,0.9895550888549137,0.05911704790628058,5.807997226715088 +OPA1,6,0.023663833737373352,0.11740092933177948,0.9701204360181893,0.127191731747168,10.127144813537598 +OR1D2,13,0.010084607638418674,0.07733000814914703,0.9869321482455214,0.01021972071334601,6.611103534698486 +OR1E2,3,0.035464007407426834,0.14619328081607819,0.9597409969553775,-0.02009108477925161,12.39762020111084 +OR2A1,19,0.007569358218461275,0.06688112765550613,0.989940107206269,-0.04098328424295004,5.727617263793945 +OR2T29,11,0.014492311514914036,0.09244739264249802,0.9814421685893197,0.1543312916860858,7.92525577545166 +OR2T5,10,0.014637011103332043,0.0934678390622139,0.9805822337908908,0.011397593164405709,7.964722633361816 +OR4F4,6,0.020786257460713387,0.11135881394147873,0.9744166690787868,0.0367238053735345,9.491451263427734 +OR4K1,2,0.05890597775578499,0.191165953874588,0.9270832529320777,-0.0712282109532439,15.978062629699707 +OR4N2,12,0.011127476580440998,0.08187204599380493,0.9860513972977759,-0.07549373448994004,6.944529056549072 +OR7A10,4,0.03177258372306824,0.13792265951633453,0.962247561867919,-0.08866194539288436,11.734665870666504 +OR7A17,11,0.012079487554728985,0.08483409136533737,0.9848126558106534,-0.09236602307779329,7.23550271987915 +ORC1,53,0.004688580986112356,0.05043966323137283,0.9936633643198994,0.3729810843584937,4.507805347442627 +ORC3,18,0.007411469239741564,0.06636456400156021,0.9901903068033566,0.14312079888968068,5.667566299438477 +ORC4,9,0.014246491715312004,0.09143821150064468,0.9831637228152189,-0.029822975462402604,7.857753753662109 +ORC5,9,0.016561472788453102,0.09965760260820389,0.97873736231313,0.037918315504354205,8.472155570983887 +ORC6,8,0.017293544486165047,0.10065366327762604,0.9773172145727007,0.16141907349408516,8.657380104064941 +OSBP,7,0.021228577941656113,0.11249453574419022,0.9731915233429538,-0.03090745872216205,9.59190559387207 +OSGEP,7,0.017725978046655655,0.10216241329908371,0.9767861228911965,0.0776222283594415,8.764951705932617 +OSTC,23,0.00657708989456296,0.06151564419269562,0.9918561194125353,-0.05146811310924142,5.339017391204834 +OTOP1,11,0.012257982045412064,0.08447607606649399,0.9843322652270614,-0.07479318741354482,7.288764953613281 +OTX1,11,0.01228669285774231,0.0848662406206131,0.9838833988818362,-0.04558318089483738,7.297295570373535 +OVCA2,7,0.018526636064052582,0.10409626364707947,0.9765169473214101,-0.1330711208374396,8.960716247558594 +OXA1L,35,0.008946468122303486,0.06731642782688141,0.9880963212924799,0.4591730476888405,6.226876735687256 +PABPC1,2,0.1345980316400528,0.28370401263237,0.8227797235897026,0.1593414017652242,24.15259552001953 +PABPN1,15,0.013905596919357777,0.08717216551303864,0.9811534403158834,0.17612099370554302,7.7631731033325195 +PAF1,24,0.033800430595874786,0.13277608156204224,0.9625611395295248,0.33472689591022387,12.103348731994629 +PAFAH1B1,12,0.011468126438558102,0.08267858624458313,0.9852135237943161,0.04385105294035434,7.050025463104248 +PAK1IP1,2,0.08227302134037018,0.2275475561618805,0.9098132431424495,0.062195090811989984,18.883094787597656 +PALB2,9,0.015795404091477394,0.0965665951371193,0.979575448707845,-0.11383184736408852,8.273892402648926 +PAM16,5,0.03145202621817589,0.13642272353172302,0.9591708934706475,0.16445919510349422,11.67531967163086 +PARN,13,0.010401920415461063,0.07884041965007782,0.9862529027231015,-0.021897631767555128,6.71430778503418 +PARS2,6,0.023186368867754936,0.11748182028532028,0.9697674365141077,0.08850801995466233,10.024456024169922 +PAXBP1,13,0.013909539207816124,0.08844664692878723,0.9810632939305555,0.29515859523336674,7.764273166656494 +PC,23,0.0064962697215378284,0.061628296971321106,0.991645725284857,-0.030733872365062635,5.306112766265869 +PCBP1,11,0.013456713408231735,0.08754569292068481,0.9826768220043346,0.0028331450511546565,7.636844158172607 +PCBP2,17,0.007660000119358301,0.06639721244573593,0.9905210526639765,-0.05417010470449651,5.7618088722229 +PCF11,50,0.007621643599122763,0.06211479753255844,0.9899469102080444,0.22790574531280813,5.747364521026611 +PCID2,30,0.005403401795774698,0.05500761419534683,0.993147581764648,-0.049168560913209855,4.839250564575195 +PCNA,13,0.02055957540869713,0.10740809887647629,0.9727668211903041,0.28989858406720975,9.439555168151855 +PDAP1,5,0.029838139191269875,0.13252966105937958,0.9605254457643598,0.17623405916677445,11.371829986572266 +PDCD11,10,0.028293900191783905,0.12453768402338028,0.9619754228206486,0.26689386479596827,11.073651313781738 +PDCD2,8,0.01726742647588253,0.10196321457624435,0.9772444380519129,-0.03150644490499276,8.650839805603027 +PDCD5,20,0.006929254624992609,0.0628335103392601,0.9908797866566579,0.11496842366114028,5.480090141296387 +PDCD6,21,0.0062753790989518166,0.06013491004705429,0.9916565432426411,0.11438361510900875,5.215121746063232 +PDCD6IP,25,0.0067606521770358086,0.06188255175948143,0.9910880408250146,-0.06414674511087995,5.413009166717529 +PDCD7,10,0.016975170001387596,0.10067829489707947,0.9789667231398584,0.1916175769296928,8.57731819152832 +PDPK1,8,0.02117009274661541,0.11304249614477158,0.9733806091024177,0.16371971532508095,9.578683853149414 +PDRG1,11,0.012984535656869411,0.08753308653831482,0.9824405896619905,0.13392501975060453,7.501665115356445 +PEF1,5,0.031317338347435,0.13533270359039307,0.9601904438992924,0.10249249498458453,11.650293350219727 +PELO,28,0.006345234811306,0.05978739261627197,0.9914406523695097,0.2964187652833066,5.244068145751953 +PELP1,7,0.024330541491508484,0.11926044523715973,0.968646387966404,0.09247215813774624,10.268815040588379 +PES1,7,0.025041893124580383,0.11967653781175613,0.9662872705285687,0.2486636805566388,10.417847633361816 +PET117,9,0.013624757528305054,0.08981358259916306,0.9821320355950328,-0.082893554994161,7.684380531311035 +PEX1,6,0.023290468379855156,0.11827804893255234,0.9696758821235208,0.02960241447334098,10.046934127807617 +PFDN1,4,0.03537952899932861,0.14648276567459106,0.9545722004712585,-0.021164402156990312,12.382845878601074 +PFDN2,8,0.02192951738834381,0.1118641197681427,0.9703989868445423,0.16229418723762967,9.74897575378418 +PFDN4,4,0.03659055382013321,0.1479330211877823,0.9518010821759108,0.15503119825182193,12.59299373626709 +PFDN5,11,0.012766918167471886,0.08549707382917404,0.9832608373190873,0.04199952049117972,7.438536643981934 +PFDN6,11,0.012242394499480724,0.08605898916721344,0.9836549344006011,0.024978349493873682,7.2841291427612305 +PFN1,26,0.006064343266189098,0.057075385004282,0.991804570001236,0.13125013555593448,5.126681327819824 +PGAM1,16,0.01075915340334177,0.07890639454126358,0.9859866195637876,0.13733812387914254,6.828629016876221 +PGD,11,0.014957965351641178,0.09200634062290192,0.9798154098005167,0.06360006416168373,8.051572799682617 +PGK1,4,0.03433169052004814,0.1453452706336975,0.9580830206455131,0.01057698786951058,12.19809627532959 +PGPEP1,5,0.03859498351812363,0.15065670013427734,0.9499087711670385,0.18789664281635798,12.933316230773926 +PGS1,11,0.011358117684721947,0.08179052174091339,0.9853120314599594,-0.027734514185009444,7.016130447387695 +PHAX,41,0.006391124799847603,0.05929219722747803,0.9913711055350528,0.37511873066180845,5.262996673583984 +PHB,25,0.028759440407156944,0.11753281205892563,0.9620753388717692,0.5811537276364588,11.164381980895996 +PHB2,41,0.023451346904039383,0.1035332977771759,0.9691503745522471,0.6085117060430989,10.081574440002441 +PHF12,11,0.01224381010979414,0.08440559357404709,0.9843767081583525,-0.02671916096205693,7.284550189971924 +PHF5A,23,0.013212127611041069,0.08381567150354385,0.982343448616727,0.4013158795435293,7.5671234130859375 +PI4KA,11,0.0127131761983037,0.08585326373577118,0.9831150175491704,-0.0003647977042824218,7.422863960266113 +PIAS4,20,0.010233843699097633,0.0747227892279625,0.9865160000549856,0.3689684772498301,6.6598405838012695 +PIGH,9,0.014417135156691074,0.09163442999124527,0.981299077307574,0.04901515585695161,7.904673099517822 +PIK3C3,11,0.014077496714890003,0.0914442166686058,0.9812967629431307,0.019979109616966497,7.811009883880615 +PIK3R4,7,0.022117726504802704,0.11422885954380035,0.9708369049714229,0.033079298147880735,9.790721893310547 +PKM,15,0.010657599195837975,0.07936231791973114,0.9863440306824371,-0.035005098745060526,6.796324729919434 +PKMYT1,9,0.016148146241903305,0.09805461019277573,0.978742289173839,0.09924969946553681,8.365767478942871 +PLA2G10,20,0.006621583364903927,0.06306632608175278,0.9912640256130004,0.060197539858836975,5.357046127319336 +PLEKHN1,11,0.011076423339545727,0.0816573053598404,0.9859067678747638,0.021609872752913168,6.928579330444336 +PLK1,6,0.02305545099079609,0.11668184399604797,0.9699493875037521,-0.0182282516614946,9.996115684509277 +PLK4,3,0.04697059467434883,0.16976986825466156,0.9412706992956432,0.029526451534096613,14.267815589904785 +PLRG1,10,0.01912687160074711,0.1040744036436081,0.9741469158294367,0.16107990364604716,9.104716300964355 +PMF1,27,0.006745744962245226,0.06110202893614769,0.9910950636649648,0.20053202612787085,5.407037734985352 +PMPCA,14,0.01208284217864275,0.08274991810321808,0.9840604645064684,0.25616909597875337,7.236506938934326 +PMPCB,56,0.007171207573264837,0.05953051522374153,0.9907350735470344,0.5785990932012502,5.574944972991943 +PNISR,19,0.00817941501736641,0.06848719716072083,0.9891169801218366,0.11216046451402674,5.953955173492432 +PNKP,19,0.007461694069206715,0.06637956202030182,0.9905930719699397,-0.06240766059431432,5.686737537384033 +PNN,8,0.019761519506573677,0.10630038380622864,0.9737494017943438,0.19746195607800565,9.254535675048828 +PNO1,14,0.019076457247138023,0.0999658852815628,0.9741233524760827,0.19977632437105297,9.0927095413208 +PNPT1,13,0.013975211419165134,0.0857214704155922,0.9833419215304722,0.12198451395982161,7.782580852508545 +POGZ,4,0.03899627551436424,0.15223626792430878,0.9520908819873237,0.02419578655680127,13.00037956237793 +POLA1,23,0.00821603462100029,0.06824187934398651,0.9887897337019644,0.18901433635845444,5.967268466949463 +POLA2,24,0.009584931656718254,0.07133671641349792,0.9868741707680887,0.27482065271554135,6.445237636566162 +POLD1,21,0.012126456946134567,0.07661040127277374,0.9833991953765079,0.24518650977988193,7.249556541442871 +POLD2,8,0.020335610955953598,0.10691624879837036,0.9727367539818071,0.10081586303741033,9.387999534606934 +POLD3,7,0.02141755446791649,0.11076031625270844,0.9711187415199081,0.1489121562339217,9.634504318237305 +POLE,17,0.012060629203915596,0.08233531564474106,0.9835376792741877,0.2635706900628168,7.229852676391602 +POLE2,5,0.021410971879959106,0.11251786351203918,0.9739846323788242,-0.053209314461541174,9.633024215698242 +POLG2,15,0.009836122393608093,0.07529096305370331,0.9873713645381972,-0.04607783332024125,6.529146671295166 +POLR1A,12,0.01737523451447487,0.09754841774702072,0.9766277024301651,0.10023108684121253,8.677803039550781 +POLR1B,18,0.020629294216632843,0.10066353529691696,0.971977872759995,0.21656148936033082,9.455546379089355 +POLR1C,18,0.009575266391038895,0.07477239519357681,0.9870425773514627,0.08101948656362126,6.441987991333008 +POLR1E,21,0.009711826220154762,0.07331893593072891,0.9874170408770675,0.0039963064493337,6.487761974334717 +POLR2A,41,0.006649587769061327,0.05967011675238609,0.9911492141552111,0.2557735675788621,5.3683624267578125 +POLR2B,32,0.013732445426285267,0.08376771956682205,0.9813681610040395,0.30652658631547935,7.714688777923584 +POLR2C,19,0.017388565465807915,0.09683648496866226,0.9764034198444301,0.27726560746059,8.681131362915039 +POLR2D,28,0.006782867014408112,0.061926476657390594,0.9906453495085593,0.1664654256012546,5.4218950271606445 +POLR2E,23,0.007362570147961378,0.06572837382555008,0.9898573228009235,0.09180894534793503,5.648838520050049 +POLR2F,33,0.006505352910608053,0.06067078933119774,0.9912612512085252,0.23445664938065414,5.309821128845215 +POLR2G,15,0.022742098197340965,0.1101682260632515,0.9692105757542626,0.2604575856690451,9.927952766418457 +POLR2H,10,0.04809264838695526,0.16064295172691345,0.9345756846416495,0.3440454619273181,14.437227249145508 +POLR2I,23,0.021473105996847153,0.10566139966249466,0.9721603863758431,0.28741439782416006,9.646990776062012 +POLR2K,2,0.11341112852096558,0.26393499970436096,0.8533507329709822,0.2756343942251064,22.170337677001953 +POLR2L,6,0.03519880771636963,0.14466135203838348,0.9529502013055612,0.11498519343252822,12.351179122924805 +POLR3A,1,0.1346093863248825,0.29228025674819946,0.8592479464285893,0.12954697184984518,24.153614044189453 +POLR3B,12,0.012844398617744446,0.08830864727497101,0.9837110555935522,0.13475937908089242,7.461073875427246 +POLR3C,17,0.009612292051315308,0.0745709016919136,0.9874646263416167,-0.061079853071340624,6.454431056976318 +POLR3D,12,0.011241747997701168,0.08097544312477112,0.985273746433581,0.0736649392504313,6.980095863342285 +POLR3E,22,0.006943808868527412,0.06382198631763458,0.9913330613169732,-0.07290604654252068,5.485842704772949 +POLR3F,17,0.009637666866183281,0.07447804510593414,0.9874760118225493,0.06266918513868863,6.462944507598877 +POLR3H,11,0.011847298592329025,0.08232291787862778,0.9856263998230844,-0.13281120590757717,7.165626049041748 +POLR3K,11,0.01522717997431755,0.09312597662210464,0.9796782822045182,0.12145028194337092,8.123706817626953 +POLRMT,22,0.011942404322326183,0.07206329703330994,0.9849553334788576,0.17575094013432874,7.194329738616943 +POP1,9,0.016527676954865456,0.09815718978643417,0.9782030567914962,-0.01326213858812719,8.463506698608398 +POP4,13,0.012729235924780369,0.08566957712173462,0.9830736153208389,0.05684123675365359,7.427550792694092 +POP5,5,0.032231949269771576,0.1383994072675705,0.9570770390230284,0.10403222239961353,11.819190979003906 +POP7,22,0.007108318619430065,0.06464802473783493,0.9904785819521303,0.03720922076902693,5.550446510314941 +POT1,1,0.12440548092126846,0.2728487551212311,0.8693247814814353,0.18185488476960252,23.220104217529297 +POTEI,12,0.014078617095947266,0.08973933011293411,0.9819691591254692,0.09780520277097188,7.8113203048706055 +POU5F1B,8,0.014918236993253231,0.09366529434919357,0.9809766632410601,0.09217229355759618,8.040873527526855 +PPA1,15,0.009273750707507133,0.07228244841098785,0.9875676520458666,0.1887731404543722,6.339750289916992 +PPA2,7,0.022169074043631554,0.11606758832931519,0.9729093171087035,0.023328139373258443,9.802079200744629 +PPAN,11,0.014711814001202583,0.09146223962306976,0.980234537057909,0.19632075599921728,7.985048294067383 +PPARGC1B,14,0.010497073642909527,0.07821256667375565,0.98610780230082,0.04193649314396799,6.744947910308838 +PPIA,16,0.009457484818994999,0.07027217000722885,0.9880007884063122,-0.012826039504548902,6.402245044708252 +PPIE,4,0.0412696897983551,0.15364491939544678,0.9474832300718907,0.12078218177014445,13.373961448669434 +PPIH,12,0.011970192193984985,0.08390141278505325,0.9845964451790064,0.09626807066234391,7.202694892883301 +PPIL1,3,0.04344158247113228,0.16257673501968384,0.9459205937368887,-0.045257989022009636,13.72136402130127 +PPIL2,13,0.010813978500664234,0.07986847311258316,0.9860001673959679,-0.013794846705561723,6.846004962921143 +PPIL4,8,0.03882861137390137,0.14755669236183167,0.9515029119963708,0.2164990652745529,12.972402572631836 +PPME1,22,0.0068769087083637714,0.06292079389095306,0.9909419259772027,0.00557724491429134,5.459351539611816 +PPP1CA,25,0.0057901558466255665,0.05794178321957588,0.9926826247669533,-0.024519246773490877,5.009444713592529 +PPP1CB,16,0.008421328850090504,0.07015437632799149,0.9890156156638661,0.05340873796153475,6.041360378265381 +PPP1R10,1,0.24444712698459625,0.3752514719963074,0.73344828829146,0.27899925072091414,32.548946380615234 +PPP1R11,7,0.019186828285455704,0.10550446063280106,0.9765576652474658,-0.04418617375485049,9.118975639343262 +PPP1R12A,18,0.006898859981447458,0.06389307975769043,0.9912121941967953,0.02828364353742897,5.468058109283447 +PPP1R15B,13,0.010177163407206535,0.07630699127912521,0.9866023412157225,0.05082443603714482,6.641372203826904 +PPP1R2,8,0.015475073829293251,0.09514836221933365,0.9800792390094492,0.062210727915651784,8.18956470489502 +PPP1R7,7,0.01872178725898266,0.10629397630691528,0.975027019991971,0.17040424695915288,9.007786750793457 +PPP1R8,7,0.02135319635272026,0.11155173182487488,0.9733929882589448,0.08842127344470146,9.620018005371094 +PPP2CA,19,0.008350308984518051,0.06967596709728241,0.9890079352772662,-0.05000884505097719,6.015832901000977 +PPP2R1A,45,0.0038444094825536013,0.04664594307541847,0.9948098138185587,0.04110490444623032,4.081871032714844 +PPP2R3C,5,0.032068971544504166,0.13846513628959656,0.9578023863595712,-0.042525433634540914,11.789271354675293 +PPP4C,17,0.008111320436000824,0.06800127774477005,0.9891720956454755,0.031904176967114645,5.929120063781738 +PPP4R2,18,0.00940210185945034,0.07297240942716599,0.9875309931024028,0.08131453868140277,6.383471488952637 +PPP6C,12,0.01044977456331253,0.079243503510952,0.9870990415758283,-0.036956199860704114,6.729733943939209 +PPRC1,12,0.020425111055374146,0.102378711104393,0.9718430339081993,-0.02911537252331459,9.408636093139648 +PPWD1,3,0.039992596954107285,0.15584500133991241,0.9504936290999282,-0.006683460372502394,13.1654052734375 +PPY,1,0.16763997077941895,0.3236531913280487,0.8220567266251224,0.006891397198453273,26.954622268676758 +PRC1,18,0.0072336397133767605,0.0651186928153038,0.9905475548231218,0.02604351497947381,5.599160194396973 +PRDM10,2,0.06425245851278305,0.19808481633663177,0.9221906249142858,0.07417843757551101,16.68742561340332 +PREB,4,0.04329650476574898,0.15980905294418335,0.9417309320326611,0.2090578504232024,13.698432922363281 +PRELID1,15,0.010360903106629848,0.07740264385938644,0.986087161405839,0.15786562981605864,6.701056480407715 +PRELID3B,28,0.008276172913610935,0.06676367670297623,0.9893667121977824,0.34572063708574813,5.989068031311035 +PRIM1,18,0.01239034067839384,0.07982860505580902,0.9831001703206402,0.25487352460579465,7.328010559082031 +PRIM2,15,0.018999852240085602,0.1001938134431839,0.9735643872235743,0.32114299598204493,9.074434280395508 +PRKRA,8,0.014491475187242031,0.09279277920722961,0.9815938522411244,0.02212757923897838,7.925026893615723 +PRKRIP1,18,0.008086295798420906,0.06882550567388535,0.9897450791833488,-0.0014695949784067704,5.919966220855713 +PRMT1,6,0.01923995278775692,0.10773351788520813,0.9763916526001506,0.04414667628783764,9.131591796875 +PRMT5,12,0.010938321240246296,0.08092080056667328,0.9857765177719479,0.08771636685884289,6.885251522064209 +PRODH,38,0.004034444689750671,0.04880344122648239,0.9949328368713657,-0.014643938812256516,4.181540489196777 +PRORP,4,0.026347646489739418,0.12585775554180145,0.9686241268668025,0.09738502283440971,10.686004638671875 +PRPF18,27,0.011744857765734196,0.08030956238508224,0.984629058866141,0.1318924443999876,7.134579181671143 +PRPF19,11,0.018621094524860382,0.10285891592502594,0.9759691309951211,0.20531983235709628,8.983530044555664 +PRPF3,15,0.010922971181571484,0.07887183129787445,0.9853800537661265,0.24049864246985558,6.88041877746582 +PRPF31,2,0.05189063400030136,0.1799754649400711,0.9372081815816474,0.04541128374513442,14.996466636657715 +PRPF38A,2,0.050994694232940674,0.1778850108385086,0.9425601577128355,-0.049986903135584286,14.866438865661621 +PRPF38B,3,0.058853402733802795,0.19031783938407898,0.9238341225404637,0.15365920631873425,15.970931053161621 +PRPF4,11,0.027468010783195496,0.12198581546545029,0.9636281759651647,0.3273443842785512,10.910836219787598 +PRPF40A,12,0.0256117582321167,0.11850251257419586,0.968108708860809,0.25450820022617676,10.535717964172363 +PRPF4B,21,0.007955089211463928,0.0676732212305069,0.9900494006295284,0.0756123344693383,5.8717427253723145 +PRPF6,23,0.008146338164806366,0.06813348084688187,0.98890125597414,0.2502707502253207,5.941904544830322 +PRPF8,15,0.012375337071716785,0.0843164473772049,0.9832338682929798,0.25865331985924533,7.323572158813477 +PRRC2A,9,0.014013275504112244,0.09118098765611649,0.9824216305819655,0.07237312853255441,7.793172359466553 +PSAT1,7,0.01925688236951828,0.1066407710313797,0.9739788463504597,0.17872801960858486,9.135607719421387 +PSMA1,17,0.009720469824969769,0.0750197097659111,0.9869825399859133,0.16176189881149305,6.49064826965332 +PSMA2,7,0.019641071557998657,0.1045914888381958,0.9732894466104169,0.16059213468639724,9.226287841796875 +PSMA3,37,0.008757389150559902,0.06604590266942978,0.9880211734067275,0.29720863019530425,6.16072416305542 +PSMA4,39,0.012667553499341011,0.07805708050727844,0.9826246913928096,0.3586425548492312,7.40953254699707 +PSMA5,3,0.06785842031240463,0.1977081298828125,0.9118334243276481,0.24131466364290735,17.1492977142334 +PSMA6,4,0.0379733182489872,0.1499483436346054,0.950409717909116,0.1599611817165872,12.82873249053955 +PSMA7,4,0.041365768760442734,0.15508393943309784,0.9476666547682636,0.10106706958389423,13.389519691467285 +PSMB1,15,0.009411326609551907,0.07218577712774277,0.9874446452875053,0.09392889510773438,6.386602401733398 +PSMB2,6,0.027050403878092766,0.12429121136665344,0.9637249443179183,0.08472555969631036,10.8275785446167 +PSMB3,3,0.06242759898304939,0.19145604968070984,0.9191376862770774,0.1209677367222712,16.44874382019043 +PSMB4,10,0.014962281100451946,0.09429337829351425,0.9804759745166759,0.03574832723730717,8.052734375 +PSMB5,27,0.009267992340028286,0.06943841278553009,0.9872352716650928,0.17157366909621927,6.33778190612793 +PSMB6,20,0.022737279534339905,0.1047045886516571,0.9685008463868507,0.30058886910341986,9.926900863647461 +PSMB7,41,0.005397668574005365,0.05386248975992203,0.9925838550477761,0.15022766330627887,4.836682319641113 +PSMC1,31,0.020345015451312065,0.09846619516611099,0.9719631698867462,0.35105633709433415,9.390170097351074 +PSMC2,43,0.009873013943433762,0.07076352834701538,0.9864882413840405,0.3273128102951738,6.541379451751709 +PSMC3,4,0.05084051191806793,0.17093735933303833,0.9357747738197425,0.08842256400672834,14.843947410583496 +PSMC4,24,0.02391868643462658,0.10965429246425629,0.9684169402090413,0.398998946182414,10.181530952453613 +PSMC5,1,0.1367119401693344,0.29437482357025146,0.8529609066215614,-0.04877947462079002,24.34151840209961 +PSMC6,15,0.013401929289102554,0.08769183605909348,0.9814150027324947,0.24700855470901814,7.621283531188965 +PSMD1,28,0.008065534755587578,0.06623494625091553,0.9889119252107166,0.23450131487092182,5.912362098693848 +PSMD11,8,0.019923683255910873,0.107527956366539,0.9726868786047179,0.12641547061592434,9.292428970336914 +PSMD12,27,0.014682997018098831,0.086164191365242,0.9796711114896199,0.2754154243825146,7.977223873138428 +PSMD13,12,0.02182742953300476,0.10865139961242676,0.969662596604148,0.17031440750472612,9.72625732421875 +PSMD14,1,0.2659401297569275,0.39939454197883606,0.6957823760625282,0.16030360009744385,33.949737548828125 +PSMD2,7,0.055891506373882294,0.17200662195682526,0.9228935232856887,0.3021864897360201,15.563860893249512 +PSMD3,7,0.023182367905974388,0.11714257299900055,0.9682395512878075,0.11836389187063912,10.023591041564941 +PSMD4,35,0.00999507401138544,0.0703437477350235,0.9861822963045462,0.19483987788262552,6.581690311431885 +PSMD6,55,0.012237672694027424,0.07396852970123291,0.9831082568171216,0.24821971380512745,7.282724380493164 +PSMD7,38,0.009935431182384491,0.07101672887802124,0.9864477052252166,0.33256318581234234,6.562024116516113 +PSMD8,2,0.06068066135048866,0.1880238801240921,0.9257109421061638,0.13248185782949384,16.21696662902832 +PSMD9,10,0.01394948735833168,0.0896267294883728,0.9812430875239555,0.04601407713373977,7.77541446685791 +PSME1,12,0.012353898957371712,0.08432105928659439,0.9835269012384662,0.11624303277766151,7.317225933074951 +PSME2,15,0.009820105507969856,0.07516805827617645,0.9872895937770397,-0.07968787682657279,6.523828506469727 +PSMG1,11,0.013522101566195488,0.08754856884479523,0.9828983808047119,-0.11062368819068209,7.65537691116333 +PSMG2,17,0.00840652734041214,0.07065724581480026,0.9889466079904802,0.03458550911454496,6.036048889160156 +PSMG3,44,0.003466235473752022,0.044576745480298996,0.9954265626549059,0.114260696318109,3.875908136367798 +PSMG4,6,0.03241964429616928,0.13556963205337524,0.9562053817220996,0.05900174359709912,11.853553771972656 +PSTK,38,0.0038807899691164494,0.04726596921682358,0.9949387264791161,0.04206306399813418,4.101139068603516 +PTBP1,15,0.011573356576263905,0.08061814308166504,0.9846931723737502,0.16669116794515454,7.082296848297119 +PTCD1,26,0.005363443400710821,0.05603887885808945,0.9930054655789784,0.2373312798950655,4.821323871612549 +PTCD3,7,0.02279813215136528,0.11699120700359344,0.9706722356381055,0.026474323383444128,9.940176963806152 +PTK2,15,0.008569817058742046,0.07072177529335022,0.9891079002518827,-0.009269350927289819,6.094389915466309 +PTMA,15,0.011120503768324852,0.07711522281169891,0.9850658268400809,0.017896944531126065,6.942352294921875 +PTPMT1,19,0.007407302502542734,0.06561674922704697,0.9903450853122475,-0.00988754696193676,5.665972709655762 +PTPN23,14,0.008950581774115562,0.0730343610048294,0.9881981607687512,0.08420439016837716,6.228308200836182 +PUF60,9,0.017950350418686867,0.10211049020290375,0.9772563228895559,0.004432936850264771,8.820250511169434 +PWP1,7,0.018542485311627388,0.10496240854263306,0.9769279266132992,0.013065364720326269,8.964548110961914 +PWP2,5,0.03162950277328491,0.1365325152873993,0.9586714673547986,0.03138341758948414,11.708213806152344 +PXN,16,0.008586007170379162,0.06994761526584625,0.9891279175884443,-0.059437546871311864,6.1001434326171875 +PYROXD1,10,0.012199065648019314,0.08589394390583038,0.9846044104737858,0.04918448840982335,7.2712273597717285 +QARS,40,0.01409184094518423,0.08385322242975235,0.9820602951621673,0.5727789095743457,7.814988136291504 +QRSL1,12,0.011051026172935963,0.08030758798122406,0.9857430268870487,0.07799738915391151,6.9206318855285645 +RAB18,5,0.023065119981765747,0.11907844990491867,0.9727850396317225,0.008953694846614334,9.998211860656738 +RAB4A,31,0.004988424014300108,0.05337520316243172,0.9933859086253448,-0.005110751133050334,4.649713039398193 +RAB6A,6,0.023022139444947243,0.11722537130117416,0.9705642356224531,0.008417486068877962,9.9888916015625 +RABGGTA,11,0.012294312007725239,0.08393710106611252,0.9842675839719114,-0.04244401136737969,7.299558162689209 +RABGGTB,17,0.00905570574104786,0.07290595769882202,0.9879124677589516,0.19811774386601974,6.264776706695557 +RAC1,20,0.008024806156754494,0.06794621050357819,0.9900401151861302,-0.08151456446264323,5.8974151611328125 +RAC3,22,0.007459241896867752,0.06621881574392319,0.9906040807498845,0.11913100008284805,5.685802936553955 +RACGAP1,17,0.01061439048498869,0.07747804373502731,0.9854573310236516,0.34663655484577377,6.782534122467041 +RAD17,13,0.011427254416048527,0.081636942923069,0.9847925502830859,0.06688254131789588,7.037451267242432 +RAD21,11,0.0158831924200058,0.09543026983737946,0.9785410004459358,0.2328674522851579,8.29685115814209 +RAD51,24,0.005809202790260315,0.055832259356975555,0.9924476560995013,0.21906647467347648,5.017677307128906 +RAD51C,7,0.019725831225514412,0.10605048388242722,0.9740022725485071,0.2090785483800765,9.246174812316895 +RAD51D,14,0.009904718957841396,0.07606405764818192,0.9873269063807494,-0.04205540550797233,6.551873683929443 +RAD9A,4,0.029776815325021744,0.13430769741535187,0.9628566801878793,0.016703894444055192,11.360136985778809 +RAE1,6,0.03325418382883072,0.13839897513389587,0.9547791862073965,0.09475373202056865,12.005151748657227 +RAN,20,0.006777429953217506,0.06331206113100052,0.9912827480566954,0.07935261684355435,5.419721603393555 +RANBP2,9,0.015281204134225845,0.09550821781158447,0.9802135641619744,0.005644489684298161,8.138104438781738 +RANGAP1,13,0.016771845519542694,0.09704560041427612,0.9789962534533792,0.24924710621438953,8.525794982910156 +RARS,41,0.008169949986040592,0.06404805183410645,0.9894366077263146,0.45989154915804165,5.950509071350098 +RARS2,15,0.009502440690994263,0.07515490800142288,0.9876425347897351,0.23597134767235897,6.417443752288818 +RBBP4,9,0.015855660662055016,0.09529395401477814,0.9792757878001646,0.12970355407301454,8.289658546447754 +RBBP5,3,0.06240738928318024,0.1855182647705078,0.9182665300719839,0.26983387190300356,16.446081161499023 +RBBP6,8,0.03127038851380348,0.13149498403072357,0.9580857438460703,0.18900406079909848,11.641558647155762 +RBBP8,11,0.011119131930172443,0.07924574613571167,0.9855048265037395,0.09794397907288785,6.941924571990967 +RBM10,13,0.009658768773078918,0.07612260431051254,0.9871668408810075,0.10641658147554578,6.470015525817871 +RBM14,23,0.011195167899131775,0.07319174706935883,0.9851819568169489,0.12378721515551178,6.965620040893555 +RBM17,7,0.020604146644473076,0.11002106219530106,0.9750119306858744,0.0009003340897229785,9.449782371520996 +RBM19,5,0.033341363072395325,0.14075835049152374,0.9553198902141199,0.045325087173715216,12.020875930786133 +RBM22,1,0.14567849040031433,0.29367566108703613,0.8497604834784332,0.29472079653497096,25.12708854675293 +RBM25,7,0.030698804184794426,0.13326384127140045,0.9602384166148203,0.33256702595114696,11.534669876098633 +RBM28,7,0.022543255239725113,0.11288395524024963,0.972972492224333,0.11278430854174641,9.884455680847168 +RBM39,8,0.02126367948949337,0.1108739823102951,0.9722818172526783,0.17621680015158414,9.599832534790039 +RBM4,22,0.006811035331338644,0.06265207380056381,0.9910489083267348,0.0770268181982429,5.433141708374023 +RBM42,8,0.023749155923724174,0.11736927181482315,0.9707168216786866,-0.0783270004524923,10.1453857421875 +RBM48,5,0.027526210993528366,0.12743353843688965,0.9655572362852578,0.05584484024792622,10.92238998413086 +RBM8A,4,0.04172171279788017,0.15492764115333557,0.944236053184105,0.3065870669071751,13.447003364562988 +RBMX,14,0.010919223539531231,0.0780516266822815,0.9852214513571091,0.14661051126461253,6.879237651824951 +RBMX2,25,0.01604568213224411,0.09185414016246796,0.9795756040903721,0.3509168876302059,8.339183807373047 +RBMXL1,6,0.020962201058864594,0.11282530426979065,0.9742374265382395,-0.037166942687358165,9.531536102294922 +RBX1,5,0.042389702051877975,0.15609431266784668,0.9438255871713318,0.1438861333754586,13.55422306060791 +RCC1,22,0.007298172451555729,0.06532160192728043,0.9903754924035656,0.1800466355398788,5.624080181121826 +RCL1,10,0.017488347366452217,0.09969774633646011,0.9775113853245866,0.06675206166540812,8.706003189086914 +RCOR1,3,0.04755533114075661,0.17291800677776337,0.9418363011257032,-0.0013417516148862326,14.35634994506836 +REV3L,10,0.013585765846073627,0.08975901454687119,0.9819669963209122,0.12890120752292286,7.673376560211182 +REXO2,8,0.021368244662880898,0.11010231822729111,0.9740671534880492,-0.013749608655341994,9.623407363891602 +RFC1,15,0.010948175564408302,0.07827267795801163,0.9851338287073014,0.2300757343261234,6.888352394104004 +RFC2,9,0.014855584129691124,0.09261906147003174,0.9805534567229116,0.15160856990276883,8.023970603942871 +RFC3,23,0.007288898807018995,0.06308552622795105,0.9902067748427352,0.23931699562216474,5.6205058097839355 +RFC4,7,0.01961541175842285,0.10622251778841019,0.9737023267731826,0.23757860483722196,9.220259666442871 +RFC5,11,0.011646582745015621,0.08268245309591293,0.9850225212990872,0.09717188451308391,7.104666709899902 +RFFL,1,0.10231176763772964,0.2540275752544403,0.8823620047754893,-0.08532032086669963,21.05752182006836 +RFT1,12,0.011197026818990707,0.08128689974546432,0.9855094949232316,0.034085669524952,6.966197967529297 +RGP1,13,0.01020841021090746,0.0774487555027008,0.9868246213351803,-0.07607368765510306,6.651559352874756 +RGPD6,20,0.006877532694488764,0.06323996186256409,0.9910980640950096,-0.07101447051476718,5.459599494934082 +RHOQ,19,0.006881368812173605,0.06336416304111481,0.9909314710004298,0.08011735347995465,5.461122035980225 +RHPN1,23,0.006235888693481684,0.060247886925935745,0.9923304568389246,-0.13264278107787972,5.198686599731445 +RINT1,15,0.009659938514232635,0.07405782490968704,0.9875208130194856,0.18551147346133812,6.470407485961914 +RIOK1,32,0.011183037422597408,0.07587835937738419,0.9855084737105495,0.1797343284893141,6.961845397949219 +RIOK2,18,0.01301406417042017,0.08376696705818176,0.9820977820965165,0.19362447055008156,7.510189533233643 +RMI1,21,0.0072289081290364265,0.06475669890642166,0.9903654364194594,0.10578215607292268,5.5973286628723145 +RNASEH2C,10,0.015284363180398941,0.09504127502441406,0.9803040614553823,-0.10359770227394892,8.138944625854492 +RNF103,11,0.01579369604587555,0.09645017981529236,0.9781644717196126,0.061299750885151126,8.273444175720215 +RNF113A,17,0.014908104203641415,0.09131180495023727,0.9814997895535921,0.2521608609499158,8.038142204284668 +RNF123,11,0.011820521205663681,0.08271516859531403,0.9847632797988622,-0.07772307247687069,7.157523155212402 +RNF14,24,0.005653252359479666,0.05768166482448578,0.9927021168287523,-0.0030300525414619834,4.949868202209473 +RNF168,6,0.021552523598074913,0.11255992203950882,0.9732810304337677,-0.0887735237194664,9.664813995361328 +RNF20,13,0.016276709735393524,0.09344334155321121,0.9776201208720409,0.22830545139738426,8.399003028869629 +RNF4,3,0.050275832414627075,0.1700444221496582,0.9358403954479058,0.1823447512339127,14.761282920837402 +RNF40,20,0.008250823244452477,0.06879033893346786,0.9889166790773297,0.04306071006094101,5.979888439178467 +RNF8,17,0.008806999772787094,0.0717075988650322,0.9885266155589632,-0.008871313404089032,6.178149700164795 +RNGTT,11,0.012545817531645298,0.08501099050045013,0.9838235013493813,-0.021002403448988363,7.373843669891357 +RNMT,12,0.013801034539937973,0.08811213076114655,0.9819564755665334,0.01950642563697748,7.733930587768555 +RNPC3,8,0.01731927879154682,0.10205940157175064,0.9770775111997848,0.17101318793028636,8.663819313049316 +RNPS1,13,0.01293508242815733,0.0866565853357315,0.9828312970889321,0.004750197396879326,7.487366199493408 +ROMO1,8,0.023574987426400185,0.11414643377065659,0.9698367621908508,0.3342098360129857,10.108116149902344 +RPA1,47,0.004663058556616306,0.05066315084695816,0.9937159118913745,0.24207096734463243,4.495519638061523 +RPA2,23,0.01427043229341507,0.08678604662418365,0.9806518806886656,0.3907122204743107,7.864353179931641 +RPA3,21,0.011799509637057781,0.08145171403884888,0.984721792545073,0.27109934878442676,7.151159286499023 +RPAIN,11,0.011612814851105213,0.08284704387187958,0.9848153673105429,0.14778254320074444,7.094359874725342 +RPAP1,9,0.024233581498265266,0.11774516850709915,0.9675431102236405,0.19908237361167389,10.248333930969238 +RPAP2,40,0.016982387751340866,0.09233427792787552,0.9778056825573815,0.3519921675261249,8.579141616821289 +RPAP3,2,0.08940443396568298,0.23301465809345245,0.8800284461988764,0.2019705620941177,19.684480667114258 +RPE,12,0.013396944850683212,0.08844684064388275,0.9829204464625683,0.03594013474229444,7.619865894317627 +RPF1,13,0.010894941166043282,0.07980802655220032,0.9855645616315016,0.15051376281368076,6.871584415435791 +RPF2,7,0.020569728687405586,0.10942292213439941,0.9744701495799147,0.06434146545476344,9.441885948181152 +RPL10,27,0.015854185447096825,0.08704383671283722,0.9803229704677182,0.3092237112879887,8.28927230834961 +RPL10A,25,0.010552968829870224,0.07328512519598007,0.9855729650366758,0.3393953307608668,6.762881755828857 +RPL11,38,0.010867024771869183,0.07144242525100708,0.9856168213708087,0.36861397401441287,6.8627753257751465 +RPL12,25,0.008100338280200958,0.0651707723736763,0.9889009082867484,0.41634388146020596,5.925105094909668 +RPL13,33,0.007635092828422785,0.06300117820501328,0.9896483895517038,0.3120976647606171,5.752433776855469 +RPL13A,16,0.00822945311665535,0.06850800663232803,0.9891448671594196,0.07728640296772787,5.972139835357666 +RPL14,22,0.020093552768230438,0.09749332815408707,0.9724934086485088,0.4242678906615058,9.331958770751953 +RPL15,17,0.021699413657188416,0.10644418001174927,0.969775887150179,0.32860923929951186,9.697693824768066 +RPL17,15,0.02977190725505352,0.12140195071697235,0.962290389580717,0.34022980736240954,11.359201431274414 +RPL18,31,0.014901882968842983,0.08640926331281662,0.9815698203550639,0.33723569792078834,8.03646469116211 +RPL18A,7,0.019542096182703972,0.10660234838724136,0.9755775084445925,-0.004086612150434792,9.203012466430664 +RPL19,24,0.016519570723176003,0.09003392606973648,0.9785680052333816,0.30479167284388586,8.461430549621582 +RPL21,26,0.0112964678555727,0.07663160562515259,0.9854826726685816,0.3357275516135607,6.997062683105469 +RPL23,40,0.015387404710054398,0.08492082357406616,0.980743352805437,0.33509482152316533,8.16633415222168 +RPL23A,10,0.0338430218398571,0.13121837377548218,0.9563964996697077,0.30892351427767156,12.11097240447998 +RPL24,46,0.011673801578581333,0.07255420088768005,0.9849588555542415,0.37965389783399683,7.112964153289795 +RPL26,21,0.017064804211258888,0.0906691923737526,0.9787224754100953,0.2576735411129207,8.599933624267578 +RPL26L1,15,0.00864429771900177,0.07092150300741196,0.9889370843543163,0.017774217650143812,6.120815753936768 +RPL27,4,0.047492966055870056,0.15978287160396576,0.9429961948353901,0.24615001419929977,14.34693431854248 +RPL27A,20,0.020305432379245758,0.1010209172964096,0.9737092103232997,0.3392492954287645,9.381031036376953 +RPL28,5,0.025790296494960785,0.12081257253885269,0.9666382083100503,0.06202467678526288,10.572376251220703 +RPL29,9,0.018667539581656456,0.1006079614162445,0.9754366585380542,-0.044569609182268134,8.99472713470459 +RPL3,199,0.0013100861106067896,0.026476601138710976,0.9983716383850878,0.15922454076225948,2.3828372955322266 +RPL30,37,0.011353639885783195,0.07365565747022629,0.985453647064684,0.3412512540777235,7.014747142791748 +RPL31,30,0.020120110362768173,0.09942793101072311,0.9740111534419569,0.3939331963763527,9.338123321533203 +RPL32,21,0.009778732433915138,0.06999389827251434,0.987680121186429,0.18244692249996153,6.510071277618408 +RPL34,33,0.01236678846180439,0.07767374068498611,0.9833857713734625,0.38491824437141625,7.321043014526367 +RPL35,4,0.03283528611063957,0.13886192440986633,0.9577315868238758,0.17892714800763201,11.92929744720459 +RPL35A,13,0.022572709247469902,0.1078275591135025,0.969655535396602,0.33708927953668927,9.890911102294922 +RPL36,21,0.016011059284210205,0.08999504894018173,0.97890864538072,0.36864892685560374,8.330182075500488 +RPL36A,9,0.015271574258804321,0.09217517077922821,0.9800877363439647,0.2553797081939646,8.135540008544922 +RPL36AL,15,0.008719549514353275,0.07064099609851837,0.9887664899774088,0.010945278141058322,6.14739990234375 +RPL37,38,0.006976784206926823,0.05999187007546425,0.9909851686160538,0.2584830723802147,5.498852729797363 +RPL37A,38,0.01327369175851345,0.07966229319572449,0.9826640043581503,0.415820875757517,7.584733486175537 +RPL38,19,0.01565193012356758,0.08897047489881516,0.9798975567575506,0.29658710744830263,8.236228942871094 +RPL39,3,0.04885720834136009,0.17064498364925385,0.9387927959655297,0.048580246008955004,14.551534652709961 +RPL4,35,0.012499053962528706,0.07851255685091019,0.9835277784239538,0.415196155770921,7.360088348388672 +RPL41,4,0.030189668759703636,0.13176453113555908,0.9635607572540027,-0.09771646729415724,11.438620567321777 +RPL5,20,0.012718712911009789,0.08053858578205109,0.9830284536178739,0.3711343284803311,7.4244794845581055 +RPL6,28,0.020540956407785416,0.0988236516714096,0.9734716373206395,0.39344079717640645,9.435279846191406 +RPL7,30,0.021266547963023186,0.09948921948671341,0.9744082612623003,0.35655016998579847,9.600480079650879 +RPL7A,1,0.2688733637332916,0.3933224081993103,0.7266107189078281,0.15006274499281164,34.13644790649414 +RPL7L1,3,0.04249398782849312,0.1600516438484192,0.9494615088940598,0.09112714296197741,13.57088565826416 +RPL8,39,0.012051638215780258,0.07541485130786896,0.9842632846165563,0.36133612559220474,7.227157115936279 +RPL9,24,0.011617576703429222,0.07846551388502121,0.9849667157573563,0.3170289536519095,7.095813751220703 +RPLP0,19,0.016109121963381767,0.0898689478635788,0.9797753883267574,0.22231972445501735,8.35565185546875 +RPLP1,11,0.014236767776310444,0.08940873295068741,0.9807932018132809,0.09013947561983461,7.855071544647217 +RPLP2,22,0.013721554540097713,0.08296933770179749,0.9826870109034723,0.012964994242250203,7.711628437042236 +RPN1,5,0.02738255448639393,0.12597741186618805,0.9656018573278365,0.02896118666369325,10.893851280212402 +RPN2,11,0.01205433625727892,0.0836680456995964,0.9844788683752816,-0.021781167983216897,7.227965831756592 +RPP14,5,0.027884166687726974,0.12634220719337463,0.9627032314450965,0.16515314459681793,10.993178367614746 +RPP21,6,0.019528066739439964,0.10886918008327484,0.9761791529859445,0.0028325514334973474,9.199708938598633 +RPP30,10,0.02017546445131302,0.10673156380653381,0.9728721246570269,0.1665487001748056,9.350959777832031 +RPP38,5,0.03278953209519386,0.14056812226772308,0.95589530930008,0.0708155950892984,11.92098331451416 +RPP40,8,0.025854473933577538,0.1220000684261322,0.9653110710209716,0.01966077677782406,10.585522651672363 +RPS10,10,0.021070193499326706,0.10773517936468124,0.9726322485482548,0.16418629136347374,9.55605697631836 +RPS10-NUDT3,17,0.01985214650630951,0.10103914886713028,0.9739933095482742,0.24387462784468028,9.275732040405273 +RPS11,27,0.014103103429079056,0.08372292667627335,0.9817487659741847,0.1920677909895305,7.818110466003418 +RPS12,49,0.015478837303817272,0.08396503329277039,0.980969162069225,0.25235071032421935,8.190561294555664 +RPS13,13,0.028720304369926453,0.1216290146112442,0.9619870906009866,0.18964285258526653,11.156782150268555 +RPS14,18,0.015753110870718956,0.08985402435064316,0.978673890502132,0.20316624904476002,8.262807846069336 +RPS15,2,0.09527872502803802,0.24300892651081085,0.8835868800015848,0.09629473697707487,20.32087516784668 +RPS15A,32,0.013317856006324291,0.08054418116807938,0.9829161320149833,0.18100642478144804,7.597341060638428 +RPS16,5,0.03907473385334015,0.14865392446517944,0.9493868154631783,0.14245678326571778,13.013450622558594 +RPS17,23,0.010783337987959385,0.07621107995510101,0.9854574394831996,0.2287221756727864,6.836298942565918 +RPS18,12,0.025380900129675865,0.11415833234786987,0.9674252452283466,0.16611396971152712,10.488127708435059 +RPS19,42,0.012063364498317242,0.07497585564851761,0.9845467085819644,0.2750021478924484,7.230672359466553 +RPS19BP1,6,0.033165592700242996,0.13631530106067657,0.9574754395782994,0.01590445053829164,11.98914909362793 +RPS2,6,0.0236431285738945,0.118343785405159,0.9685687180051712,0.06892643012101136,10.122713088989258 +RPS20,28,0.008947228081524372,0.06960910558700562,0.9889255923664693,0.24722744512856898,6.2271409034729 +RPS21,7,0.0337502658367157,0.13596375286579132,0.9548072465022517,0.1236972965637751,12.094365119934082 +RPS23,12,0.02281002700328827,0.11052148789167404,0.9705497588614398,0.22495942363888288,9.942769050598145 +RPS24,25,0.020406130701303482,0.0991881862282753,0.9733097200139417,0.29431515305698414,9.40426254272461 +RPS25,4,0.033298417925834656,0.1383759081363678,0.9586915166552228,-0.007125308968295074,12.013132095336914 +RPS26,8,0.03742750734090805,0.14232395589351654,0.9531337533484989,0.22433045271351038,12.736201286315918 +RPS27,18,0.015922270715236664,0.09085818380117416,0.9795268579819278,0.1160934544281075,8.307052612304688 +RPS27A,9,0.020621715113520622,0.10718382894992828,0.9722752343985744,0.23179335101854817,9.45380973815918 +RPS28,31,0.011068313382565975,0.07512308657169342,0.9861811083969596,0.2171138507985472,6.9260430335998535 +RPS29,41,0.006690803449600935,0.059529416263103485,0.9915590110069217,0.1794321085104085,5.384973526000977 +RPS3,27,0.020929647609591484,0.10108396410942078,0.9738693725548199,0.30414834850892364,9.524131774902344 +RPS3A,16,0.016801675781607628,0.0932268351316452,0.9787872062914575,0.08269466706900146,8.533373832702637 +RPS4X,18,0.025007307529449463,0.11177413165569305,0.9688786681566285,0.23516519245013146,10.410651206970215 +RPS5,15,0.017547955736517906,0.0964965969324112,0.977804503158141,0.3052945265097312,8.72082805633545 +RPS6,18,0.022225359454751015,0.10475403815507889,0.9713075873318088,0.23023349404913832,9.814515113830566 +RPS7,27,0.020584916695952415,0.09906833618879318,0.9728685186487885,0.3048520526520591,9.4453706741333 +RPS8,2,0.11950170993804932,0.26521557569503784,0.8440304763390901,0.14404518712358025,22.757863998413086 +RPS9,27,0.01700913906097412,0.09169404208660126,0.978766696763445,0.23379786657504525,8.585895538330078 +RPSA,23,0.010623046196997166,0.07524339109659195,0.9864338493604703,0.17469791945116556,6.785298824310303 +RPTOR,8,0.021286578848958015,0.11148655414581299,0.9720710527195174,0.10315108744783712,9.605000495910645 +RPUSD3,17,0.008215603418648243,0.06940058618783951,0.9893726570273934,0.05417804803298256,5.967111587524414 +RPUSD4,12,0.010005582123994827,0.0772542729973793,0.9876420043369285,-0.0047551907924582495,6.585149765014648 +RRM1,11,0.016200603917241096,0.09312306344509125,0.9779029105025384,0.25358072147816374,8.37934398651123 +RRM2,57,0.0032433338928967714,0.042880505323410034,0.9955756772047925,0.2541144493887639,3.7492144107818604 +RRN3,4,0.05589751526713371,0.17480379343032837,0.9281148731375759,0.0801618584235004,15.564698219299316 +RRP1,11,0.013059758581221104,0.08770290017127991,0.9834945212019787,0.056554555176612474,7.52336311340332 +RRP12,21,0.015481281094253063,0.08972626179456711,0.9796140493410994,0.09887936215821555,8.191206932067871 +RRP15,3,0.04447837546467781,0.16543085873126984,0.942817618152671,0.13964817031819612,13.884137153625488 +RRP7A,4,0.04534858465194702,0.1623845100402832,0.9408726870970052,0.11071802485634988,14.019299507141113 +RRP9,11,0.013984999619424343,0.09070036560297012,0.9808848951668869,0.1183468206609955,7.785305976867676 +RRS1,7,0.023962901905179024,0.11829303205013275,0.9679691030155257,0.23372829550934485,10.190937995910645 +RSL1D1,21,0.0070851705968379974,0.06367678940296173,0.9908511613490222,0.10601529040723333,5.541401386260986 +RSL24D1,18,0.011012553237378597,0.07849273085594177,0.9852395430147219,0.3220926976496661,6.90857458114624 +RSRC2,10,0.014716371893882751,0.09312000125646591,0.9801649391003187,0.12233456090034237,7.986285209655762 +RTCB,17,0.00847212877124548,0.07057000696659088,0.9890648881440173,0.09277148316299333,6.059554576873779 +RTEL1,14,0.00931672565639019,0.07467175275087357,0.9877177461568052,0.03424081025078145,6.354422569274902 +RTF1,6,0.04465211555361748,0.15711970627307892,0.9460167719483489,0.26406193964004115,13.91122817993164 +RTTN,13,0.010680782608687878,0.08041395992040634,0.9861703658830864,-0.005250156893223365,6.803713321685791 +RUVBL1,21,0.019175909459590912,0.10086914896965027,0.973978542305903,0.4457014416901267,9.116379737854004 +RUVBL2,2,0.0699898973107338,0.20692317187786102,0.9152733903062632,0.016544794658480126,17.416549682617188 +SACM1L,20,0.007211714051663876,0.06535448133945465,0.9900746563317698,0.10169108550701272,5.590667724609375 +SAE1,12,0.014049474149942398,0.0855780616402626,0.9811076798256271,0.1810536238629793,7.803231716156006 +SAMM50,10,0.019780918955802917,0.10549608618021011,0.9731634547317021,0.4373645729037146,9.259076118469238 +SAP130,12,0.01268621627241373,0.08640236407518387,0.983008790757177,-0.050439359561831466,7.414988994598389 +SAP18,5,0.03325304016470909,0.1405806988477707,0.9560647721504912,0.0825250323177583,12.00494384765625 +SAP30BP,4,0.050844836980104446,0.17466957867145538,0.9356725387498047,0.1986338817912021,14.844579696655273 +SARNP,10,0.013540354557335377,0.08938506990671158,0.9826999470795813,-0.043692872859620986,7.660541534423828 +SARS,3,0.07567234337329865,0.21238605678081512,0.907441294954267,0.32522214271908184,18.109773635864258 +SARS2,17,0.008795971982181072,0.07154092192649841,0.9885089986870876,0.08357823468324707,6.174280166625977 +SART1,30,0.011776603758335114,0.08022681623697281,0.9858452543458873,0.36053322758640755,7.144215106964111 +SART3,16,0.015544977970421314,0.09303974360227585,0.980913680539181,0.16570110586742756,8.208040237426758 +SASS6,7,0.017365755513310432,0.10208487510681152,0.9769034983519933,0.07742496895057352,8.675436019897461 +SBDS,9,0.021416496485471725,0.11203496903181076,0.9723292784216816,0.06659688681546211,9.63426685333252 +SBNO1,19,0.008319769985973835,0.0699734091758728,0.9888704462145845,0.005790100229776699,6.00482177734375 +SCAF1,18,0.01079908199608326,0.0788344070315361,0.9854173119425361,0.4080019606857276,6.841288089752197 +SCAP,14,0.009820114821195602,0.0755695030093193,0.9872554599815163,-0.0011250462385381068,6.523831844329834 +SCD,3,0.04544033482670784,0.1666668802499771,0.9421938234768547,-0.037079849943110206,14.033474922180176 +SCFD1,15,0.009902969002723694,0.0756693184375763,0.9865334680365523,0.05982593237586035,6.551294803619385 +SCNM1,10,0.013361714780330658,0.08924704045057297,0.9829071827986047,-0.08791716954502372,7.609840393066406 +SDAD1,24,0.00928622204810381,0.070489801466465,0.9881710095149121,0.22761852444144332,6.3440117835998535 +SDC1,17,0.007615255191922188,0.06681666523218155,0.989897044094552,0.08099120325578925,5.744955539703369 +SDE2,15,0.009231102652847767,0.07334248721599579,0.9882177767388507,0.09542787835507546,6.325156211853027 +SDHA,6,0.024984538555145264,0.12024027109146118,0.9687321637261965,-0.0736017516988215,10.405911445617676 +SDHAF2,12,0.01225586049258709,0.08377594500780106,0.9844783316730693,-0.15669918135830796,7.2881340980529785 +SDHC,11,0.012988422997295856,0.08793582022190094,0.9829907268935306,0.023014433347029288,7.5027875900268555 +SDHD,5,0.024698913097381592,0.12180310487747192,0.9685529749738317,-0.04979327341032112,10.346260070800781 +SEC13,5,0.026324855163693428,0.12481993436813354,0.9665619406063601,0.03863639263840743,10.681382179260254 +SEC16A,16,0.009330014698207378,0.07421506196260452,0.9878036424855273,-0.11387301809511714,6.35895299911499 +SEC61A1,18,0.00990821048617363,0.07588331401348114,0.9873155595270282,0.31395063014334823,6.5530290603637695 +SEC61B,9,0.014782503247261047,0.09334083646535873,0.9810284535213146,0.033262395631765335,8.0042085647583 +SEC61G,10,0.01934555359184742,0.10543186962604523,0.974897553114499,0.33622328239305677,9.1566162109375 +SEC62,1,0.08513205498456955,0.23355641961097717,0.9060431314121745,-0.04348668316710252,19.208391189575195 +SEC63,8,0.017572494223713875,0.1016155481338501,0.9766837199389277,0.06936263182194377,8.726922988891602 +SEH1L,6,0.03131570667028427,0.13691021502017975,0.9611646794411731,0.13815882058922463,11.64999008178711 +SEM1,9,0.016482442617416382,0.09431755542755127,0.9796257387965305,0.10561041846125031,8.451916694641113 +SENP6,10,0.012434725649654865,0.08583074808120728,0.9844651024578251,0.10269540138912331,7.341124057769775 +SEPHS2,16,0.00859487522393465,0.07076603919267654,0.9888162730021004,-0.0733308319986972,6.103292942047119 +SEPSECS,8,0.015263279899954796,0.09526222944259644,0.9806004729706655,-0.059747469213407,8.133329391479492 +SERBP1,11,0.013599059544503689,0.08805234730243683,0.9821870137227784,-0.0924069063273706,7.677130222320557 +SET,1,0.06797833740711212,0.20924890041351318,0.9249499544906317,-0.055898218210433864,17.16444206237793 +SETD1A,7,0.0386013425886631,0.1474480926990509,0.9469337619418361,0.30621686314720614,12.934380531311035 +SETD2,11,0.011502225883305073,0.08239792287349701,0.9850229343708871,0.019606263352454942,7.0604987144470215 +SETDB1,13,0.01193921361118555,0.08268449455499649,0.9849868929344274,0.0689175439924191,7.193368434906006 +SF1,4,0.038994576781988144,0.15320847928524017,0.9509426690796543,0.14567304111311738,13.00009536743164 +SF3A1,33,0.004604047629982233,0.05195317044854164,0.994106080669508,0.05242099030822772,4.466983318328857 +SF3A2,32,0.009327522478997707,0.07001151889562607,0.9874610479754317,0.36597044797189054,6.358103275299072 +SF3A3,3,0.07004021108150482,0.20391422510147095,0.9109934454192633,0.23493310452330804,17.422807693481445 +SF3B1,17,0.007112374063581228,0.06493773311376572,0.9906223465635411,0.062106718510918965,5.552029132843018 +SF3B2,50,0.006029628217220306,0.057399868965148926,0.9924925918157826,0.2983611164065261,5.1119866371154785 +SF3B3,38,0.011352727189660072,0.0784047544002533,0.9859324534274645,0.3470590720679047,7.014464855194092 +SF3B4,24,0.007389454171061516,0.06483186781406403,0.9902721266007521,0.27759219268532287,5.65914249420166 +SF3B5,9,0.015371155925095081,0.0952746644616127,0.9797604874750615,0.12008265006700271,8.16202163696289 +SFPQ,31,0.01892184466123581,0.09673426300287247,0.9772455875158244,0.45325345164193115,9.055787086486816 +SFSWAP,17,0.009290257468819618,0.07387293875217438,0.9880694542765391,0.24662519483816855,6.345390319824219 +SGPP1,16,0.008330203592777252,0.07046379148960114,0.9895921009935486,0.05982912059536313,6.0085859298706055 +SHQ1,10,0.01193968765437603,0.0844823569059372,0.9847812838576149,-0.07001297454723716,7.193511962890625 +SIGLEC14,9,0.016041388735175133,0.09656702727079391,0.97997617498501,-0.1280653547108629,8.338068008422852 +SIN3A,30,0.005770503077656031,0.05691666156053543,0.9921841937881045,0.04071483825593017,5.000936031341553 +SKA1,7,0.02057691477239132,0.10813138633966446,0.973292424455482,0.02200790394149762,9.443534851074219 +SKA3,11,0.013903031125664711,0.08911972492933273,0.981678748154394,0.24518785252598502,7.762457370758057 +SKP1,6,0.03202889487147331,0.13459809124469757,0.9567539681318201,0.1316920507691227,11.781902313232422 +SKP2,20,0.006838671863079071,0.06335237622261047,0.9908603913908431,0.24569443583448675,5.444153308868408 +SLBP,9,0.018175480887293816,0.09847313910722733,0.9756045749508634,0.2893302595618514,8.875389099121094 +SLC16A5,18,0.008411403745412827,0.0702618733048439,0.9886160559585305,0.08400490152185723,6.03779935836792 +SLC1A5,49,0.0032861705403774977,0.042712870985269547,0.9957843184286614,-0.004577527472947553,3.7738921642303467 +SLC25A10,31,0.005024438723921776,0.05352362245321274,0.9931391119705745,0.3566801900778065,4.666467189788818 +SLC25A26,3,0.036350056529045105,0.14944347739219666,0.9553957413223583,0.016926807887550202,12.551538467407227 +SLC25A3,12,0.012025644071400166,0.08241552859544754,0.9845480317209512,-0.11532129129556633,7.219358921051025 +SLC35B1,7,0.01886853761970997,0.10719064623117447,0.9760441955992372,-0.05543714628793028,9.043023109436035 +SLC35G6,17,0.008297979831695557,0.06980255991220474,0.9892058284929891,-0.03478232520988003,5.996953010559082 +SLC39A10,16,0.009551526978611946,0.073881134390831,0.98736942851175,-0.08155636851620816,6.433996677398682 +SLC39A7,13,0.01256254967302084,0.08505872637033463,0.9830720682864195,0.15873877053788135,7.378759384155273 +SLC39A9,75,0.0024485315661877394,0.03668458014726639,0.9967146876680797,0.2972068940946243,3.257596492767334 +SLC4A5,22,0.006244679447263479,0.06059744581580162,0.9915803159176576,0.09996262202564231,5.202349662780762 +SLC7A5,38,0.004982460290193558,0.05250067636370659,0.993357256564979,0.30642111227075186,4.646932601928711 +SLC7A6OS,4,0.05907014012336731,0.18511854112148285,0.9248869733046996,0.20202733906598566,16.00031089782715 +SLC9B1,8,0.016785094514489174,0.1002277135848999,0.9783330477361192,-0.11224331055569373,8.52916145324707 +SLU7,23,0.032372426241636276,0.13002561032772064,0.9596364623268026,0.33984010347216553,11.844919204711914 +SMAGP,11,0.012185914441943169,0.08611083030700684,0.9836096546796937,0.17840776867878536,7.267307281494141 +SMARCB1,18,0.007460877299308777,0.06607141345739365,0.9901475639387465,0.13307385439594535,5.686426162719727 +SMARCE1,3,0.05096656084060669,0.17381519079208374,0.9382054765402386,-0.11167107393466334,14.862337112426758 +SMC1A,18,0.02306479960680008,0.10834670066833496,0.9682755757144467,0.323607983614157,9.99814224243164 +SMC2,14,0.010404575616121292,0.07722911983728409,0.9856374719510557,0.3834842868134547,6.715164661407471 +SMC3,6,0.03915161266922951,0.14870738983154297,0.9479168911161784,0.30379806224880285,13.026246070861816 +SMC4,34,0.004747577011585236,0.05091234669089317,0.9939131481382113,0.33447086997898345,4.536077499389648 +SMC5,12,0.011023640632629395,0.07872288674116135,0.9868103542855121,-0.14422418954253463,6.912051677703857 +SMC6,13,0.011124706827104092,0.08091103285551071,0.985153292108957,0.14204059865580904,6.943665027618408 +SMG1,3,0.05068569257855415,0.17529889941215515,0.9362943496889395,0.21466563022643692,14.821329116821289 +SMG5,9,0.020605139434337616,0.10183518379926682,0.9729008776664879,0.17087884212295856,9.4500093460083 +SMG6,6,0.022519104182720184,0.11413879692554474,0.971374391742238,0.08388080526859058,9.879159927368164 +SMN2,32,0.018777813762426376,0.09652794897556305,0.9765284346721211,0.42510603375411693,9.021255493164062 +SMNDC1,1,0.12893922626972198,0.2864246070384979,0.8597685533741373,0.0601171214730747,23.63943099975586 +SMU1,15,0.014494996517896652,0.0892835408449173,0.980182054994024,0.33383169419115327,7.925990104675293 +SMUG1,7,0.0223202146589756,0.11302472651004791,0.9715127489063482,-0.04341088012732308,9.835436820983887 +SNAPC1,9,0.02115006558597088,0.10814975947141647,0.9719382735586621,0.21180044479216997,9.574151992797852 +SNAPC2,7,0.023344676941633224,0.11601000279188156,0.9688602071461873,0.10747966217683848,10.05862045288086 +SNAPC3,21,0.008111766539514065,0.0670776292681694,0.9890605970165325,0.06661781048627889,5.929283618927002 +SNAPC4,11,0.013532762415707111,0.08903985470533371,0.9817554504285179,0.1091481463983388,7.658393859863281 +SNAPC5,10,0.013532065786421299,0.08981345593929291,0.9825601042147056,0.03641756595223652,7.658195972442627 +SNF8,13,0.010997178964316845,0.0799252986907959,0.9853648117437107,-0.07856192201825296,6.903750896453857 +SNIP1,5,0.030188236385583878,0.13302138447761536,0.9605096590701812,0.26656706448003814,11.438348770141602 +SNRNP200,17,0.015290309675037861,0.09178491681814194,0.9794900390974893,0.34612157363624,8.140528678894043 +SNRNP25,13,0.01100330613553524,0.07940410822629929,0.9854495539224846,0.06732412925601255,6.905673503875732 +SNRNP27,4,0.0440327450633049,0.16227106750011444,0.9445341829699023,0.0237925181322913,13.814409255981445 +SNRNP35,17,0.01208829041570425,0.08251434564590454,0.983678264450784,0.2800509662079618,7.238138675689697 +SNRNP40,21,0.007027327083051205,0.06361424922943115,0.9909979630749971,-0.0019870068845619353,5.518735408782959 +SNRNP48,19,0.010279135778546333,0.07541567832231522,0.9861015758341481,0.1542889318138353,6.674561500549316 +SNRNP70,13,0.019641131162643433,0.10479036718606949,0.9727849388271176,0.4357244931672975,9.226303100585938 +SNRPA1,16,0.024249283596873283,0.11371704190969467,0.9716849688541156,0.3372996103422001,10.251652717590332 +SNRPB,27,0.010788931511342525,0.07716520875692368,0.9867989933140001,0.36287977932814774,6.838071823120117 +SNRPC,4,0.05502982810139656,0.177860826253891,0.925289164350781,0.28421246154169993,15.443422317504883 +SNRPD1,14,0.013830497860908508,0.08906420320272446,0.9825652639891075,0.32545879293191887,7.742181777954102 +SNRPD2,2,0.09384793043136597,0.2366250455379486,0.8923107008416639,0.40964481422769666,20.167720794677734 +SNRPD3,18,0.02629701979458332,0.11737708002328873,0.9664158965688183,0.4252177284579694,10.67573356628418 +SNRPE,19,0.02154851146042347,0.10744334757328033,0.9728464879036282,0.43008216425386825,9.663914680480957 +SNRPF,13,0.03388732671737671,0.13369280099868774,0.9587237833952178,0.367291274147531,12.118896484375 +SNRPG,40,0.01057109609246254,0.07415280491113663,0.9855367724655245,0.4444976696010937,6.768687725067139 +SNUPN,14,0.012201907113194466,0.08493002504110336,0.9839116836554083,0.10258464366231763,7.2720746994018555 +SNW1,22,0.007444088812917471,0.06566095352172852,0.9899426866497388,-0.009507695499296061,5.680024147033691 +SNX15,63,0.002733559813350439,0.03912247344851494,0.9962697214786624,0.17577147269772603,3.441983222961426 +SOD1,17,0.008495263755321503,0.0697174146771431,0.9884472638846256,0.15992216201677956,6.0678229331970215 +SOD2,20,0.006969911511987448,0.0642121285200119,0.9908910088162607,0.08086813887140609,5.496143817901611 +SON,51,0.008180474862456322,0.06471942365169525,0.9893390856100541,0.5084242802703112,5.954340934753418 +SP2,12,0.01271600741893053,0.08502276241779327,0.9833494888425265,-0.00853563719863442,7.423689842224121 +SPATA5,7,0.01769198849797249,0.10249299556016922,0.9772254447362135,-0.028477757017658492,8.756545066833496 +SPATA5L1,15,0.010932227596640587,0.08026320487260818,0.9849634993780143,0.21470639103600017,6.883333683013916 +SPC24,10,0.015620487742125988,0.0942435935139656,0.9784499910410709,0.3276867251049504,8.227952003479004 +SPC25,20,0.021376946941018105,0.10572197288274765,0.9711226196378937,0.4965921519284438,9.625367164611816 +SPCS2,12,0.01019352301955223,0.07629159092903137,0.9870676023564409,0.01628463604202295,6.6467084884643555 +SPCS3,16,0.008966250345110893,0.07126719504594803,0.9885930628364302,-0.05511164251933409,6.233757495880127 +SPDL1,9,0.016337646171450615,0.09654022753238678,0.9778216758748505,0.1156998551165972,8.414710998535156 +SPDYE2,18,0.007770214229822159,0.06739407032728195,0.9901182961655357,-0.0802969934453352,5.803111553192139 +SPRTN,14,0.010226244106888771,0.07680833339691162,0.986552979099116,-0.07345455007138935,6.65736722946167 +SPTLC1,3,0.03672075271606445,0.14943231642246246,0.9538308249696051,0.07775914451170411,12.615377426147461 +SPTSSA,13,0.010416611097753048,0.07838715612888336,0.9862941825806001,0.0019434200574234414,6.719046592712402 +SRBD1,15,0.007631285116076469,0.06757526099681854,0.9904571508174645,0.06512508197848192,5.750998497009277 +SRCAP,20,0.008111209608614445,0.06893382966518402,0.9893038129844473,0.14922791531861632,5.929079055786133 +SRF,13,0.010886621661484241,0.08037129044532776,0.9864806308942706,-0.09690120850271822,6.868960380554199 +SRFBP1,7,0.02625117637217045,0.12369195371866226,0.9649571133862339,-0.038145945044176795,10.666423797607422 +SRP14,13,0.01269601285457611,0.08458071202039719,0.9837356942521543,-0.008583671000718053,7.417851448059082 +SRP19,13,0.00987269263714552,0.07589327543973923,0.9872018705831793,0.11677097286408843,6.5412726402282715 +SRP54,8,0.020251406356692314,0.11074836552143097,0.9740566128940344,0.2891797291033499,9.368542671203613 +SRP68,17,0.014180527999997139,0.08852759003639221,0.9816212996931521,0.4426249259403404,7.839541435241699 +SRP72,23,0.0076793949119746685,0.06588562577962875,0.9895169910763264,0.3345256162707594,5.76909875869751 +SRP9,17,0.009494646452367306,0.07226556539535522,0.9872749523522204,0.16291438666522387,6.414811134338379 +SRPRB,17,0.011912933550775051,0.08316006511449814,0.9854437015392978,0.1661106858901444,7.1854472160339355 +SRRM1,18,0.00921401847153902,0.0721970871090889,0.9879848675056219,0.08089869199413405,6.319300174713135 +SRRM2,18,0.01065005362033844,0.07623616605997086,0.9854225291511426,0.16886820145330406,6.793919086456299 +SRRT,30,0.01853771321475506,0.0952184647321701,0.974891823463933,0.26519859078325336,8.963396072387695 +SRSF1,16,0.014256884343922138,0.08720957487821579,0.9804419846777739,0.16207082216089383,7.860619068145752 +SRSF10,14,0.014641936868429184,0.08948497474193573,0.9803431439714053,0.20964388119941696,7.966062545776367 +SRSF11,43,0.0038555499631911516,0.04706636071205139,0.9950209950621879,-0.08769775653523972,4.0877814292907715 +SRSF2,7,0.022487269714474678,0.11293930560350418,0.9708379610263016,0.0667712093519152,9.872174263000488 +SRSF3,3,0.06127534061670303,0.18929649889469147,0.9208745188697178,0.17962216178046714,16.29623794555664 +SRSF6,23,0.006290549412369728,0.060089316219091415,0.9918698412465301,-0.06597604420120677,5.221421241760254 +SRSF7,12,0.014468232169747353,0.08969563245773315,0.9817460009456432,0.029301405142675516,7.9186692237854 +SS18L2,15,0.012093593366444111,0.0819631889462471,0.9835598594329074,0.15737728781276178,7.2397260665893555 +SSBP1,12,0.015365786850452423,0.09124364703893661,0.9816166460628967,0.06999505714537928,8.160595893859863 +SSBP3,10,0.01384098269045353,0.08999694138765335,0.9817148356142702,0.024925722577197145,7.745115756988525 +SSRP1,56,0.008575358428061008,0.06551270186901093,0.9887048470727942,0.31835165511425517,6.096359729766846 +SSU72,9,0.020391147583723068,0.10786446183919907,0.9726421579556105,0.1736620237924357,9.400810241699219 +SSX4,10,0.013981581665575504,0.09082869440317154,0.9821254481728323,-0.0762889900564377,7.7843546867370605 +STARD7,11,0.013498689979314804,0.08756014704704285,0.9835808396240937,0.09997151098406647,7.648746490478516 +STAT5A,21,0.006678865756839514,0.06265442073345184,0.9908476881124358,0.2737826059926242,5.3801679611206055 +STIL,17,0.008669370785355568,0.07105761021375656,0.9884635428001826,-0.012374284652626763,6.129685878753662 +STRAP,10,0.015020119957625866,0.09238007664680481,0.9809728631457472,0.06943698025289417,8.068283081054688 +STRIP1,10,0.013144796714186668,0.08728352189064026,0.98358438802438,-0.11874788432403437,7.547817230224609 +STX18,8,0.018648957833647728,0.10395700484514236,0.974541965952011,0.29183791927584823,8.990248680114746 +STX5,5,0.029197227209806442,0.13216981291770935,0.9610175139635524,0.01975304702257984,11.249034881591797 +SUGP1,15,0.009892569854855537,0.07632908225059509,0.9872410620852392,0.005570857995478389,6.547854900360107 +SUGT1,10,0.015032375231385231,0.09268427640199661,0.9803757396719918,-0.08402641421273473,8.071574211120605 +SUMO2,7,0.018856098875403404,0.10467088222503662,0.9759514965535327,0.0287631845259487,9.040040016174316 +SUPT16H,10,0.035909175872802734,0.1389482021331787,0.9524481423317693,0.3327855519907485,12.475190162658691 +SUPT4H1,8,0.023287009447813034,0.11560013890266418,0.9681963780853229,0.014243973820494485,10.046188354492188 +SUPT5H,40,0.034642744809389114,0.12844179570674896,0.9583048928644176,0.32142648808822916,12.253230094909668 +SUPT6H,29,0.039195768535137177,0.1382206529378891,0.9504123643328904,0.3561204882522083,13.033589363098145 +SUPV3L1,8,0.02152353525161743,0.10869903862476349,0.9722810777435106,0.2378359502004524,9.658312797546387 +SURF6,13,0.01135246828198433,0.0811467170715332,0.9852908509395633,0.12364323980093991,7.014384746551514 +SYF2,8,0.01696363277733326,0.0992889478802681,0.9786502663088393,0.19571398394681194,8.574402809143066 +SYMPK,9,0.032540105283260345,0.13245056569576263,0.9560750518945061,0.27038838427128664,11.875555992126465 +SYS1,6,0.023533327504992485,0.11688999831676483,0.9687160900192913,0.12327889420415314,10.099180221557617 +TACC3,13,0.010298173874616623,0.07722753286361694,0.9866746739351279,0.04642024728171935,6.680739879608154 +TADA2A,4,0.032301709055900574,0.13848569989204407,0.9605731560635282,0.050978466928940404,11.831974029541016 +TADA3,15,0.010149678215384483,0.07684574276208878,0.9866856355823806,-0.027478486754200154,6.6323981285095215 +TAF1,9,0.03397699445486069,0.13683855533599854,0.9547352893688206,0.12741272229680758,12.134919166564941 +TAF10,4,0.05837028846144676,0.18512894213199615,0.9233034403732128,0.11123640022667472,15.905245780944824 +TAF12,6,0.04903869703412056,0.1641102433204651,0.9309931114303434,0.20585865492764896,14.578536033630371 +TAF1B,10,0.015603587031364441,0.09611198306083679,0.9795662755568861,-0.04198058887878837,8.223499298095703 +TAF1C,14,0.012716817669570446,0.0856352299451828,0.983413376162003,-0.04080498651984601,7.423926830291748 +TAF1D,14,0.010905073955655098,0.07881882786750793,0.9855200963754516,0.08251404979178138,6.87477970123291 +TAF2,15,0.02599012851715088,0.11469068378210068,0.9651058118508706,0.1547009821917522,10.613256454467773 +TAF5,3,0.09748861193656921,0.2354317158460617,0.8698027167079179,0.189426284995415,20.555185317993164 +TAF6,12,0.022704467177391052,0.10826634615659714,0.96848203998284,0.0169647789215145,9.9197359085083 +TAF7,10,0.022763554006814957,0.11301912367343903,0.9690428188806189,0.0662281760298184,9.932635307312012 +TAF8,11,0.02638683095574379,0.11818434298038483,0.9636850992878426,0.04807695428594332,10.693947792053223 +TAMM41,12,0.010436651296913624,0.07958102226257324,0.9871933534150445,0.021003241346440944,6.7255072593688965 +TANGO6,9,0.023159220814704895,0.11550066620111465,0.9696787814689982,0.18647134482261668,10.018586158752441 +TARDBP,17,0.00945816095918417,0.07356928288936615,0.9870580197430782,0.0951905148616422,6.402473449707031 +TARS,13,0.014555408619344234,0.09053163975477219,0.9806016320878458,0.3272334905325534,7.9424896240234375 +TARS2,8,0.019511865451931953,0.10736220329999924,0.9754737583031715,0.10197107703965702,9.195891380310059 +TBC1D28,8,0.013407598249614239,0.0893218070268631,0.9833074103447853,0.025449228888593974,7.622895240783691 +TBC1D3,3,0.05357621982693672,0.18113365769386292,0.9330857695848543,0.13638652786171496,15.238088607788086 +TBCA,18,0.008810390718281269,0.07106306403875351,0.9885722787067097,0.14833128421453853,6.1793389320373535 +TBCB,10,0.015367839485406876,0.09443670511245728,0.9798857583700021,-0.0038461084329370673,8.161140441894531 +TBCC,14,0.009750466793775558,0.07508432865142822,0.9881257120716082,-0.13283232764370237,6.500655651092529 +TBCD,3,0.03993479162454605,0.1546785980463028,0.9527268524140746,-0.06723750812774101,13.155887603759766 +TBCE,13,0.010453004390001297,0.07872768491506577,0.9865032271555106,0.006298616050684612,6.730774402618408 +TBL3,7,0.023931298404932022,0.11646293848752975,0.9692670938334675,-0.009260771057583644,10.184215545654297 +TBP,15,0.01229230035096407,0.08244750648736954,0.9836233072248084,-0.07526017664489978,7.2989606857299805 +TCERG1,11,0.01310065295547247,0.08561516553163528,0.9827225747721526,0.30561974608857007,7.535132884979248 +TCOF1,9,0.019423743709921837,0.10750193893909454,0.9746578400340479,0.15020007586197023,9.175102233886719 +TCP1,6,0.035007916390895844,0.1426468789577484,0.9565447203759055,0.138776007561193,12.317642211914062 +TDGF1,10,0.012537751346826553,0.08608336001634598,0.9834295511598719,0.10206593384661447,7.37147331237793 +TEAD3,26,0.005430696997791529,0.05607299134135246,0.9931195804731163,-0.05199062184958663,4.851457595825195 +TEFM,21,0.008630146272480488,0.06822765618562698,0.9899980473378183,0.013748199961270672,6.115803241729736 +TELO2,8,0.018479857593774796,0.1040075495839119,0.9753359993345025,0.2156435695523846,8.949397087097168 +TEN1,8,0.015552881173789501,0.09626546502113342,0.9792477410965589,0.09412434054425355,8.210126876831055 +TERF1,9,0.014287593774497509,0.09207798540592194,0.9820344708742709,-0.11414710309804349,7.869080066680908 +TERF2,5,0.027966754510998726,0.12717300653457642,0.9647509475172426,0.11231000781425084,11.00944709777832 +TEX10,16,0.0108908386901021,0.07828571647405624,0.9865927894676724,0.08163932588751938,6.870290279388428 +TFAM,8,0.022199459373950958,0.11227189749479294,0.9709294929184231,0.13636328147034915,9.808794975280762 +TFB2M,16,0.00789661705493927,0.06781477481126785,0.9896180904948365,0.13020308205927497,5.850122928619385 +TFDP1,17,0.009303729049861431,0.07258407771587372,0.987389979650087,0.06595873321123828,6.34998893737793 +TFIP11,6,0.03816103935241699,0.14867928624153137,0.9549919045811296,0.17870235265652581,12.86040210723877 +TFRC,5,0.024725880473852158,0.1214999258518219,0.9675552597931533,0.10800786067176178,10.351906776428223 +TGS1,16,0.008571485988795757,0.07115296274423599,0.988824204025559,-0.004677854314495948,6.094982624053955 +THAP1,11,0.01456918753683567,0.09159333258867264,0.9801360857053641,0.1912435708375827,7.946248531341553 +THAP11,7,0.018642457202076912,0.10346457362174988,0.9758457804834306,-0.011824208172364261,8.988682746887207 +THG1L,5,0.032717879861593246,0.13822175562381744,0.960338399249852,0.29054632759161564,11.907950401306152 +THOC1,4,0.04488270357251167,0.1606660783290863,0.9397865690957856,0.19631518973290368,13.947101593017578 +THOC2,19,0.01530875638127327,0.09042427688837051,0.9798539995770985,0.35901407970212107,8.145437240600586 +THOC3,14,0.012213706970214844,0.08254079520702362,0.9837852605569101,0.05400104253430481,7.275589942932129 +THOC5,3,0.05181995406746864,0.17544259130954742,0.9362731806505434,0.20358986101506865,14.986249923706055 +THOC6,11,0.011867891997098923,0.08389490842819214,0.9840931543707064,0.050879088590410515,7.171850681304932 +THOC7,13,0.011048437096178532,0.07999695837497711,0.9856409403875408,0.16904679903985356,6.919821262359619 +TICRR,15,0.012338765896856785,0.08086773753166199,0.9829794205211132,0.22195221144935226,7.312743186950684 +TIGD1,6,0.02118530124425888,0.11109822243452072,0.973159720889907,-0.053002446613884154,9.582123756408691 +TIMELESS,7,0.01506605464965105,0.09415115416049957,0.9808768135811385,0.04894949254669568,8.080611228942871 +TIMM10,6,0.019783036783337593,0.1082082986831665,0.9755573703217361,0.029955873653016652,9.25957202911377 +TIMM13,18,0.0119324941188097,0.07965333014726639,0.9839208012342258,0.34684408136684436,7.191343784332275 +TIMM22,8,0.01616242714226246,0.09761101752519608,0.9797542763588782,-0.08608126551092098,8.369465827941895 +TIMM23B,50,0.017156777903437614,0.08869113773107529,0.9773342387433501,0.6158797166138726,8.623078346252441 +TIMM44,5,0.02465188503265381,0.11971672624349594,0.9677805483341508,0.15340908628122646,10.336404800415039 +TIMM8A,16,0.008142556995153427,0.06902895867824554,0.9894646339494778,0.06734404364544612,5.940525054931641 +TIMM9,20,0.007379862014204264,0.06553266197443008,0.9900815091253372,0.25851827950104533,5.655468463897705 +TINF2,7,0.021552851423621178,0.1125391349196434,0.9705322822999636,0.19053382192359436,9.664888381958008 +TIPIN,11,0.011356879957020283,0.0808049663901329,0.9856926114582593,-0.11297708677172201,7.015748023986816 +TKT,20,0.007781952619552612,0.06729301065206528,0.9895652601347653,-0.008590925410721038,5.807493209838867 +TLCD1,15,0.008723230101168156,0.07174419611692429,0.9889268785743796,-0.054309745745551696,6.148697853088379 +TLN1,8,0.016341280192136765,0.09835246205329895,0.9789332345969697,-0.019181044145979168,8.41564655303955 +TM7SF2,10,0.013618916273117065,0.08850327134132385,0.9820913885555118,0.024025710268225677,7.682732582092285 +TMA16,17,0.008338685147464275,0.06910622119903564,0.9900072318936827,0.14909981181867835,6.011643886566162 +TMED2,13,0.010023689828813076,0.07700133323669434,0.9872645480042825,0.010630609593068295,6.5911054611206055 +TMEM127,13,0.010507306084036827,0.07908120006322861,0.9863082312899423,0.036050387608963876,6.748234748840332 +TMEM199,10,0.012422441504895687,0.08588384091854095,0.9841232128421491,-0.028917397163659583,7.337496757507324 +TMEM240,17,0.007939939387142658,0.06950125843286514,0.9899241472482082,0.0018681172709304299,5.866148471832275 +TMEM258,20,0.007357167545706034,0.06550733745098114,0.9902706929020104,0.14945155101216806,5.64676570892334 +TMSB10,23,0.006122077815234661,0.05855239927768707,0.9918145875389198,0.07007768566630403,5.151027202606201 +TMX2,11,0.017450913786888123,0.10052676498889923,0.9761403088813447,0.23286855248639945,8.696681022644043 +TNNT2,3,0.033148083835840225,0.1414029449224472,0.9592448440258734,-0.043540917835386134,11.985982894897461 +TNPO3,10,0.015377293340861797,0.09488704800605774,0.9802251616626391,-0.0928277118403412,8.163650512695312 +TOE1,14,0.011146843433380127,0.08136000484228134,0.9853855294684862,-0.030790224648081108,6.9505696296691895 +TOMM20,11,0.014960994012653828,0.09217751026153564,0.980219457360021,0.05800668503817753,8.052388191223145 +TOMM22,11,0.015802010893821716,0.09561146050691605,0.9804252242902682,0.26848256417762373,8.275622367858887 +TOMM40,11,0.014331222511827946,0.0907917395234108,0.9820661785710847,0.08435747840184844,7.881086349487305 +TONSL,3,0.034645549952983856,0.14435964822769165,0.9575832310874456,-0.020082980751235213,12.2537260055542 +TOP1,13,0.013677265495061874,0.08918147534132004,0.9815665798109032,-0.00551582102723542,7.6991729736328125 +TOP2A,4,0.0344337560236454,0.140372171998024,0.9562845396955841,0.1174528587991617,12.216214179992676 +TOP3A,16,0.008648497983813286,0.07229629904031754,0.9885684582601892,-0.01936021452794487,6.122303009033203 +TOPBP1,5,0.026540018618106842,0.1257091462612152,0.9651058210188671,0.18734767065362046,10.724944114685059 +TOR2A,5,0.03159550204873085,0.136213019490242,0.959607079218747,-0.05410447675670009,11.701918601989746 +TP53RK,9,0.013238013722002506,0.08906589448451996,0.9833422631133012,0.01920376734623943,7.574532985687256 +TPI1,4,0.03383143991231918,0.14165855944156647,0.9571468956090273,0.04384312392710157,12.10890007019043 +TPR,7,0.02352900616824627,0.11523233354091644,0.9683212246876479,0.1509874493995282,10.09825325012207 +TPRKB,4,0.03733165189623833,0.15048137307167053,0.952197321762939,0.17622015791821996,12.719882011413574 +TPT1,5,0.031355682760477066,0.1349741816520691,0.9601887132803697,-0.08521809820176099,11.65742301940918 +TPX2,36,0.007983319461345673,0.0661562979221344,0.9899700576564593,0.3917035292965806,5.882151126861572 +TRA2B,22,0.008442599326372147,0.06803255528211594,0.9891547763067889,0.17812257429817396,6.048985481262207 +TRAIP,8,0.01680346578359604,0.09804175049066544,0.9790881978226381,-0.03986348850037979,8.533828735351562 +TRAPPC1,10,0.013604005798697472,0.09045201539993286,0.9819091602516995,0.1538006812367719,7.678525924682617 +TRAPPC11,11,0.01230649370700121,0.08586276322603226,0.9837126966595144,0.17054982457056214,7.303174018859863 +TRAPPC3,17,0.00731827924028039,0.06527438014745712,0.9902823787290223,0.13262511060798338,5.631822109222412 +TRAPPC4,19,0.007716836873441935,0.06683073937892914,0.9898480168790482,0.01959193819373532,5.783145427703857 +TRAPPC5,13,0.009911232627928257,0.0754658579826355,0.986986753674252,0.08515084552761788,6.554028034210205 +TRAPPC8,15,0.008935828693211079,0.07193421572446823,0.9881644644332241,0.125681074820698,6.223173141479492 +TREML2,6,0.024691911414265633,0.12155969440937042,0.9685127814209089,-0.0024162035304476615,10.344793319702148 +TREX2,27,0.007185542955994606,0.06372468173503876,0.9902063034006078,0.34681294587388256,5.580514430999756 +TRIAP1,6,0.021813176572322845,0.11491794884204865,0.9737754169284916,0.07593145259604807,9.723081588745117 +TRIM49C,9,0.01509122084826231,0.09441427141427994,0.9804668395326348,0.016419011646040916,8.087356567382812 +TRMT10C,9,0.016316598281264305,0.09636234492063522,0.9795468503168798,0.16615806122334253,8.409289360046387 +TRMT112,20,0.011199422180652618,0.07851774245500565,0.9853001614490431,0.1313214810993201,6.966943264007568 +TRMT5,10,0.018030909821391106,0.10159894824028015,0.9765077335152617,0.2858804685222091,8.840020179748535 +TRMT6,12,0.01065501943230629,0.07919466495513916,0.9859611442573872,0.05885028704320138,6.795502185821533 +TRNP1,15,0.009498397819697857,0.07469722628593445,0.9875865333795739,-0.0884148648868573,6.416077613830566 +TRNT1,21,0.006651474162936211,0.06284656375646591,0.9911611924769671,0.12068198324421092,5.369123935699463 +TRPM7,13,0.011376050300896168,0.08139313757419586,0.9849464131436306,0.030314060535900772,7.021666526794434 +TRRAP,42,0.011371725238859653,0.07423219084739685,0.9846976677605405,0.3617872883336215,7.020331859588623 +TSEN2,17,0.009744130074977875,0.0738549679517746,0.9877495504566514,0.20292942190690594,6.498542785644531 +TSEN34,27,0.004830645862966776,0.05390601605176926,0.9938036077396392,0.012955519850629399,4.575589656829834 +TSEN54,5,0.026470523327589035,0.1242271289229393,0.9654838005166626,0.07593934564865629,10.710893630981445 +TSFM,10,0.01414169929921627,0.09214983135461807,0.9811634709162641,0.06426142545018025,7.828801155090332 +TSG101,8,0.026278870180249214,0.12107788026332855,0.9646903157174993,0.3452564433323729,10.672049522399902 +TSPYL5,10,0.014741980470716953,0.09420062601566315,0.9809294553189766,-0.010391362040545277,7.993230819702148 +TSR1,19,0.010033834725618362,0.07508181780576706,0.9866365105333724,0.16530363723487917,6.59443998336792 +TSR2,53,0.013251625001430511,0.08130565285682678,0.9838917618512463,0.3188084339239664,7.578425884246826 +TSSK3,21,0.006439092569053173,0.061786238104104996,0.99148689602875,0.04686138045537123,5.282710075378418 +TTC1,7,0.019262144342064857,0.10764122754335403,0.9756966117772763,0.12489050134350482,9.136856079101562 +TTC27,9,0.021796727553009987,0.11100640892982483,0.9730933766738021,-0.07095345303360831,9.719414710998535 +TTC4,11,0.01398180890828371,0.09139589220285416,0.9820120846270091,-0.09996022169253613,7.784418106079102 +TTF2,7,0.017652150243520737,0.10318481922149658,0.9778824059677235,0.005430998976398384,8.74668025970459 +TTI1,15,0.013283725827932358,0.0871470645070076,0.9816365979004853,0.2672388331474676,7.587599754333496 +TTI2,5,0.029761552810668945,0.13217946887016296,0.9608929260525811,0.058123980702458965,11.357224464416504 +TTK,20,0.010126667097210884,0.07544383406639099,0.9862900187787721,0.4092566677771817,6.624875068664551 +TUBA1B,8,0.020828373730182648,0.10887370258569717,0.9718825719335725,0.16744811853817468,9.50106143951416 +TUBA1C,5,0.022051339969038963,0.11316630989313126,0.9752074576382486,-0.13095861690716168,9.776017189025879 +TUBA3D,7,0.016731426119804382,0.09872116148471832,0.9793496566370242,-0.13321920829301814,8.515515327453613 +TUBB,10,0.012516975402832031,0.0857134759426117,0.9834751402214474,0.11942114968207386,7.365362644195557 +TUBD1,19,0.007834818214178085,0.06721103936433792,0.9897062262799888,-0.04055198799378675,5.827186584472656 +TUBE1,14,0.009972273372113705,0.076640285551548,0.9869364942606146,0.013756039436049151,6.574179172515869 +TUBG1,2,0.0840904489159584,0.21480409801006317,0.9051907968857225,-0.01880346156491893,19.09052085876465 +TUBGCP2,24,0.006245517171919346,0.05999884009361267,0.991507995952368,0.1466021390499132,5.202698707580566 +TUBGCP3,3,0.04079175740480423,0.15333124995231628,0.948733457265379,0.013630025993974537,13.296295166015625 +TUBGCP4,4,0.03878512606024742,0.1531182825565338,0.9485070959834652,0.16964903650666277,12.96513557434082 +TUBGCP5,7,0.016696371138095856,0.10048845410346985,0.9786841414741554,0.15271234042786117,8.506589889526367 +TUBGCP6,16,0.009175202809274197,0.07288374751806259,0.9879080041617819,-0.07300364210432861,6.305975914001465 +TULP1,13,0.010786053724586964,0.0796990767121315,0.9859640584913146,-0.045383870005246754,6.837159633636475 +TUT1,21,0.006616397760808468,0.06228424236178398,0.9914723753388025,0.006185836157165445,5.354948043823242 +TVP23C,12,0.013256714679300785,0.0880705714225769,0.9825857659501711,-0.05765510452684351,7.579881191253662 +TWF1,14,0.009849943220615387,0.07647671550512314,0.9868627547858139,0.018410733490229935,6.533731937408447 +TWISTNB,17,0.0226520337164402,0.10794002562761307,0.9708876942869152,0.1844876319881129,9.908273696899414 +TXN,10,0.013499812223017216,0.08835228532552719,0.9818390531542986,0.1034838523564953,7.649064540863037 +TXNL4A,9,0.015541203320026398,0.09519749134778976,0.9800561519385812,0.14476866492182228,8.20704460144043 +TXNL4B,13,0.009971587918698788,0.07680705189704895,0.986997792418746,0.011412679364347308,6.573953151702881 +U2AF1,14,0.02995886094868183,0.12540748715400696,0.960088880526199,0.3850693820375077,11.394810676574707 +U2AF2,10,0.014812123030424118,0.09474483877420425,0.9812755381391274,0.151837032015367,8.012224197387695 +U2SURP,29,0.011724471114575863,0.0774233490228653,0.9851280621586963,0.3243690499070294,7.128384113311768 +UBA1,28,0.017250586301088333,0.09236106276512146,0.9761415620241052,0.233455844790408,8.646620750427246 +UBA2,17,0.011377759277820587,0.07706444710493088,0.9850841727830363,0.14585489659065187,7.022193908691406 +UBA3,9,0.019034884870052338,0.1057778000831604,0.9752319215142722,-0.05451081410170547,9.082796096801758 +UBA5,10,0.012066320516169071,0.08352447301149368,0.9846339971737079,0.04327316567078736,7.231558322906494 +UBA52,6,0.02704175002872944,0.12532517313957214,0.9654027076009389,0.04499191692515979,10.825845718383789 +UBAP1,11,0.015964090824127197,0.0945318192243576,0.9783753258430701,0.22654691043962155,8.317954063415527 +UBC,10,0.01435830444097519,0.09141569584608078,0.9814112938701478,-0.03968983005279134,7.888528823852539 +UBE2D3,14,0.009258843027055264,0.07230384647846222,0.9879062349312803,0.12827079139039024,6.334652900695801 +UBE2H,20,0.007204448338598013,0.06398093700408936,0.9905915932084011,0.01777665515852286,5.587851047515869 +UBE2I,7,0.022643493488430977,0.11286604404449463,0.9695592957111155,0.22932884464389575,9.90640640258789 +UBE2L3,20,0.00785864144563675,0.06455983966588974,0.9894828197867274,0.07139094921129407,5.836039066314697 +UBE2M,17,0.00818493403494358,0.06928849220275879,0.9892534734466357,0.041382435933682626,5.955963611602783 +UBE2N,18,0.007730467244982719,0.06684580445289612,0.9901566162447749,0.012351382523702218,5.78825044631958 +UBE2Z,7,0.019818058237433434,0.10971144586801529,0.9736536934612479,0.07193493859780885,9.267765045166016 +UBL5,23,0.010987324640154839,0.07747252285480499,0.9860224820434739,0.38466669782736906,6.900656700134277 +UBQLN4,15,0.011821797117590904,0.0825318843126297,0.983846784363632,0.19793474583018247,7.157909870147705 +UBR4,2,0.049304232001304626,0.17747391760349274,0.9410633967673335,0.08042017371346843,14.617953300476074 +UBR5,4,0.025584911927580833,0.1227094754576683,0.9688194703949856,-0.0008634498521615362,10.530195236206055 +UBTF,20,0.007103284355252981,0.06430111080408096,0.9907784602627229,-0.005466066282087735,5.548480033874512 +UFL1,16,0.00888040941208601,0.0712575614452362,0.9881972555804648,0.23205674570886775,6.203845024108887 +UFM1,13,0.011694380082190037,0.08210938423871994,0.984700506444761,0.12633046140748505,7.1192307472229 +UNC45A,18,0.007440216839313507,0.06586813181638718,0.9902537026616819,-0.00880936132993576,5.678547382354736 +UNC50,9,0.013566109351813793,0.0898197665810585,0.9820220928284558,0.07496756270380192,7.667823314666748 +UPF1,13,0.01727350614964962,0.09345169365406036,0.9777703842059953,0.2925264857097182,8.652361869812012 +UPF2,16,0.011854340322315693,0.08190006762742996,0.9843954109074342,0.10329626141516268,7.167755126953125 +UPF3A,7,0.018304921686649323,0.10297586023807526,0.9767759185191581,-0.03372748899241418,8.906936645507812 +UQCC2,10,0.0161129143089056,0.09647276252508163,0.9787176440337924,-0.017860543881452542,8.356637001037598 +UQCRB,9,0.01634877920150757,0.09763950854539871,0.9779373602948488,0.18380724178935423,8.417577743530273 +UQCRC1,31,0.004753298591822386,0.05285213142633438,0.9939501451925583,-0.017536927214363087,4.5388102531433105 +UQCRC2,7,0.019943874329328537,0.10966570675373077,0.9748362704801822,0.07360399897112581,9.297136306762695 +UQCRFS1,12,0.01292388141155243,0.08585602045059204,0.9833339814629098,0.15757005681101396,7.484123229980469 +UQCRH,7,0.020349465310573578,0.10758306831121445,0.9730968143038996,0.12934294366039076,9.391196250915527 +UQCRQ,16,0.008847094140946865,0.07134110480546951,0.9890180507054298,0.14557714597222704,6.192196846008301 +URB1,4,0.0330624133348465,0.14204218983650208,0.9584902570694187,-0.034092717521252966,11.970484733581543 +URB2,20,0.007440105080604553,0.06533011794090271,0.9906221316628575,0.05326200758796781,5.678504943847656 +URI1,8,0.017837654799222946,0.10381298512220383,0.9768053939095206,0.10682210008264062,8.792518615722656 +URM1,13,0.009408497251570225,0.07472007721662521,0.9879289582681673,0.05142856844829301,6.3856425285339355 +UROD,20,0.006909619085490704,0.06326758116483688,0.9916505248078102,-0.12472772641824478,5.472320079803467 +USF2,19,0.007589854300022125,0.06644067913293839,0.9898388665936327,0.05875212028578918,5.735366344451904 +USP10,9,0.013241725973784924,0.08785819262266159,0.9829060107496863,-0.018116302342765598,7.575594902038574 +USP19,7,0.01706649735569954,0.10038940608501434,0.9788936736072226,0.0033167183784861953,8.600359916687012 +USP36,18,0.008399464190006256,0.06979150325059891,0.9889507958704171,-0.08664499062225595,6.033512592315674 +USP37,7,0.021611057221889496,0.11316774040460587,0.9709513015748169,0.05868135491369781,9.677929878234863 +USP39,22,0.013017382472753525,0.0837102085351944,0.9829154568747588,0.24259860300639857,7.5111470222473145 +USP5,20,0.0068737524561584,0.0630076676607132,0.9909098679181841,0.05586236363558653,5.458098888397217 +USP7,27,0.004978914745151997,0.054008323699235916,0.9936889805197613,-0.0029110734976593904,4.645279407501221 +USP8,8,0.016092723235487938,0.09736993163824081,0.9800624119290017,-0.046912311703746476,8.351399421691895 +USP9X,11,0.011781175620853901,0.0842999741435051,0.9843979319385683,0.031187513950971878,7.14560079574585 +USPL1,10,0.012489682994782925,0.08504298329353333,0.9838140787541331,0.18631607513620288,7.35732889175415 +UTP15,15,0.025402963161468506,0.11378637701272964,0.9664590463041911,0.20019990594741752,10.492685317993164 +UTP18,7,0.023361459374427795,0.11630997806787491,0.9688351287773724,0.04670140887179227,10.062233924865723 +UTP20,11,0.019265703856945038,0.10395654290914536,0.9739091914693598,0.13827609545790456,9.137700080871582 +UTP23,17,0.012354033999145031,0.08264660090208054,0.9832063661559943,0.10944806660538449,7.31726598739624 +UTP3,2,0.08490511029958725,0.2299906611442566,0.8994751875902994,0.021360506858924728,19.182771682739258 +UTP6,7,0.033107690513134,0.13634616136550903,0.9545373905286902,0.13304426807658368,11.978678703308105 +UXS1,13,0.012102648615837097,0.0835350751876831,0.9837582233861544,-0.0030113920942789326,7.242435932159424 +UXT,21,0.00953314546495676,0.07381711900234222,0.9869059411475438,0.15885279926382076,6.4278035163879395 +VARS,22,0.012463170103728771,0.08246967196464539,0.9841875001166304,0.4540986023381752,7.349515914916992 +VARS2,12,0.01063119899481535,0.08009042590856552,0.9867204783016036,0.08331131953457409,6.787901878356934 +VBP1,11,0.011473584920167923,0.0823897272348404,0.9849185460026563,0.0020576073416304864,7.051703453063965 +VCP,24,0.005747468676418066,0.057876113802194595,0.9924690661060273,0.06710982349743382,4.990944862365723 +VEZT,21,0.007401581387966871,0.06515365839004517,0.9902812585989819,-0.07030158269744867,5.663784503936768 +VHL,16,0.008257029578089714,0.06934219598770142,0.9896419165003331,-0.13986818676728627,5.982137203216553 +VMP1,10,0.012961243279278278,0.08823489397764206,0.9828909261326721,0.022825243481325915,7.494933605194092 +VPS13D,14,0.010046221315860748,0.07704170793294907,0.986687012205512,0.05666208911968356,6.598508834838867 +VPS18,15,0.008263207972049713,0.07016190141439438,0.9891171822658757,0.22171854981540406,5.984375476837158 +VPS25,9,0.01671457476913929,0.10005355626344681,0.9781388771861392,0.0996654026286253,8.511226654052734 +VPS28,6,0.054901544004678726,0.16805200278759003,0.9255329478146053,0.37299827261233615,15.425410270690918 +VPS29,17,0.010868323966860771,0.07840119302272797,0.9857778918250429,0.04266688541047286,6.863185882568359 +VPS33A,18,0.007146280724555254,0.06596013903617859,0.9910087885345433,0.018992042770301667,5.565248012542725 +VPS35,5,0.03124847263097763,0.1367388218641281,0.9595311923420456,0.18959860509723542,11.637476921081543 +VPS37A,7,0.019262181594967842,0.10743778944015503,0.9763920302427319,-0.09531921233912014,9.136863708496094 +VPS37C,17,0.007373556029051542,0.06597873568534851,0.9902209369520302,0.18414246056629138,5.653051376342773 +VPS41,33,0.0041557783260941505,0.04915015771985054,0.9947153318526101,0.07513335138692707,4.243953704833984 +VPS51,10,0.012275820598006248,0.08560874313116074,0.9843681688653652,0.008272796237014057,7.294066429138184 +VPS54,32,0.004519382957369089,0.05135660618543625,0.9938606153627098,0.169071869218135,4.425720691680908 +VPS72,10,0.015289822593331337,0.09362185746431351,0.981385456501599,-0.014026603118073673,8.140398025512695 +WAC,16,0.009443357586860657,0.074175164103508,0.9874137579360945,-0.03498579439889874,6.3974609375 +WARS,7,0.01834517903625965,0.10495618730783463,0.9766349638368528,0.04338027764499216,8.916726112365723 +WBP1,10,0.01520505454391241,0.09509201347827911,0.980543542648348,-0.01571106185067821,8.117802619934082 +WBP11,31,0.006044091656804085,0.05824965611100197,0.9918295660788791,0.3291102575877407,5.118114471435547 +WBP1L,20,0.006547678727656603,0.06243147328495979,0.9916364164246598,-0.04990426248066694,5.327066421508789 +WDHD1,18,0.008050310425460339,0.06888289004564285,0.9896530595610572,0.012803508436195621,5.906779766082764 +WDR1,10,0.012311778031289577,0.08561747521162033,0.9842814729684419,0.004712565848882922,7.304741382598877 +WDR12,22,0.009768412448465824,0.07252897322177887,0.9876331146289964,0.21956816076823307,6.5066351890563965 +WDR18,11,0.013267974369227886,0.08761776238679886,0.9826658669229887,0.1147244312151447,7.583099842071533 +WDR24,8,0.022756649181246758,0.11505008488893509,0.9683600598429751,0.36389895579025694,9.93112850189209 +WDR25,16,0.00880481582134962,0.07283679395914078,0.9887102664366383,0.09421747216464514,6.177383899688721 +WDR26,22,0.007194094825536013,0.06443335860967636,0.9905112582139974,0.15418281197080452,5.583834648132324 +WDR3,37,0.009903899393975735,0.07026632130146027,0.9865904126506034,0.12265545407692215,6.551602840423584 +WDR33,11,0.01806844398379326,0.10294095426797867,0.974903147979215,0.2754696100192149,8.849217414855957 +WDR36,29,0.011003917083144188,0.07449629157781601,0.9861399825775902,0.13496830353258302,6.90586519241333 +WDR4,34,0.00407235836610198,0.04892512038350105,0.9948763521099129,0.033078589394243095,4.20114278793335 +WDR43,8,0.0228355061262846,0.11388839036226273,0.9692833828541094,0.08678453256479589,9.948320388793945 +WDR46,7,0.02252095937728882,0.11543191224336624,0.9698293507581642,0.0351703139504681,9.879566192626953 +WDR5,8,0.027580877766013145,0.12164447456598282,0.9627159328909202,-0.025612863506626785,10.93323040008545 +WDR54,21,0.006171120330691338,0.06028981879353523,0.9920862390569186,0.01883779439729768,5.171618461608887 +WDR55,9,0.01445443369448185,0.09272872656583786,0.9820648447337802,0.022202709307299166,7.914892196655273 +WDR61,11,0.01668477989733219,0.0976121723651886,0.9788885701275106,0.05689634231166746,8.503636360168457 +WDR7,11,0.013064657337963581,0.08836497366428375,0.9828009088186607,0.07552030382591483,7.524774074554443 +WDR70,3,0.0423579104244709,0.16185235977172852,0.950190357210916,-0.04135106311987099,13.549139976501465 +WDR74,18,0.008271653205156326,0.06965620815753937,0.9891991063760716,0.04553729563761807,5.987432479858398 +WDR75,16,0.014339549466967583,0.0888596847653389,0.9804307302516846,0.11310533337192373,7.88337516784668 +WDR77,4,0.032189805060625076,0.13885895907878876,0.9601216804535023,0.07258580272187433,11.811461448669434 +WDR82,10,0.02856752835214138,0.1262008100748062,0.9606371409095298,0.25579494671128405,11.127069473266602 +WDR83OS,9,0.014814041554927826,0.09378628432750702,0.9804076844285018,-0.016578549902592634,8.01274299621582 +WDR92,6,0.02127138525247574,0.11167006939649582,0.9723326300601784,0.060282504438999705,9.601572036743164 +WDTC1,11,0.019786259159445763,0.10541108250617981,0.9732663041533673,0.13986757949024595,9.260327339172363 +WEE1,4,0.048176221549510956,0.1713148057460785,0.9391997952029463,0.14464801197444985,14.449766159057617 +WNK1,13,0.012080518528819084,0.08395043760538101,0.9834584052371116,0.05787026297873123,7.235811710357666 +WTAP,17,0.008051400072872639,0.06835758686065674,0.9894831921470301,0.034107330536345196,5.90717887878418 +XAB2,16,0.021675489842891693,0.10975368320941925,0.9747085686221643,0.24300299296413913,9.692346572875977 +XPO1,20,0.016815710812807083,0.09214353561401367,0.9771587153565154,0.46688605340164946,8.53693675994873 +XPO5,22,0.007043187972158194,0.063331238925457,0.9905692220827973,0.0017767051531727234,5.524959564208984 +XRCC1,26,0.005608218722045422,0.05717993155121803,0.9925929665975218,0.016709918478335004,4.930113792419434 +XRCC2,15,0.00795446615666151,0.0676121860742569,0.9899111311603774,0.08285339138900454,5.871511936187744 +XRCC3,6,0.023573681712150574,0.11718075722455978,0.9685739264987804,0.15668952102881054,10.107834815979004 +XRCC5,9,0.018038831651210785,0.09964171051979065,0.9763042839730607,0.22996966101969618,8.841961860656738 +XRCC6,3,0.052315372973680496,0.17416086792945862,0.936550401497906,0.27744902490830836,15.057716369628906 +XRN1,1,0.1870291531085968,0.33662334084510803,0.7768359805123422,0.14818892563494074,28.470762252807617 +XRN2,6,0.024412408471107483,0.12011521309614182,0.9682926380462251,0.02568023840621844,10.286076545715332 +YARS,19,0.00752585893496871,0.06603159010410309,0.9899720291098734,0.18861794103636265,5.711135387420654 +YARS2,19,0.007333823479712009,0.06562016904354095,0.990252429363562,0.15654668170338237,5.637800216674805 +YBX1,8,0.018577078357338905,0.10495198518037796,0.9748714194305728,0.10208200069773185,8.972907066345215 +YEATS2,24,0.009449165314435959,0.07014908641576767,0.9873711652915418,-0.08523934700700903,6.399428367614746 +YEATS4,23,0.008697123266756535,0.06951503455638885,0.9885467106408343,0.1179020076941858,6.139489650726318 +YJEFN3,18,0.0074986545369029045,0.0662989541888237,0.9909595332097446,-0.04885845393612864,5.700804233551025 +YKT6,7,0.020297443494200706,0.10842215269804001,0.9729545540014007,0.08586803584777257,9.379185676574707 +YPEL1,7,0.015288359485566616,0.09591826796531677,0.9810439729449353,-0.017825134687740263,8.140009880065918 +YPEL5,7,0.025011137127876282,0.12129590660333633,0.9662927020400095,0.23156050983403204,10.411449432373047 +YRDC,10,0.01585557498037815,0.09666337817907333,0.9791856432461896,-0.08424711689277543,8.289636611938477 +YTHDC1,23,0.007840614765882492,0.06646702438592911,0.9891972964628962,0.2131138106258124,5.829341888427734 +YY1,8,0.01828484795987606,0.10308992862701416,0.9771116260635053,-0.16340425980359188,8.90205192565918 +ZBTB11,14,0.009957795031368732,0.07677105069160461,0.9863489942090704,0.22759571557406008,6.5694050788879395 +ZBTB17,16,0.012577114626765251,0.08450339734554291,0.9830964985923033,0.13111814824436288,7.383035659790039 +ZC3H18,3,0.04393145814538002,0.16039033234119415,0.9460044474847638,-0.04062990846443442,13.79851245880127 +ZC3H8,2,0.07014454901218414,0.2062426656484604,0.9178047430275457,-0.0205566803476788,17.435781478881836 +ZCCHC9,6,0.030888929963111877,0.13536378741264343,0.9588193634249081,0.047741469267450896,11.570333480834961 +ZCRB1,14,0.012537932954728603,0.08570334315299988,0.9847211078097428,0.12233432062946684,7.37152624130249 +ZFC3H1,11,0.021018613129854202,0.09638811647891998,0.9717386333067382,0.10087229537975456,9.544352531433105 +ZFP69B,15,0.010883954353630543,0.07915312051773071,0.9857111716741752,0.09032883794107481,6.868118762969971 +ZMAT2,43,0.007355641573667526,0.06267890334129333,0.9907535095154826,0.22531505356335726,5.646180152893066 +ZMAT5,10,0.015355784446001053,0.09572921693325043,0.9795135090120489,0.0215855602289741,8.157938957214355 +ZNF100,9,0.014100718311965466,0.09177818894386292,0.9821284929363491,-0.03084214960269022,7.817449569702148 +ZNF131,3,0.05051064118742943,0.1746661216020584,0.9343313445703761,0.18697907786992055,14.7957124710083 +ZNF133,10,0.014171119779348373,0.09096353501081467,0.9818474923941131,-0.060600210904330376,7.836940765380859 +ZNF207,29,0.0053264545276761055,0.05541844666004181,0.9929282445111719,-0.0048050969066642255,4.8046698570251465 +ZNF236,9,0.020129188895225525,0.10573771595954895,0.9725579912705276,0.10218662843567193,9.340230941772461 +ZNF24,6,0.018272385001182556,0.10446935892105103,0.9770744100302091,-0.04666281492939508,8.899018287658691 +ZNF253,12,0.01091052871197462,0.07954785972833633,0.9862964474400113,-0.0975936794017408,6.876498699188232 +ZNF284,22,0.005718690808862448,0.05808350443840027,0.9924868185169284,0.05102665917722626,4.978434085845947 +ZNF317,13,0.013022570870816708,0.08574147522449493,0.982630072935363,0.011095399587941647,7.512644290924072 +ZNF335,14,0.009644405916333199,0.07503920793533325,0.9875681790519475,-0.008694021724630211,6.465203285217285 +ZNF407,11,0.012943503446877003,0.0875292643904686,0.9835514325964293,-0.046898032441667015,7.489802360534668 +ZNF468,20,0.007484672125428915,0.06590942293405533,0.9904151891013557,-0.09301726510481055,5.695487022399902 +ZNF492,15,0.009567959234118462,0.07419706135988235,0.9874912036052621,-0.06053321266946598,6.4395294189453125 +ZNF506,17,0.009288049302995205,0.07328097522258759,0.9880311877151977,-0.08235597174977143,6.3446364402771 +ZNF559,11,0.011632799170911312,0.082163006067276,0.9858289245662536,-0.11832228146072553,7.100461483001709 +ZNF574,32,0.004372669383883476,0.050665128976106644,0.9943039441667333,0.060930218084395055,4.353291988372803 +ZNF658,11,0.012376382946968079,0.08529982715845108,0.9841373657770396,-0.09341283865762462,7.32388162612915 +ZNF674,12,0.010939852334558964,0.08022402226924896,0.9857526141645024,-0.11911826472344231,6.885733127593994 +ZNF676,9,0.012469204142689705,0.08551384508609772,0.9843002738357786,-0.017255276572181708,7.35129451751709 +ZNF687,21,0.007894613780081272,0.06604570150375366,0.9898193571392426,-0.07015671687134008,5.849380970001221 +ZNF718,33,0.004260999150574207,0.050094787031412125,0.9944197115646405,0.013169380133358001,4.29734468460083 +ZNF720,8,0.01654002070426941,0.09950552880764008,0.9788343355298961,-0.005717040458715238,8.466666221618652 +ZNF763,15,0.008776444010436535,0.0718751922249794,0.9888199610711941,-0.029212907154299816,6.167422771453857 +ZNF830,1,0.08690223097801208,0.23494845628738403,0.9011282209933712,0.10087747147560729,19.407068252563477 +ZNHIT2,9,0.013833815231919289,0.09110983461141586,0.9821941192231659,-0.03956703017574956,7.743110179901123 +ZNHIT3,7,0.020872216671705246,0.1111532598733902,0.9731433837607372,-0.07156673191927028,9.511055946350098 +ZNHIT6,11,0.015292126685380936,0.0938938707113266,0.979452447005631,-0.049490325369562514,8.141012191772461 +ZNRD1,8,0.02127443440258503,0.11279561370611191,0.9712908075088498,0.12047627883817053,9.60226058959961 +ZRANB2,4,0.034264933317899704,0.14232084155082703,0.9581014812939754,-0.1082545026223163,12.186230659484863 +ZRSR2,12,0.011263447813689709,0.08198262006044388,0.9858263693757766,0.016546780111162302,6.9868292808532715 +ZW10,20,0.0072879246436059475,0.06519079953432083,0.9904926931090465,0.15064331934021252,5.6201300621032715 +ZWINT,4,0.0389794260263443,0.14827421307563782,0.9487072298166827,0.06484896249198781,12.99756908416748 diff --git a/baseline/outputs/conditional_flow_matching_method/pred_conditional_flow_matching_method_seen_cell_split_test.h5ad b/baseline/outputs/conditional_flow_matching_method/pred_conditional_flow_matching_method_seen_cell_split_test.h5ad index fa4b5d4ae7e8aed35721e07f2e802c6a7f98d164..8930bab3feeb6e5e68c1b08cbcf738a216f69cae 100644 --- a/baseline/outputs/conditional_flow_matching_method/pred_conditional_flow_matching_method_seen_cell_split_test.h5ad +++ b/baseline/outputs/conditional_flow_matching_method/pred_conditional_flow_matching_method_seen_cell_split_test.h5ad @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:685db3257cd231731f33e451c6f75152f8d61a087801a523b25329f7ad8e8a4c -size 319587490 +oid sha256:2154d6473c5087e27e20bfe35788d095343a78b73f5d3c5299f2724989e6829e +size 312789702 diff --git a/baseline/outputs/conditional_flow_matching_method/real_conditional_flow_matching_method_seen_cell_split_test.h5ad b/baseline/outputs/conditional_flow_matching_method/real_conditional_flow_matching_method_seen_cell_split_test.h5ad index 39ec73d74e9a0140e7881547054c96e11e592f40..7ea3805570fd9f47031e96b30e763ae9f29f5fa8 100644 --- a/baseline/outputs/conditional_flow_matching_method/real_conditional_flow_matching_method_seen_cell_split_test.h5ad +++ b/baseline/outputs/conditional_flow_matching_method/real_conditional_flow_matching_method_seen_cell_split_test.h5ad @@ -1,3 +1,3 @@ version https://git-lfs.github.com/spec/v1 -oid sha256:50ae5ae19dd79cc40a9f26285f3cd2f2a6f706375a40be1f0d0034ece14d8b76 -size 136058500 +oid sha256:8076e2f5b333f9f46120320ce77965062bd1f0268719e4169f5121c95da5c50c +size 136059684 diff --git a/baseline/outputs/conditional_flow_matching_method/summary_conditional_flow_matching_method_seen_cell_split_test.csv b/baseline/outputs/conditional_flow_matching_method/summary_conditional_flow_matching_method_seen_cell_split_test.csv index 8ff98e7a8785b8325ce6da3f53ec27e922e4e80c..d99d0289cdf263a6018e11db47dcbc61392b8be2 100644 --- a/baseline/outputs/conditional_flow_matching_method/summary_conditional_flow_matching_method_seen_cell_split_test.csv +++ b/baseline/outputs/conditional_flow_matching_method/summary_conditional_flow_matching_method_seen_cell_split_test.csv @@ -9,8 +9,8 @@ n_eval_cells,29965 n_train_genes,2056 n_eval_genes,2056 n_eval_genes_missing_from_train,0 -mse_mean_avg,0.019163491268137628 -mae_mean_avg,0.09751620552328628 -pearson_mean_avg,0.9756440617310962 -pearson_delta_avg,0.08957929501688626 -delta_l2_avg,8.494131993922743 +mse_mean_avg,0.01911219263722004 +mae_mean_avg,0.09772463235369945 +pearson_mean_avg,0.9756348678854861 +pearson_delta_avg,0.11158561568526223 +delta_l2_avg,8.491365785728632 diff --git a/baseline/outputs/conditional_flow_matching_method/training_history_conditional_flow_matching_method_seen_cell_split_test.csv b/baseline/outputs/conditional_flow_matching_method/training_history_conditional_flow_matching_method_seen_cell_split_test.csv index c23fd78b7326c0dac8457fe319b568c8a1a854b1..712b6fae0ec46abe65be929f155bee88f270f111 100644 --- a/baseline/outputs/conditional_flow_matching_method/training_history_conditional_flow_matching_method_seen_cell_split_test.csv +++ b/baseline/outputs/conditional_flow_matching_method/training_history_conditional_flow_matching_method_seen_cell_split_test.csv @@ -1,20 +1,21 @@ -epoch,train_total,train_recon,train_flow,train_endpoint,train_mmd,train_mean,val_total,val_recon,val_flow,val_endpoint,val_mmd,val_mean -1,0.7099311868846416,0.2872591935098171,0.23419513962231575,0.07880723661859519,0.19041752696037292,0.012933913624146953,0.33599394125243026,0.2699047566081087,0.015705714855963986,0.0056495392830887186,0.23188400765260062,0.016517441457835957 -2,0.31066063418984413,0.2765745238959789,0.005475435626576655,0.002225396469584666,0.18214683830738068,0.006545574656338431,0.307092376674215,0.26801348601778346,0.0006996398036183867,0.0006301841434227148,0.22149600088596344,0.015564734911701331 -3,0.2988686554133892,0.2729715509712696,0.0009557047189446167,0.000734101812413428,0.18187685549259186,0.005908813348505646,0.3069025973478953,0.2684025314326088,0.0002666692028772862,0.0004973066479578847,0.22006762524445853,0.015844644212241594 -4,0.2980132968723774,0.27295295111835005,0.0005763415638648439,0.0006081948781502433,0.18103148102760314,0.005788588010473177,0.3046994327257077,0.26719280735900003,0.00017590964641082488,0.000477578246621609,0.2179656277100245,0.015207408117324425 -5,0.30012667298316953,0.2750719152390957,0.0003905085723090451,0.000544906579743838,0.1818532133102417,0.006011220881482586,0.30512714323898155,0.26758854153255623,9.953832621552768e-05,0.00045426935381935135,0.21861535807450613,0.015300623073320216 -6,0.30163356229662897,0.2762745556235313,0.0002674222952919081,0.0005013478714681696,0.1827651083469391,0.0064306868525454775,0.30395802296698093,0.2669009516636531,6.245124359338661e-05,0.0004191172956780065,0.21730807423591614,0.015023034056260562 -7,0.29759583368897435,0.27278817735612393,0.00021779624177725054,0.00048540595395024866,0.18329548478126526,0.005908708658535033,0.30457736427585286,0.26726406036565703,5.985522019121466e-05,0.0004444642369586897,0.21792652209599814,0.015208637744459944 -8,0.29631013214588164,0.27197444200515747,0.00016812849309644662,0.00046844744429108687,0.18179718792438507,0.00566955502901692,0.3056660642226537,0.26808740260700387,3.9517619522181725e-05,0.0004276314660576948,0.21765197813510895,0.01554036965050424 -9,0.2976865042746067,0.27330474071204663,0.0001400934579578461,0.00045558969519333915,0.18226113498210908,0.005717715128557756,0.3036455375452836,0.2667286802704136,5.030420936691371e-05,0.00042721984997721546,0.21662546942631403,0.01496523811268465 -10,0.30110554188489913,0.2761254480481148,0.00010798695308039896,0.0004481538051913958,0.1827368062734604,0.006320357949007303,0.3054047754655282,0.2679311639318864,3.1379622858670096e-05,0.00043262542749289423,0.21773227800925574,0.015437006145172441 -11,0.2978330807387829,0.27362941272556784,9.492984409007477e-05,0.0004426611168310046,0.18028389990329743,0.0058115554082905875,0.3036332322905461,0.2667977775757511,2.2094751310911914e-05,0.0004274136511715672,0.21580631534258524,0.015007970616958724 -12,0.2998507963120937,0.275599198192358,7.656231164219208e-05,0.00043531879491638393,0.18083509385585786,0.005835586863686331,0.3037345012029012,0.2669174854333202,1.7477341752207092e-05,0.0004054381715832278,0.21595444033543268,0.014992633077781647 -13,0.29901653453707694,0.27481679126620295,6.598191204830072e-05,0.00043480782478582113,0.1807200115919113,0.005811366098350845,0.30384495419760543,0.26692460166911286,1.6282691698658407e-05,0.0004257825967215467,0.21637233346700668,0.015045803016012846 -14,0.2968123908340931,0.2731849317997694,4.770272604218917e-05,0.0004260506186255952,0.18066644966602324,0.0052762343967333435,0.30315500870347023,0.266526793440183,6.479716044320109e-06,0.0004105396998663006,0.21572753290335336,0.014840473726508208 -15,0.2943950039148331,0.27130497187376024,4.11799237372179e-05,0.0004259626036218833,0.17903094708919526,0.004912184645072557,0.30446299848457176,0.2672804457445939,4.855887643392028e-06,0.0004148717841114073,0.21794762214024863,0.015173066218267195 -16,0.2980514550209045,0.2745214889198542,3.7895015248068374e-05,0.0004240531886171084,0.17962264001369477,0.005298834676505066,0.30319284585615,0.26659159113963443,4.310457801655805e-06,0.00039280177528174437,0.2154802903532982,0.014850359395495616 -17,0.2954455578327179,0.2721803789585829,3.5111359302391066e-05,0.0004220420584897511,0.17855823934078216,0.005145665962481871,0.30359477922320366,0.2668211593603094,4.321080851354964e-06,0.00042292078493725665,0.21621116250753403,0.014934559013151253 -18,0.29714504763484,0.2734395454078913,3.234614638131461e-05,0.0004210493169375695,0.1795671707391739,0.00548974140023347,0.3034466914832592,0.2667315676808357,4.556634356352636e-06,0.0003984413900373814,0.2162515843907992,0.01488391135353595 -19,0.2945520044863224,0.27154500477015975,2.9618433245559572e-05,0.00042023393063573167,0.17822119176387788,0.004930334103410133,0.303397794564565,0.2667580641185244,3.0363262434699814e-06,0.0004082458414510863,0.21533614148696265,0.014897432944659764 +epoch,train_total,train_recon,train_flow,train_endpoint,train_mmd,train_mean,train_ot_decode_mmd,train_ot_decode_mean,train_disp_reg,train_vel_l2,val_total,val_recon,val_flow,val_endpoint,val_mmd,val_mean,val_ot_decode_mmd,val_ot_decode_mean,val_disp_reg,val_vel_l2 +1,0.7260855907201766,0.2875967035442591,0.23561803043819965,0.07883021616144106,0.19043146014213563,0.01306097540888004,0.1907109647989273,0.012873750185826793,0.07886278878664597,0.23192312025930734,0.35723286929229897,0.27002522846062976,0.01928534151132529,0.006376065730970974,0.23260554671287537,0.016577831469476223,0.22088205814361572,0.016581362287979573,0.006376123317750171,0.0192852493103904 +2,0.32195078864693644,0.27683196097612384,0.005564161220099777,0.002244915282644797,0.18226666152477264,0.006630028394283727,0.18226015627384184,0.006631795310531743,0.002247825816157274,0.005247413141478319,0.32235846916834515,0.2681408676629265,0.0006243309762794524,0.000624367375470077,0.22279327611128488,0.015617763516881192,0.22322356700897217,0.015617636876413599,0.0006243859036961416,0.0006243336993065896 +3,0.30961272642016413,0.2731092883646488,0.0009283065810450353,0.0007219279487617314,0.1819450443983078,0.005959683986147866,0.1819866967201233,0.005957014043233357,0.000723688489815686,0.0007834495035058353,0.3218673101315896,0.26851155888289213,0.0002466099206988777,0.0004995577855879674,0.22180494169394174,0.01589724389486946,0.2135800967613856,0.0158972875700177,0.0004995410642247103,0.0002465787601977354 +4,0.308655117303133,0.2730628021806478,0.0005543926644895692,0.0006006049244024326,0.18111132979393005,0.005831923610530794,0.1810938000679016,0.005826686712098308,0.0006003224308369681,0.0004841640405356884,0.3196913208812475,0.26723536010831594,0.00010943980608620525,0.0004519751922392364,0.21998917808135351,0.015223852848672928,0.2206505537033081,0.015223954212463772,0.00045198322247112327,0.00010943980866310692 +5,0.3108049909770489,0.27514853194355965,0.0003628557214688044,0.0005342075794760603,0.18192058503627778,0.006041289322893135,0.18191982984542845,0.006041554365074262,0.0005344023353245575,0.00032915555333602243,0.31955820694565773,0.2676338916644454,0.00010771373968054831,0.00045780570750745636,0.21824985990921655,0.015323652711231261,0.21099440505107245,0.01532362414097103,0.00045780078592846013,0.00010770534898559465 +6,0.31252818182110786,0.2763585484772921,0.0002805642613384407,0.0005050578701775521,0.18282712519168853,0.006453430550172925,0.18282259464263917,0.0064557314780540765,0.0005050609388854354,0.0002637496498209657,0.31856723440190154,0.26699944275120896,7.950710528348282e-05,0.00044621725707353715,0.217384472489357,0.015066945668271122,0.21297617504994074,0.015067030331313921,0.0004462219670434327,7.950325796931186e-05 +7,0.30833912193775176,0.272866594940424,0.00021079509875562507,0.0004830249144288246,0.18334017038345338,0.005927924775751308,0.18333252429962157,0.005925687187118456,0.0004831311563611962,0.00020053234096849336,0.319122218216459,0.2673648300891121,5.68908002757477e-05,0.0004273955206978523,0.2185997242728869,0.015262578444283767,0.2103173608581225,0.015262630540140284,0.0004273941922292579,5.688975935906152e-05 +8,0.30688960731029513,0.2720266804099083,0.00015900595411949325,0.0004649149902979843,0.18184158623218535,0.005689510627998971,0.1818227016925812,0.005690020026522688,0.00046492919951560905,0.00015248536263243295,0.320815106232961,0.2681772510210673,3.582879912755743e-05,0.0004208194586681202,0.21868030975262323,0.015579573926515877,0.22053943326075873,0.015579682202466453,0.0004208174726348564,3.582788349376642e-05 +9,0.30831721290946007,0.2733631574362516,0.00013217186733527342,0.00046024185663554815,0.18232015073299407,0.005737415289622731,0.18232950389385225,0.005740664183394984,0.00046041724635870196,0.0001272475088626379,0.31832230649888515,0.26674903618792695,3.3840355058600835e-05,0.00041462214782465406,0.21733773996432623,0.014970206262660213,0.21729062994321188,0.014970165657966087,0.00041462436941704556,3.3841780653650254e-05 +10,0.31195622861385347,0.2761949112266302,0.00011890640944329789,0.0004491935684927739,0.1828007584810257,0.006346461405046284,0.1828000032901764,0.006349315259139985,0.00044928388655534945,0.0001151123175441171,0.3203591462224722,0.268011637032032,3.206731973174707e-05,0.00043089837193595787,0.21804419656594595,0.01547027142563214,0.2187472606698672,0.01547041125013493,0.00043089924232238747,3.206799067356769e-05 +11,0.3083870202302933,0.27368386574089526,9.396014942467446e-05,0.00044237337744561955,0.18034431338310242,0.005828264245064929,0.18034050166606902,0.005827533179544844,0.00044227787089766934,9.09016510559013e-05,0.31858168604473275,0.2668596425404151,2.418360763082698e-05,0.0004083190024175565,0.21655274430910745,0.015040880389278755,0.22042307257652283,0.015040872705867514,0.00040832356110816664,2.4184080833341188e-05 +12,0.3104447920620441,0.27566283866763114,7.652406678971601e-05,0.0004379084061656613,0.18084345042705535,0.005851832777261734,0.18088712751865388,0.005853201521676965,0.0004379050472925883,7.409204519717605e-05,0.3183728369573752,0.2669541717817386,1.4015850884637379e-05,0.0004215567575253469,0.21626121799151102,0.015008687764445009,0.21591316163539886,0.015008722960677309,0.00042155334692021523,1.4017012783066699e-05 +13,0.3095723737776279,0.2748602519184351,6.444809620006708e-05,0.0004314646300917957,0.18077716290950774,0.005822989161824808,0.18077527821063996,0.0058237342309439555,0.00043147400225279853,6.241272891202243e-05,0.3190456007917722,0.26695811407019693,1.3425717706165111e-05,0.00040766025449556764,0.21660149097442627,0.015053599141538143,0.22764570017655691,0.015053640599944629,0.0004076607971607397,1.3424829717223474e-05 +14,0.3071902035176754,0.27320717737078665,4.848282325838227e-05,0.0004266230497887591,0.18064003705978393,0.0052754978369921445,0.18068985044956207,0.0052749221987323834,0.00042656717108911833,4.6783715497440425e-05,0.31753033958375454,0.2665523908411463,6.012266103046689e-06,0.000408976659552233,0.2155005211631457,0.014851601760407599,0.21291605879863104,0.014851588678235808,0.0004089788105072027,6.012255151214656e-06 +15,0.30460422158241274,0.27132960088551045,4.2999617853638483e-05,0.00042387160923681224,0.17900700390338897,0.004914346578298137,0.1790170431137085,0.004912187009467744,0.00042383127074572256,4.148846801399486e-05,0.3192891739308834,0.26737873163074255,5.344201089959218e-06,0.0004150052215360726,0.2174759308497111,0.015218772716859045,0.21839422235886255,0.0152188020098644,0.00041500089052230277,5.344023226901602e-06 +16,0.30840020433068277,0.274547743499279,3.9564086837344806e-05,0.0004239401369704865,0.17964680790901183,0.005303582356427796,0.17964699923992156,0.005305172667722218,0.0004239616051199846,3.817286093180883e-05,0.31774958968162537,0.2665997852260868,5.252525464053785e-06,0.0004090565920099228,0.21603586028019586,0.014851153187919408,0.21531562507152557,0.014851159081445076,0.00040905796716591186,5.252250465067239e-06 +17,0.3056986400485039,0.27221252381801603,3.661385340819834e-05,0.0004228895182313863,0.17855076670646666,0.005146461731055751,0.17854874074459076,0.005146189735387452,0.00042287239906727334,3.5325822163940754e-05,0.31822768909235793,0.26685668155550957,4.755999488755454e-06,0.0004082766251182572,0.21623564014832178,0.014948594088006454,0.21692686279614767,0.014948586574367559,0.0004082792759921479,4.755679488728977e-06 +18,0.3075269614160061,0.27346095390617847,3.407484312447195e-05,0.0004194395513331983,0.17958719730377198,0.005490880048018881,0.17958061635494232,0.0054894074978074055,0.00041947011894080785,3.2901750673772765e-05,0.3173826492081086,0.2667658564945062,4.960979955133856e-06,0.0004111661443554719,0.21571681400140127,0.01489716376333187,0.20413036892811456,0.01489715842277898,0.00041116879765468184,4.960996439725325e-06 +19,0.30507050022482873,0.2717678413540125,3.1630149887860174e-05,0.0004208134714281186,0.17840054631233215,0.005024521636078134,0.17840739011764525,0.005022601757082157,0.0004208155744709074,3.0520781301675015e-05,0.31796940229833126,0.2670774096623063,4.5087711792272484e-06,0.0004074913540534908,0.2162719095746676,0.01506858216210579,0.20428979893525442,0.01506858554300076,0.00040749192855097743,4.508942870036965e-06 +20,0.3028733278810978,0.2694020713865757,3.0199566554074407e-05,0.00042116680851904674,0.17940658390522002,0.005040690860478207,0.1794028949737549,0.0050386664643883704,0.00042120331825572064,2.916663064752356e-05,0.31788071307043236,0.2667013940711816,6.956826145672797e-06,0.0004122269850389178,0.21595931549866995,0.014903112887016809,0.21467652916908264,0.014903124397581754,0.00041222217078029644,6.956513061595615e-06 diff --git a/method/flow_matching.ipynb b/method/flow_matching.ipynb index 2ecd3005d2c86cb4a100ae4bae916628f0f34836..9384078b6ebd0656f42f1b62b1d7d38a31eb4292 100644 --- a/method/flow_matching.ipynb +++ b/method/flow_matching.ipynb @@ -1,2589 +1,2861 @@ { - "cells": [ + "cells": [ + { + "cell_type": "markdown", + "id": "da3151cf", + "metadata": {}, + "source": [ + "# Conditional Flow Matching Method\n", + "\n", + "This notebook implements a thesis-specific method for single-gene perturbation prediction on single-cell RNA-seq data.\n", + "\n", + "Core idea:\n", + "\n", + "1. Encode control and perturbed cells into a latent space.\n", + "2. Use perturbation gene identity plus batch background as the condition signal.\n", + "3. Build pseudo pairs with mini-batch optimal transport.\n", + "4. Train a FiLM-conditioned flow matching model that transports control-cell states toward the perturbation distribution.\n", + "5. Decode generated latent states back to expression space and evaluate with distribution-aware metrics.\n", + "6. **CellOT-style coupling**: besides matching the flow endpoint to the Sinkhorn latent barycenter, the loss also aligns **decode(z1_bar)** with the perturbed batch (MMD + pseudo-bulk) and applies a light **displacement prior** on `pred_z1 − z0` (optional **velocity L2**). Mini-batches resample controls if a spurious index collision occurs.\n", + "\n", + "By default, the notebook uses a `seen_cell_split`, which matches the algorithm design based on learned perturbation-gene embeddings. If you switch to the original unseen-gene split, unseen perturbation genes are mapped to an `__unk__` token and should be interpreted as an ablation rather than the main setting.\n" + ] + }, + { + "cell_type": "markdown", + "id": "9e3b1b7d", + "metadata": {}, + "source": [ + "## Method Summary\n", + "\n", + "**条件（与代码一致）：「基因 + batch」** — `ConditionEncoder` 仅包含扰动基因嵌入与技术 batch 嵌入（经 `MLP` 得到条件向量 `h`）。不包含单独的扰动类型（如 KO/activation）或细胞类型嵌入。\n", + "\n", + "Let `x` be the preprocessed gene-expression vector of a control cell, and let `p` be a single-gene perturbation. The model learns:\n", + "\n", + "`x -> z = E(x) -> z_hat(1) via conditional flow -> x_hat = D(z_hat(1))`\n", + "\n", + "The vector field is conditioned on **gene + batch** (`h`), plus **continuous time embedding** `phi(t)` (sinusoidal features + small MLP).\n", + "\n", + "### Training objective（总损失，与 `compute_loss_terms` 一致）\n", + "\n", + "```\n", + "L = RECON_WEIGHT * L_recon\n", + " + FLOW_WEIGHT * L_flow\n", + " + ENDPOINT_WEIGHT * L_endpoint\n", + " + MMD_WEIGHT * L_mmd\n", + " + MEAN_WEIGHT * L_mean\n", + " + OT_DECODE_MMD_WEIGHT * L_ot_decode_mmd\n", + " + OT_DECODE_MEAN_WEIGHT * L_ot_decode_mean\n", + " + DISPLACEMENT_REG_WEIGHT * L_disp_reg\n", + " + VELOCITY_L2_WEIGHT * L_vel_l2\n", + " + DELTA_MSE_WEIGHT * L_delta_mse\n", + " + DELTA_COS_WEIGHT * L_delta_cos\n", + " + DELTA_MMD_WEIGHT * L_delta_mmd\n", + " + DELTA_BULK_WEIGHT * L_delta_bulk\n", + "```\n", + "\n", + "- **L_recon（重构损失）**: latent autoencoder 对 **source 与 target** 表达的重构 MSE，即 `MSE(D(E(x)), x)` 在 control 与 perturbed 样本上求和（见 `source_recon` / `target_recon`）。默认权重下调，避免重构主导、挤压对扰动效应的学习。\n", + "- **L_flow（流匹配损失）**: 预测速度场与 OT 重心目标速度 `z1_bar - z0` 的 MSE。\n", + "- **L_endpoint（潜空间 endpoint 损失）**: 由 `z_t` 与预测速度一步推到 `t=1` 的潜向量 `pred_z1`，与 Sinkhorn 重心 `z1_bar` 的 MSE。\n", + "- **L_mmd**: 解码后表达空间上，预测 batch 与真实 perturbed 细胞的 RBF-MMD（与扰动后分布对齐）。\n", + "- **L_mean**: 解码后按细胞维平均的 pseudo-bulk 向量与真实 perturbed 均值的 MSE。\n", + "- **L_ot_decode_mmd / L_ot_decode_mean**: 将 **Sinkhorn 潜空间重心** `z1_bar` 经解码器映射到表达空间，与当前 batch 的真实 perturbed 细胞做 **MMD** 与 **pseudo-bulk MSE**；约束「OT 传输几何」经解码器落到观测空间（与仅监督 `pred_z1` 路径互补）。\n", + "- **L_disp_reg**: `mean(||pred_z1 - z0||^2)`，抑制过大的 latent 位移（CellOT-style 运输幅度先验）。\n", + "- **L_vel_l2**: `mean(||v||^2)`，可选，惩罚过强的速度场。\n", + "- **L_delta_mse / L_delta_cos**: 以配对 control 为基线，`pred_x1 - source_x` 与 `target_x - source_x` 的逐细胞 MSE 与 `1 - cos`（强化扰动方向与幅度，而非仅拟合绝对表达）。\n", + "- **L_delta_mmd**: 预测 delta 与真实 delta 在基因维上的 RBF-MMD（与扰动一致的分布形状）。\n", + "- **L_delta_bulk**: 基因维平均 delta（pseudo-bulk 扰动向量）的 MSE，与评估中的 delta 指标更直接相关。\n" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "id": "1c11f06f", + "metadata": {}, + "outputs": [ { - "cell_type": "markdown", - "id": "da3151cf", - "metadata": {}, - "source": [ - "# Conditional Flow Matching Method\n", - "\n", - "This notebook implements a thesis-specific method for single-gene perturbation prediction on single-cell RNA-seq data.\n", - "\n", - "Core idea:\n", - "\n", - "1. Encode control and perturbed cells into a latent space.\n", - "2. Use perturbation gene identity plus batch background as the condition signal.\n", - "3. Build pseudo pairs with mini-batch optimal transport.\n", - "4. Train a FiLM-conditioned flow matching model that transports control-cell states toward the perturbation distribution.\n", - "5. Decode generated latent states back to expression space and evaluate with distribution-aware metrics.\n", - "\n", - "By default, the notebook uses a `seen_cell_split`, which matches the algorithm design based on learned perturbation-gene embeddings. If you switch to the original unseen-gene split, unseen perturbation genes are mapped to an `__unk__` token and should be interpreted as an ablation rather than the main setting.\n" - ] + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.\u001b[0m\u001b[33m\n", + "\u001b[0m" + ] }, { - "cell_type": "markdown", - "id": "9e3b1b7d", - "metadata": {}, - "source": [ - "## Method Summary\n", - "\n", - "**条件（与代码一致）：「基因 + batch」** — `ConditionEncoder` 仅包含扰动基因嵌入与技术 batch 嵌入（经 `MLP` 得到条件向量 `h`）。不包含单独的扰动类型（如 KO/activation）或细胞类型嵌入。\n", - "\n", - "Let `x` be the preprocessed gene-expression vector of a control cell, and let `p` be a single-gene perturbation. The model learns:\n", - "\n", - "`x -> z = E(x) -> z_hat(1) via conditional flow -> x_hat = D(z_hat(1))`\n", - "\n", - "The vector field is conditioned on **gene + batch** (`h`), plus **continuous time embedding** `phi(t)` (sinusoidal features + small MLP).\n", - "\n", - "### Training objective（总损失，与 `compute_loss_terms` 一致）\n", - "\n", - "```\n", - "L = RECON_WEIGHT * L_recon\n", - " + FLOW_WEIGHT * L_flow\n", - " + ENDPOINT_WEIGHT * L_endpoint\n", - " + MMD_WEIGHT * L_mmd\n", - " + MEAN_WEIGHT * L_mean\n", - " + DELTA_MSE_WEIGHT * L_delta_mse\n", - " + DELTA_COS_WEIGHT * L_delta_cos\n", - " + DELTA_MMD_WEIGHT * L_delta_mmd\n", - " + DELTA_BULK_WEIGHT * L_delta_bulk\n", - "```\n", - "\n", - "- **L_recon（重构损失）**: latent autoencoder 对 **source 与 target** 表达的重构 MSE，即 `MSE(D(E(x)), x)` 在 control 与 perturbed 样本上求和（见 `source_recon` / `target_recon`）。默认权重下调，避免重构主导、挤压对扰动效应的学习。\n", - "- **L_flow（流匹配损失）**: 预测速度场与 OT 重心目标速度 `z1_bar - z0` 的 MSE。\n", - "- **L_endpoint（潜空间 endpoint 损失）**: 由 `z_t` 与预测速度一步推到 `t=1` 的潜向量 `pred_z1`，与 Sinkhorn 重心 `z1_bar` 的 MSE。\n", - "- **L_mmd**: 解码后表达空间上，预测 batch 与真实 perturbed 细胞的 RBF-MMD（与扰动后分布对齐）。\n", - "- **L_mean**: 解码后按细胞维平均的 pseudo-bulk 向量与真实 perturbed 均值的 MSE。\n", - "- **L_delta_mse / L_delta_cos**: 以配对 control 为基线，`pred_x1 - source_x` 与 `target_x - source_x` 的逐细胞 MSE 与 `1 - cos`（强化扰动方向与幅度，而非仅拟合绝对表达）。\n", - "- **L_delta_mmd**: 预测 delta 与真实 delta 在基因维上的 RBF-MMD（与扰动一致的分布形状）。\n", - "- **L_delta_bulk**: 基因维平均 delta（pseudo-bulk 扰动向量）的 MSE，与评估中的 delta 指标更直接相关。\n" - ] - }, + "name": "stdout", + "output_type": "stream", + "text": [ + "Project root: /root/final\n", + "Input h5ad: /root/final/data/processed/scPerturb/prediction_ready/replogle_k562_essential_single_gene_prediction_ready_full.h5ad\n", + "Output dir: /root/final/baseline/outputs/conditional_flow_matching_method\n", + "Device: cuda\n" + ] + } + ], + "source": [ + "! pip install -q anndata numpy pandas scipy torch matplotlib\n", + "from pathlib import Path\n", + "from copy import deepcopy\n", + "import json\n", + "import math\n", + "import random\n", + "\n", + "import anndata as ad\n", + "import matplotlib as mpl\n", + "import matplotlib.pyplot as plt\n", + "import numpy as np\n", + "import pandas as pd\n", + "import scipy.sparse as sp\n", + "import torch\n", + "import torch.nn as nn\n", + "import torch.nn.functional as F\n", + "\n", + "\n", + "def find_project_root(start: Path | None = None) -> Path:\n", + " start = (start or Path.cwd()).resolve()\n", + " for path in [start, *start.parents]:\n", + " if (path / \"src\" / \"data\").exists() and (path / \"baseline\").exists():\n", + " return path\n", + " return start\n", + "\n", + "\n", + "PROJECT_ROOT = find_project_root()\n", + "DATASET_NAME = \"ReplogleWeissman2022_K562_essential\"\n", + "METHOD_NAME = \"conditional_flow_matching_method\"\n", + "RANDOM_SEED = 42\n", + "\n", + "SPLIT_STRATEGY = \"seen_cell_split\" # choices: \"seen_cell_split\", \"use_existing\"\n", + "TRAIN_FRAC = 0.80\n", + "VAL_FRAC = 0.10\n", + "EVAL_SPLIT = \"test\"\n", + "CONTROL_SPLIT = \"train\"\n", + "TRAIN_SPLIT = \"train\"\n", + "VAL_SPLIT = \"val\"\n", + "\n", + "USE_FEATURE_GENES = True\n", + "FEATURE_GENE_COLUMNS = [\"is_feature_gene\", \"highly_variable\"]\n", + "\n", + "# Optional runtime caps. Set to None for the full run.\n", + "MAX_TRAIN_CONDITIONS = None\n", + "MAX_EVAL_GENES = None\n", + "MAX_EVAL_CELLS_PER_GENE = None\n", + "\n", + "LATENT_DIM = 128\n", + "HIDDEN_DIM = 512\n", + "CONDITION_DIM = 256\n", + "TIME_DIM = 128\n", + "DROPOUT = 0.08\n", + "NUM_FLOW_BLOCKS = 6\n", + "\n", + "BATCH_SIZE = 128\n", + "EPOCHS = 20\n", + "STEPS_PER_EPOCH = 200\n", + "VALIDATION_STEPS = 48\n", + "LEARNING_RATE = 1e-3\n", + "WEIGHT_DECAY = 1e-5\n", + "GRAD_CLIP_NORM = 1.0\n", + "\n", + "RECON_WEIGHT = 1.0\n", + "FLOW_WEIGHT = 1.5\n", + "ENDPOINT_WEIGHT = 0.5\n", + "MMD_WEIGHT = 0.1\n", + "MEAN_WEIGHT = 1.0\n", + "DELTA_MSE_WEIGHT = 0.0\n", + "DELTA_COS_WEIGHT = 0.0\n", + "DELTA_MMD_WEIGHT = 0.0\n", + "DELTA_BULK_WEIGHT = 0.0\n", + "\n", + "# --- CellOT-style OT coupling (same mini-batch Sinkhorn z1_bar; extra decode + transport priors) ---\n", + "OT_DECODE_MMD_WEIGHT = 0.05 # MMD(decode(z1_bar)), target_x) ties decoder to OT geometry\n", + "OT_DECODE_MEAN_WEIGHT = 0.25 # pseudo-bulk for decode(z1_bar) vs target\n", + "DISPLACEMENT_REG_WEIGHT = 0.01 # mean ||pred_z1 - z0||^2 — discourages extreme latent jumps\n", + "VELOCITY_L2_WEIGHT = 0.0 # mean ||v||^2; set e.g. 1e-4 to penalize large speeds\n", + "\n", + "NOISE_STD = 0.05\n", + "SINKHORN_EPSILON = 0.5\n", + "SINKHORN_ITERS = 30\n", + "INFERENCE_STEPS = 24\n", + "PRED_BATCH_SIZE = 256\n", + "\n", + "# --- Optimization & validation (tuning) ---\n", + "LR_SCHEDULER = \"plateau\" # \"none\", \"cosine\", \"plateau\", \"onecycle\"\n", + "PLATEAU_FACTOR = 0.5\n", + "PLATEAU_PATIENCE = 3\n", + "PLATEAU_MIN_LR = 1e-6\n", + "COSINE_MIN_LR = 1e-6\n", + "EARLY_STOP_PATIENCE = 5 # 0 = disabled\n", + "FIXED_VAL_SUBSET = True\n", + "FIXED_VAL_BATCH_SEED = True\n", + "VAL_BATCH_SEED = RANDOM_SEED + 777\n", + "FLOW_LR_MULT = 1.5\n", + "UNCERTAINTY_SAMPLES = 1 # >1: stochastic control resampling; metrics get pred_stochastic_std_mean\n", + "\n", + "INPUT_CANDIDATES = [\n", + " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready_full.h5ad\",\n", + " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready.h5ad\",\n", + " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready_fast_debug.h5ad\",\n", + "]\n", + "INPUT_PATH = next((path for path in INPUT_CANDIDATES if path.exists()), None)\n", + "if INPUT_PATH is None:\n", + " raise FileNotFoundError(\"Could not find a prediction-ready Replogle h5ad. Checked:\\n\" + \"\\n\".join(map(str, INPUT_CANDIDATES)))\n", + "\n", + "OUTPUT_DIR = PROJECT_ROOT / \"baseline\" / \"outputs\" / METHOD_NAME\n", + "OUTPUT_DIR.mkdir(parents=True, exist_ok=True)\n", + "MODEL_PATH = OUTPUT_DIR / \"conditional_flow_matching_model.pt\"\n", + "\n", + "DEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", + "\n", + "np.random.seed(RANDOM_SEED)\n", + "random.seed(RANDOM_SEED)\n", + "torch.manual_seed(RANDOM_SEED)\n", + "if torch.cuda.is_available():\n", + " torch.cuda.manual_seed_all(RANDOM_SEED)\n", + "if torch.backends.cudnn.is_available():\n", + " torch.backends.cudnn.deterministic = True\n", + " torch.backends.cudnn.benchmark = False\n", + "\n", + "print(f\"Project root: {PROJECT_ROOT}\")\n", + "print(f\"Input h5ad: {INPUT_PATH}\")\n", + "print(f\"Output dir: {OUTPUT_DIR}\")\n", + "print(f\"Device: {DEVICE}\")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 2, + "id": "e4871faf", + "metadata": {}, + "outputs": [], + "source": [ + "SCIENCE_COLORS = {\n", + " \"blue\": \"#0072B2\",\n", + " \"orange\": \"#E69F00\",\n", + " \"green\": \"#009E73\",\n", + " \"magenta\": \"#CC79A7\",\n", + " \"dark\": \"#333333\",\n", + " \"gray\": \"#999999\",\n", + "}\n", + "\n", + "\n", + "def setup_science_style() -> None:\n", + " mpl.rcParams.update(\n", + " {\n", + " \"figure.dpi\": 110,\n", + " \"savefig.dpi\": 300,\n", + " \"font.family\": \"sans-serif\",\n", + " \"font.sans-serif\": [\"DejaVu Sans\", \"Helvetica\", \"Arial\"],\n", + " \"font.size\": 9,\n", + " \"axes.titlesize\": 10,\n", + " \"axes.labelsize\": 9,\n", + " \"axes.linewidth\": 0.6,\n", + " \"axes.edgecolor\": SCIENCE_COLORS[\"dark\"],\n", + " \"axes.labelcolor\": SCIENCE_COLORS[\"dark\"],\n", + " \"xtick.labelsize\": 8,\n", + " \"ytick.labelsize\": 8,\n", + " \"xtick.major.width\": 0.6,\n", + " \"ytick.major.width\": 0.6,\n", + " \"legend.frameon\": False,\n", + " \"legend.fontsize\": 8,\n", + " \"axes.spines.top\": False,\n", + " \"axes.spines.right\": False,\n", + " \"figure.facecolor\": \"white\",\n", + " \"axes.facecolor\": \"white\",\n", + " \"grid.alpha\": 0.3,\n", + " \"grid.linewidth\": 0.4,\n", + " \"lines.linewidth\": 1.25,\n", + " \"pdf.fonttype\": 42,\n", + " \"ps.fonttype\": 42,\n", + " }\n", + " )\n", + "\n", + "\n", + "FIGURE_DIR = OUTPUT_DIR / \"figures\"\n", + "FIGURE_DIR.mkdir(parents=True, exist_ok=True)\n", + "setup_science_style()\n", + "\n", + "\n", + "def plot_training_history(history: pd.DataFrame, tag: str) -> None:\n", + " setup_science_style()\n", + " h = history.copy()\n", + " if h.empty:\n", + " return\n", + " fig, axes = plt.subplots(1, 2, figsize=(7.0, 2.45), constrained_layout=True)\n", + " e = h[\"epoch\"].to_numpy()\n", + " ax = axes[0]\n", + " if \"train_total\" in h.columns:\n", + " ax.plot(\n", + " e,\n", + " h[\"train_total\"],\n", + " \"o-\",\n", + " ms=3,\n", + " lw=1.1,\n", + " color=SCIENCE_COLORS[\"blue\"],\n", + " label=\"Train (total)\",\n", + " )\n", + " if \"val_total\" in h.columns and h[\"val_total\"].notna().any():\n", + " ax.plot(\n", + " e,\n", + " h[\"val_total\"],\n", + " \"s-\",\n", + " ms=3,\n", + " lw=1.1,\n", + " color=SCIENCE_COLORS[\"orange\"],\n", + " label=\"Validation (total)\",\n", + " )\n", + " best_i = int(h[\"val_total\"].idxmin())\n", + " best_ep = float(h.loc[best_i, \"epoch\"])\n", + " best_v = float(h.loc[best_i, \"val_total\"])\n", + " ax.scatter(\n", + " [best_ep],\n", + " [best_v],\n", + " s=55,\n", + " zorder=5,\n", + " facecolors=\"none\",\n", + " edgecolors=SCIENCE_COLORS[\"dark\"],\n", + " linewidths=1.1,\n", + " label=\"Best val\",\n", + " )\n", + " ax.set_xlabel(\"Epoch\")\n", + " ax.set_ylabel(\"Loss\")\n", + " ax.legend(loc=\"upper right\", handlelength=2.0)\n", + "\n", + " ax2 = axes[1]\n", + " for col, lab, c in (\n", + " (\"train_flow\", \"Train flow\", SCIENCE_COLORS[\"green\"]),\n", + " (\"train_recon\", \"Train reconstruction\", SCIENCE_COLORS[\"magenta\"]),\n", + " (\"train_ot_decode_mmd\", \"Train OT-decode MMD\", \"#0072B2\"),\n", + " (\"train_disp_reg\", \"Train displacement reg\", \"#009E73\"),\n", + " (\"train_vel_l2\", \"Train velocity L2\", \"#CC79A7\"),\n", + " (\"train_delta_mse\", \"Train delta MSE\", \"#D55E00\"),\n", + " (\"train_delta_mmd\", \"Train delta MMD\", \"#6A3D9A\"),\n", + " ):\n", + " if col in h.columns:\n", + " ax2.plot(e, h[col], \"o-\", ms=2.5, lw=1.0, color=c, label=lab)\n", + " ax2.set_xlabel(\"Epoch\")\n", + " ax2.set_ylabel(\"Component loss\")\n", + " ax2.set_yscale(\"symlog\", linthresh=1e-4)\n", + " ax2.legend(loc=\"upper right\", handlelength=2.0)\n", + "\n", + " fig.suptitle(\n", + " \"Training dynamics — conditional flow matching\",\n", + " fontsize=10.5,\n", + " color=SCIENCE_COLORS[\"dark\"],\n", + " y=1.02,\n", + " )\n", + " for ext in (\"png\", \"pdf\"):\n", + " fig.savefig(FIGURE_DIR / f\"training_history_{tag}.{ext}\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " plt.show()\n", + "\n", + "\n", + "def plot_evaluation_metrics(\n", + " metrics: pd.DataFrame,\n", + " tag: str,\n", + " train_genes: list[str] | None = None,\n", + ") -> None:\n", + " setup_science_style()\n", + " if metrics.empty:\n", + " return\n", + " m = metrics.copy()\n", + " if train_genes is not None:\n", + " seen = set(train_genes)\n", + " m[\"cohort\"] = np.where(m[\"perturbation_gene\"].isin(seen), \"Seen at train\", \"Unseen at train\")\n", + " else:\n", + " m[\"cohort\"] = \"All\"\n", + "\n", + " fig = plt.figure(figsize=(7.2, 5.9), constrained_layout=True)\n", + " gs = fig.add_gridspec(2, 2, height_ratios=[1.05, 1.0], hspace=0.28, wspace=0.28)\n", + "\n", + " ax1 = fig.add_subplot(gs[0, 0])\n", + " if m[\"cohort\"].nunique() > 1:\n", + " palette = {\"Seen at train\": SCIENCE_COLORS[\"blue\"], \"Unseen at train\": SCIENCE_COLORS[\"orange\"]}\n", + " for lab, g in m.groupby(\"cohort\"):\n", + " s = 14 + np.clip(g[\"n_cells\"].to_numpy(dtype=float), 0, 120) * 0.12\n", + " ax1.scatter(\n", + " g[\"mse_mean\"],\n", + " g[\"pearson_mean\"],\n", + " s=s,\n", + " alpha=0.78,\n", + " edgecolors=\"none\",\n", + " c=palette.get(lab, SCIENCE_COLORS[\"gray\"]),\n", + " label=lab,\n", + " )\n", + " ax1.legend(loc=\"lower left\")\n", + " else:\n", + " ax1.scatter(\n", + " m[\"mse_mean\"],\n", + " m[\"pearson_mean\"],\n", + " s=22,\n", + " alpha=0.75,\n", + " edgecolors=\"none\",\n", + " c=SCIENCE_COLORS[\"blue\"],\n", + " )\n", + " ax1.set_xlabel(\"MSE (predicted vs true, gene means)\")\n", + " ax1.set_ylabel(\"Pearson r (gene means)\")\n", + " ax1.set_xscale(\"log\")\n", + " ax1.text(0.02, 0.98, \"a\", transform=ax1.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " ax2 = fig.add_subplot(gs[0, 1])\n", + " m_sorted = m.sort_values(\"pearson_mean\", ascending=True)\n", + " max_rows = 36\n", + " if len(m_sorted) > max_rows:\n", + " head_n = max_rows // 2\n", + " plot_df = pd.concat([m_sorted.head(head_n), m_sorted.tail(head_n)], axis=0)\n", + " else:\n", + " plot_df = m_sorted\n", + " y = np.arange(len(plot_df))\n", + " ax2.barh(y, plot_df[\"pearson_mean\"], height=0.68, color=SCIENCE_COLORS[\"blue\"], alpha=0.88)\n", + " ax2.set_yticks(y)\n", + " ax2.set_yticklabels(plot_df[\"perturbation_gene\"], fontsize=5.8)\n", + " ax2.set_xlabel(\"Pearson r (gene means)\")\n", + " ax2.set_xlim(0, 1.02)\n", + " ax2.text(0.02, 0.98, \"b\", transform=ax2.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " ax3 = fig.add_subplot(gs[1, 0])\n", + " d = m[\"delta_l2\"].dropna().to_numpy()\n", + " if d.size:\n", + " nbin = int(np.clip(len(d) // 2, 10, 32))\n", + " ax3.hist(\n", + " d,\n", + " bins=nbin,\n", + " color=SCIENCE_COLORS[\"green\"],\n", + " alpha=0.88,\n", + " edgecolor=\"white\",\n", + " linewidth=0.35,\n", + " )\n", + " ax3.set_xlabel(\"L2 distance (pred delta vs true delta)\")\n", + " ax3.set_ylabel(\"Perturbations\")\n", + " ax3.text(0.02, 0.98, \"c\", transform=ax3.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " ax4 = fig.add_subplot(gs[1, 1])\n", + " ax4.scatter(\n", + " m[\"pearson_mean\"],\n", + " m[\"pearson_delta\"],\n", + " s=20,\n", + " c=SCIENCE_COLORS[\"orange\"],\n", + " alpha=0.72,\n", + " edgecolors=\"none\",\n", + " )\n", + " pm = m[\"pearson_mean\"].to_numpy(dtype=float)\n", + " pd_ = m[\"pearson_delta\"].to_numpy(dtype=float)\n", + " lo = float(np.nanmin([np.nanmin(pm), np.nanmin(pd_), 0.0]))\n", + " hi = float(np.nanmax([np.nanmax(pm), np.nanmax(pd_), 1.0]))\n", + " pad = 0.03 * (hi - lo + 1e-6)\n", + " ax4.plot([lo, hi], [lo, hi], ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.9, label=\"Identity\")\n", + " ax4.set_xlim(lo - pad, hi + pad)\n", + " ax4.set_ylim(lo - pad, hi + pad)\n", + " ax4.set_xlabel(\"Pearson r (means)\")\n", + " ax4.set_ylabel(\"Pearson r (delta vs control)\")\n", + " ax4.legend(loc=\"lower right\")\n", + " ax4.text(0.02, 0.98, \"d\", transform=ax4.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", + "\n", + " fig.suptitle(\n", + " \"Held-out evaluation — per-perturbation metrics\",\n", + " fontsize=10.5,\n", + " color=SCIENCE_COLORS[\"dark\"],\n", + " )\n", + " for ext in (\"png\", \"pdf\"):\n", + " fig.savefig(FIGURE_DIR / f\"evaluation_metrics_{tag}.{ext}\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " plt.show()\n" + ] + }, + { + "cell_type": "code", + "execution_count": 3, + "id": "b3c3b561", + "metadata": {}, + "outputs": [], + "source": [ + "def normalize_bool(series: pd.Series) -> pd.Series:\n", + " if pd.api.types.is_bool_dtype(series):\n", + " return series.fillna(False).astype(bool)\n", + " text = series.astype(str).str.lower().str.strip()\n", + " return text.isin([\"true\", \"1\", \"yes\", \"y\"])\n", + "\n", + "\n", + "def require_columns(obs: pd.DataFrame, columns: list[str]) -> None:\n", + " missing = [col for col in columns if col not in obs.columns]\n", + " if missing:\n", + " raise KeyError(f\"Missing required obs columns: {missing}\")\n", + "\n", + "\n", + "def take_dense_rows(X, idx: np.ndarray) -> np.ndarray:\n", + " out = X[idx]\n", + " if sp.issparse(out):\n", + " out = out.toarray()\n", + " return np.asarray(out, dtype=np.float32)\n", + "\n", + "\n", + "def mean_vector(X) -> np.ndarray:\n", + " if sp.issparse(X):\n", + " return np.asarray(X.mean(axis=0)).ravel().astype(np.float32)\n", + " return np.asarray(X, dtype=np.float32).mean(axis=0)\n", + "\n", + "\n", + "def safe_pearson(x: np.ndarray, y: np.ndarray) -> float:\n", + " x = np.asarray(x).ravel()\n", + " y = np.asarray(y).ravel()\n", + " if x.size < 2 or np.std(x) == 0 or np.std(y) == 0:\n", + " return np.nan\n", + " return float(np.corrcoef(x, y)[0, 1])\n", + "\n", + "\n", + "def split_array_indices(indices: np.ndarray, rng: np.random.Generator) -> tuple[np.ndarray, np.ndarray, np.ndarray]:\n", + " indices = np.asarray(indices)\n", + " if len(indices) == 0:\n", + " return indices, indices, indices\n", + " shuffled = indices.copy()\n", + " rng.shuffle(shuffled)\n", + " if len(shuffled) < 3:\n", + " return np.sort(shuffled), np.array([], dtype=indices.dtype), np.array([], dtype=indices.dtype)\n", + " n_train = max(1, int(round(len(shuffled) * TRAIN_FRAC)))\n", + " n_val = int(round(len(shuffled) * VAL_FRAC))\n", + " n_train = min(n_train, len(shuffled) - 2)\n", + " n_val = max(1, min(n_val, len(shuffled) - n_train - 1))\n", + " train = np.sort(shuffled[:n_train])\n", + " val = np.sort(shuffled[n_train : n_train + n_val])\n", + " test = np.sort(shuffled[n_train + n_val :])\n", + " if len(test) == 0:\n", + " test = val[-1:].copy()\n", + " val = val[:-1]\n", + " if len(val) == 0:\n", + " val = train[-1:].copy()\n", + " train = train[:-1]\n", + " return train, val, test\n", + "\n", + "\n", + "def assign_seen_cell_split(obs: pd.DataFrame, seed: int) -> pd.Categorical:\n", + " rng = np.random.default_rng(seed)\n", + " split = np.full(len(obs), \"train\", dtype=object)\n", + " is_control = normalize_bool(obs[\"is_control\"]).to_numpy(dtype=bool)\n", + " genes = obs[\"perturbation_gene\"].astype(str).to_numpy()\n", + "\n", + " control_idx = np.where(is_control)[0]\n", + " train_idx, val_idx, test_idx = split_array_indices(control_idx, rng)\n", + " split[train_idx] = \"train\"\n", + " split[val_idx] = \"val\"\n", + " split[test_idx] = \"test\"\n", + "\n", + " for gene in sorted(set(genes[~is_control])):\n", + " idx = np.where((~is_control) & (genes == gene))[0]\n", + " train_idx, val_idx, test_idx = split_array_indices(idx, rng)\n", + " split[train_idx] = \"train\"\n", + " split[val_idx] = \"val\"\n", + " split[test_idx] = \"test\"\n", + "\n", + " return pd.Categorical(split, categories=[\"train\", \"val\", \"test\"], ordered=True)\n", + "\n", + "\n", + "def choose_feature_mask(var: pd.DataFrame) -> np.ndarray:\n", + " if not USE_FEATURE_GENES:\n", + " return np.ones(var.shape[0], dtype=bool)\n", + " for column in FEATURE_GENE_COLUMNS:\n", + " if column in var.columns:\n", + " mask = var[column].astype(bool).to_numpy()\n", + " if mask.any():\n", + " print(f\"Using feature mask from var['{column}']: {int(mask.sum())} genes\")\n", + " return mask\n", + " print(\"Feature-gene columns not found or empty; using all genes.\")\n", + " return np.ones(var.shape[0], dtype=bool)\n", + "\n", + "\n", + "def median_heuristic_sigma(x: torch.Tensor, y: torch.Tensor, max_points: int = 256) -> float:\n", + " z = torch.cat([x[:max_points], y[:max_points]], dim=0)\n", + " if z.shape[0] < 2:\n", + " return 1.0\n", + " d2 = torch.cdist(z, z, p=2).pow(2)\n", + " mask = ~torch.eye(d2.shape[0], dtype=torch.bool, device=d2.device)\n", + " valid = d2[mask]\n", + " valid = valid[valid > 0]\n", + " if valid.numel() == 0:\n", + " return 1.0\n", + " return float(torch.sqrt(valid.median()).item())\n", + "\n", + "\n", + "def rbf_mmd_loss(x: torch.Tensor, y: torch.Tensor, sigma: float | None = None) -> torch.Tensor:\n", + " if x.shape[0] < 2 or y.shape[0] < 2:\n", + " return x.new_tensor(0.0)\n", + " sigma = sigma or median_heuristic_sigma(x, y)\n", + " gamma = 1.0 / (2.0 * sigma * sigma + 1e-8)\n", + " d_xx = torch.cdist(x, x, p=2).pow(2)\n", + " d_yy = torch.cdist(y, y, p=2).pow(2)\n", + " d_xy = torch.cdist(x, y, p=2).pow(2)\n", + " k_xx = torch.exp(-gamma * d_xx)\n", + " k_yy = torch.exp(-gamma * d_yy)\n", + " k_xy = torch.exp(-gamma * d_xy)\n", + " return k_xx.mean() + k_yy.mean() - 2.0 * k_xy.mean()\n", + "\n", + "\n", + "def sinkhorn_barycentric_targets(x0: torch.Tensor, x1: torch.Tensor, epsilon: float, n_iters: int) -> torch.Tensor:\n", + " cost = torch.cdist(x0, x1, p=2).pow(2)\n", + " kernel = torch.exp(-cost / max(epsilon, 1e-4)).clamp_min(1e-8)\n", + "\n", + " a = torch.full((x0.shape[0],), 1.0 / x0.shape[0], device=x0.device, dtype=x0.dtype)\n", + " b = torch.full((x1.shape[0],), 1.0 / x1.shape[0], device=x1.device, dtype=x1.dtype)\n", + " u = torch.ones_like(a)\n", + " v = torch.ones_like(b)\n", + "\n", + " for _ in range(n_iters):\n", + " u = a / (kernel @ v + 1e-8)\n", + " v = b / (kernel.t() @ u + 1e-8)\n", + "\n", + " plan = u[:, None] * kernel * v[None, :]\n", + " row_mass = plan.sum(dim=1, keepdim=True).clamp_min(1e-8)\n", + " return (plan / row_mass) @ x1\n", + "\n", + "\n", + "def build_aggregate_frame(metrics: pd.DataFrame) -> pd.DataFrame:\n", + " numeric = metrics.select_dtypes(include=[np.number])\n", + " if numeric.empty:\n", + " return pd.DataFrame()\n", + " return numeric.agg([\"mean\", \"median\"]).T\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def rk2_integrate_latent(model, z0, gene_idx, batch_idx, steps: int) -> torch.Tensor:\n", + " z = z0\n", + " dt = 1.0 / steps\n", + " for step in range(steps):\n", + " t0 = torch.full((z.shape[0],), step / steps, device=z.device, dtype=z.dtype)\n", + " k1 = model.velocity(z, t0, gene_idx, batch_idx)\n", + " z_mid = z + 0.5 * dt * k1\n", + " t_mid = torch.full((z.shape[0],), (step + 0.5) / steps, device=z.device, dtype=z.dtype)\n", + " k2 = model.velocity(z_mid, t_mid, gene_idx, batch_idx)\n", + " z = z + dt * k2\n", + " return z\n" + ] + }, + { + "cell_type": "markdown", + "id": "1296b782", + "metadata": {}, + "source": [ + "## Load And Prepare Data\n" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "id": "b1ead7e2", + "metadata": {}, + "outputs": [ { - "cell_type": "code", - "execution_count": 21, - "id": "1c11f06f", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n", - "\u001b[0mProject root: /root/final\n", - "Input h5ad: /root/final/data/processed/scPerturb/prediction_ready/replogle_k562_essential_single_gene_prediction_ready_full.h5ad\n", - "Output dir: /root/final/baseline/outputs/conditional_flow_matching_method\n", - "Device: cuda\n" - ] - } - ], - "source": [ - "! pip install -q anndata numpy pandas scipy torch matplotlib\n", - "from pathlib import Path\n", - "from copy import deepcopy\n", - "import json\n", - "import math\n", - "import random\n", - "\n", - "import anndata as ad\n", - "import matplotlib as mpl\n", - "import matplotlib.pyplot as plt\n", - "import numpy as np\n", - "import pandas as pd\n", - "import scipy.sparse as sp\n", - "import torch\n", - "import torch.nn as nn\n", - "import torch.nn.functional as F\n", - "\n", - "\n", - "def find_project_root(start: Path | None = None) -> Path:\n", - " start = (start or Path.cwd()).resolve()\n", - " for path in [start, *start.parents]:\n", - " if (path / \"src\" / \"data\").exists() and (path / \"baseline\").exists():\n", - " return path\n", - " return start\n", - "\n", - "\n", - "PROJECT_ROOT = find_project_root()\n", - "DATASET_NAME = \"ReplogleWeissman2022_K562_essential\"\n", - "METHOD_NAME = \"conditional_flow_matching_method\"\n", - "RANDOM_SEED = 42\n", - "\n", - "SPLIT_STRATEGY = \"seen_cell_split\" # choices: \"seen_cell_split\", \"use_existing\"\n", - "TRAIN_FRAC = 0.80\n", - "VAL_FRAC = 0.10\n", - "EVAL_SPLIT = \"test\"\n", - "CONTROL_SPLIT = \"train\"\n", - "TRAIN_SPLIT = \"train\"\n", - "VAL_SPLIT = \"val\"\n", - "\n", - "USE_FEATURE_GENES = True\n", - "FEATURE_GENE_COLUMNS = [\"is_feature_gene\", \"highly_variable\"]\n", - "\n", - "# Optional runtime caps. Set to None for the full run.\n", - "MAX_TRAIN_CONDITIONS = None\n", - "MAX_EVAL_GENES = None\n", - "MAX_EVAL_CELLS_PER_GENE = None\n", - "\n", - "LATENT_DIM = 128\n", - "HIDDEN_DIM = 512\n", - "CONDITION_DIM = 256\n", - "TIME_DIM = 128\n", - "DROPOUT = 0.08\n", - "NUM_FLOW_BLOCKS = 6\n", - "\n", - "BATCH_SIZE = 128\n", - "EPOCHS = 20\n", - "STEPS_PER_EPOCH = 200\n", - "VALIDATION_STEPS = 48\n", - "LEARNING_RATE = 1e-3\n", - "WEIGHT_DECAY = 1e-5\n", - "GRAD_CLIP_NORM = 1.0\n", - "\n", - "RECON_WEIGHT = 1.0\n", - "FLOW_WEIGHT = 1.5\n", - "ENDPOINT_WEIGHT = 0.5\n", - "MMD_WEIGHT = 0.1\n", - "MEAN_WEIGHT = 1.0\n", - "DELTA_MSE_WEIGHT = 0.0\n", - "DELTA_COS_WEIGHT = 0.0\n", - "DELTA_MMD_WEIGHT = 0.0\n", - "DELTA_BULK_WEIGHT = 0.0\n", - "\n", - "NOISE_STD = 0.05\n", - "SINKHORN_EPSILON = 0.5\n", - "SINKHORN_ITERS = 30\n", - "INFERENCE_STEPS = 24\n", - "PRED_BATCH_SIZE = 256\n", - "\n", - "# --- Optimization & validation (tuning) ---\n", - "LR_SCHEDULER = \"plateau\" # \"none\", \"cosine\", \"plateau\", \"onecycle\"\n", - "PLATEAU_FACTOR = 0.5\n", - "PLATEAU_PATIENCE = 3\n", - "PLATEAU_MIN_LR = 1e-6\n", - "COSINE_MIN_LR = 1e-6\n", - "EARLY_STOP_PATIENCE = 5 # 0 = disabled\n", - "FIXED_VAL_SUBSET = True\n", - "FIXED_VAL_BATCH_SEED = True\n", - "VAL_BATCH_SEED = RANDOM_SEED + 777\n", - "FLOW_LR_MULT = 1.5\n", - "UNCERTAINTY_SAMPLES = 1 # >1: stochastic control resampling; metrics get pred_stochastic_std_mean\n", - "\n", - "INPUT_CANDIDATES = [\n", - " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready_full.h5ad\",\n", - " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready.h5ad\",\n", - " PROJECT_ROOT / \"data\" / \"processed\" / \"scPerturb\" / \"prediction_ready\" / \"replogle_k562_essential_single_gene_prediction_ready_fast_debug.h5ad\",\n", - "]\n", - "INPUT_PATH = next((path for path in INPUT_CANDIDATES if path.exists()), None)\n", - "if INPUT_PATH is None:\n", - " raise FileNotFoundError(\"Could not find a prediction-ready Replogle h5ad. Checked:\\n\" + \"\\n\".join(map(str, INPUT_CANDIDATES)))\n", - "\n", - "OUTPUT_DIR = PROJECT_ROOT / \"baseline\" / \"outputs\" / METHOD_NAME\n", - "OUTPUT_DIR.mkdir(parents=True, exist_ok=True)\n", - "MODEL_PATH = OUTPUT_DIR / \"conditional_flow_matching_model.pt\"\n", - "\n", - "DEVICE = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n", - "\n", - "np.random.seed(RANDOM_SEED)\n", - "random.seed(RANDOM_SEED)\n", - "torch.manual_seed(RANDOM_SEED)\n", - "if torch.cuda.is_available():\n", - " torch.cuda.manual_seed_all(RANDOM_SEED)\n", - "if torch.backends.cudnn.is_available():\n", - " torch.backends.cudnn.deterministic = True\n", - " torch.backends.cudnn.benchmark = False\n", - "\n", - "print(f\"Project root: {PROJECT_ROOT}\")\n", - "print(f\"Input h5ad: {INPUT_PATH}\")\n", - "print(f\"Output dir: {OUTPUT_DIR}\")\n", - "print(f\"Device: {DEVICE}\")\n" - ] - }, + "name": "stdout", + "output_type": "stream", + "text": [ + "Using feature mask from var['is_feature_gene']: 4334 genes\n", + "adata: AnnData object with n_obs × n_vars = 310385 × 4334\n", + " obs: 'batch', 'gene', 'gene_id', 'transcript', 'gene_transcript', 'guide_id', 'percent_mito', 'UMI_count', 'z_gemgroup_UMI', 'core_scale_factor', 'core_adjusted_UMI_count', 'disease', 'cancer', 'cell_line', 'sex', 'age', 'perturbation', 'organism', 'perturbation_type', 'tissue_type', 'ncounts', 'ngenes', 'nperts', 'percent_ribo', 'perturbation_label', 'perturbation_gene', 'is_control', 'is_single_perturbation', 'split', 'condition', 'batch_model', 'source_split'\n", + " var: 'chr', 'start', 'end', 'class', 'strand', 'length', 'in_matrix', 'mean', 'std', 'cv', 'fano', 'ensembl_id', 'ncounts', 'ncells', 'highly_variable', 'highly_variable_rank', 'means', 'variances', 'variances_norm', 'is_target_gene', 'is_feature_gene'\n", + " uns: 'hvg', 'log1p', 'prediction_ready'\n", + " layers: 'counts'\n", + "split counts:\n", + "split\n", + "train 248311\n", + "test 31046\n", + "val 31028\n", + "Name: count, dtype: int64\n", + "train conditions: 2056\n", + "val conditions: 2056\n", + "eval conditions: 2056\n", + "eval genes missing from training: 0\n" + ] + } + ], + "source": [ + "adata = ad.read_h5ad(INPUT_PATH)\n", + "require_columns(adata.obs, [\"perturbation_gene\", \"is_control\", \"split\"])\n", + "\n", + "adata.obs[\"is_control\"] = normalize_bool(adata.obs[\"is_control\"])\n", + "adata.obs[\"perturbation_gene\"] = adata.obs[\"perturbation_gene\"].astype(str).str.strip()\n", + "adata.obs[\"perturbation_gene\"] = adata.obs[\"perturbation_gene\"].mask(adata.obs[\"is_control\"], \"__ctrl__\")\n", + "adata.obs[\"condition\"] = adata.obs[\"perturbation_gene\"].astype(str)\n", + "adata.obs[\"batch_model\"] = adata.obs[\"batch\"].astype(str) if \"batch\" in adata.obs.columns else \"global\"\n", + "\n", + "feature_mask = choose_feature_mask(adata.var)\n", + "adata = adata[:, feature_mask].copy()\n", + "\n", + "if SPLIT_STRATEGY == \"seen_cell_split\":\n", + " adata.obs[\"source_split\"] = adata.obs[\"split\"].astype(str)\n", + " adata.obs[\"split\"] = assign_seen_cell_split(adata.obs, RANDOM_SEED)\n", + "elif SPLIT_STRATEGY == \"use_existing\":\n", + " adata.obs[\"split\"] = pd.Categorical(\n", + " adata.obs[\"split\"].astype(str),\n", + " categories=[\"train\", \"val\", \"test\"],\n", + " ordered=True,\n", + " )\n", + "else:\n", + " raise ValueError(f\"Unknown SPLIT_STRATEGY: {SPLIT_STRATEGY}\")\n", + "\n", + "is_control = adata.obs[\"is_control\"].to_numpy(dtype=bool)\n", + "split = adata.obs[\"split\"].astype(str).to_numpy()\n", + "genes = adata.obs[\"perturbation_gene\"].astype(str).to_numpy()\n", + "batches = adata.obs[\"batch_model\"].astype(str).to_numpy()\n", + "\n", + "control_train_idx = np.where(is_control & (split == CONTROL_SPLIT))[0]\n", + "if len(control_train_idx) == 0:\n", + " print(f\"No controls found in CONTROL_SPLIT={CONTROL_SPLIT}; falling back to all controls.\")\n", + " control_train_idx = np.where(is_control)[0]\n", + "if len(control_train_idx) == 0:\n", + " raise ValueError(\"No control cells found.\")\n", + "\n", + "train_conditions = sorted(set(genes[(~is_control) & (split == TRAIN_SPLIT)]))\n", + "val_conditions = sorted(set(genes[(~is_control) & (split == VAL_SPLIT)]))\n", + "eval_conditions = sorted(set(genes[(~is_control) & (split == EVAL_SPLIT)]))\n", + "\n", + "if MAX_TRAIN_CONDITIONS is not None:\n", + " train_conditions = train_conditions[:MAX_TRAIN_CONDITIONS]\n", + " train_condition_set = set(train_conditions)\n", + " keep_mask = is_control | np.isin(genes, list(train_condition_set)) | np.isin(genes, eval_conditions) | np.isin(genes, val_conditions)\n", + " adata = adata[keep_mask].copy()\n", + " is_control = adata.obs[\"is_control\"].to_numpy(dtype=bool)\n", + " split = adata.obs[\"split\"].astype(str).to_numpy()\n", + " genes = adata.obs[\"perturbation_gene\"].astype(str).to_numpy()\n", + " batches = adata.obs[\"batch_model\"].astype(str).to_numpy()\n", + " control_train_idx = np.where(is_control & (split == CONTROL_SPLIT))[0]\n", + " train_conditions = sorted(set(genes[(~is_control) & (split == TRAIN_SPLIT)]))\n", + " val_conditions = sorted(set(genes[(~is_control) & (split == VAL_SPLIT)]))\n", + " eval_conditions = sorted(set(genes[(~is_control) & (split == EVAL_SPLIT)]))\n", + "\n", + "if MAX_EVAL_GENES is not None:\n", + " eval_conditions = eval_conditions[:MAX_EVAL_GENES]\n", + "\n", + "indices_by_split_gene = {}\n", + "for split_name in [TRAIN_SPLIT, VAL_SPLIT, EVAL_SPLIT]:\n", + " indices_by_split_gene[split_name] = {}\n", + " split_mask = (~is_control) & (split == split_name)\n", + " for gene in sorted(set(genes[split_mask])):\n", + " indices = np.where(split_mask & (genes == gene))[0]\n", + " if len(indices) > 0:\n", + " indices_by_split_gene[split_name][gene] = indices\n", + "\n", + "control_idx_by_batch = {}\n", + "for batch_name in sorted(set(batches[control_train_idx])):\n", + " batch_idx = control_train_idx[batches[control_train_idx] == batch_name]\n", + " if len(batch_idx) > 0:\n", + " control_idx_by_batch[batch_name] = batch_idx\n", + "\n", + "gene_vocab = [\"__ctrl__\", \"__unk__\", *train_conditions]\n", + "gene_to_idx = {gene: idx for idx, gene in enumerate(gene_vocab)}\n", + "batch_vocab = [\"__unk__\", *sorted(set(batches))]\n", + "batch_to_idx = {batch: idx for idx, batch in enumerate(batch_vocab)}\n", + "\n", + "gene_index_per_obs = np.array([gene_to_idx.get(gene, gene_to_idx[\"__unk__\"]) for gene in genes], dtype=np.int64)\n", + "batch_index_per_obs = np.array([batch_to_idx.get(batch, batch_to_idx[\"__unk__\"]) for batch in batches], dtype=np.int64)\n", + "\n", + "control_mean = mean_vector(adata.X[control_train_idx])\n", + "unseen_eval_genes = sorted(set(eval_conditions) - set(train_conditions))\n", + "\n", + "print(\"adata:\", adata)\n", + "print(\"split counts:\")\n", + "print(adata.obs[\"split\"].astype(str).value_counts())\n", + "print(\"train conditions:\", len(train_conditions))\n", + "print(\"val conditions:\", len(val_conditions))\n", + "print(\"eval conditions:\", len(eval_conditions))\n", + "print(\"eval genes missing from training:\", len(unseen_eval_genes))\n", + "if unseen_eval_genes[:10]:\n", + " print(\"first unseen eval genes:\", unseen_eval_genes[:10])\n" + ] + }, + { + "cell_type": "markdown", + "id": "3fa220a2", + "metadata": {}, + "source": [ + "## Model Definition\n" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "id": "ec5b92e5", + "metadata": {}, + "outputs": [ { - "cell_type": "code", - "execution_count": 22, - "id": "e4871faf", - "metadata": {}, - "outputs": [], - "source": [ - "SCIENCE_COLORS = {\n", - " \"blue\": \"#0072B2\",\n", - " \"orange\": \"#E69F00\",\n", - " \"green\": \"#009E73\",\n", - " \"magenta\": \"#CC79A7\",\n", - " \"dark\": \"#333333\",\n", - " \"gray\": \"#999999\",\n", - "}\n", - "\n", - "\n", - "def setup_science_style() -> None:\n", - " mpl.rcParams.update(\n", - " {\n", - " \"figure.dpi\": 110,\n", - " \"savefig.dpi\": 300,\n", - " \"font.family\": \"sans-serif\",\n", - " \"font.sans-serif\": [\"DejaVu Sans\", \"Helvetica\", \"Arial\"],\n", - " \"font.size\": 9,\n", - " \"axes.titlesize\": 10,\n", - " \"axes.labelsize\": 9,\n", - " \"axes.linewidth\": 0.6,\n", - " \"axes.edgecolor\": SCIENCE_COLORS[\"dark\"],\n", - " \"axes.labelcolor\": SCIENCE_COLORS[\"dark\"],\n", - " \"xtick.labelsize\": 8,\n", - " \"ytick.labelsize\": 8,\n", - " \"xtick.major.width\": 0.6,\n", - " \"ytick.major.width\": 0.6,\n", - " \"legend.frameon\": False,\n", - " \"legend.fontsize\": 8,\n", - " \"axes.spines.top\": False,\n", - " \"axes.spines.right\": False,\n", - " \"figure.facecolor\": \"white\",\n", - " \"axes.facecolor\": \"white\",\n", - " \"grid.alpha\": 0.3,\n", - " \"grid.linewidth\": 0.4,\n", - " \"lines.linewidth\": 1.25,\n", - " \"pdf.fonttype\": 42,\n", - " \"ps.fonttype\": 42,\n", - " }\n", - " )\n", - "\n", - "\n", - "FIGURE_DIR = OUTPUT_DIR / \"figures\"\n", - "FIGURE_DIR.mkdir(parents=True, exist_ok=True)\n", - "setup_science_style()\n", - "\n", - "\n", - "def plot_training_history(history: pd.DataFrame, tag: str) -> None:\n", - " setup_science_style()\n", - " h = history.copy()\n", - " if h.empty:\n", - " return\n", - " fig, axes = plt.subplots(1, 2, figsize=(7.0, 2.45), constrained_layout=True)\n", - " e = h[\"epoch\"].to_numpy()\n", - " ax = axes[0]\n", - " if \"train_total\" in h.columns:\n", - " ax.plot(\n", - " e,\n", - " h[\"train_total\"],\n", - " \"o-\",\n", - " ms=3,\n", - " lw=1.1,\n", - " color=SCIENCE_COLORS[\"blue\"],\n", - " label=\"Train (total)\",\n", - " )\n", - " if \"val_total\" in h.columns and h[\"val_total\"].notna().any():\n", - " ax.plot(\n", - " e,\n", - " h[\"val_total\"],\n", - " \"s-\",\n", - " ms=3,\n", - " lw=1.1,\n", - " color=SCIENCE_COLORS[\"orange\"],\n", - " label=\"Validation (total)\",\n", - " )\n", - " best_i = int(h[\"val_total\"].idxmin())\n", - " best_ep = float(h.loc[best_i, \"epoch\"])\n", - " best_v = float(h.loc[best_i, \"val_total\"])\n", - " ax.scatter(\n", - " [best_ep],\n", - " [best_v],\n", - " s=55,\n", - " zorder=5,\n", - " facecolors=\"none\",\n", - " edgecolors=SCIENCE_COLORS[\"dark\"],\n", - " linewidths=1.1,\n", - " label=\"Best val\",\n", - " )\n", - " ax.set_xlabel(\"Epoch\")\n", - " ax.set_ylabel(\"Loss\")\n", - " ax.legend(loc=\"upper right\", handlelength=2.0)\n", - "\n", - " ax2 = axes[1]\n", - " for col, lab, c in (\n", - " (\"train_flow\", \"Train flow\", SCIENCE_COLORS[\"green\"]),\n", - " (\"train_recon\", \"Train reconstruction\", SCIENCE_COLORS[\"magenta\"]),\n", - " (\"train_delta_mse\", \"Train delta MSE\", \"#D55E00\"),\n", - " (\"train_delta_mmd\", \"Train delta MMD\", \"#6A3D9A\"),\n", - " ):\n", - " if col in h.columns:\n", - " ax2.plot(e, h[col], \"o-\", ms=2.5, lw=1.0, color=c, label=lab)\n", - " ax2.set_xlabel(\"Epoch\")\n", - " ax2.set_ylabel(\"Component loss\")\n", - " ax2.set_yscale(\"symlog\", linthresh=1e-4)\n", - " ax2.legend(loc=\"upper right\", handlelength=2.0)\n", - "\n", - " fig.suptitle(\n", - " \"Training dynamics — conditional flow matching\",\n", - " fontsize=10.5,\n", - " color=SCIENCE_COLORS[\"dark\"],\n", - " y=1.02,\n", - " )\n", - " for ext in (\"png\", \"pdf\"):\n", - " fig.savefig(FIGURE_DIR / f\"training_history_{tag}.{ext}\", bbox_inches=\"tight\", facecolor=\"white\")\n", - " plt.show()\n", - "\n", - "\n", - "def plot_evaluation_metrics(\n", - " metrics: pd.DataFrame,\n", - " tag: str,\n", - " train_genes: list[str] | None = None,\n", - ") -> None:\n", - " setup_science_style()\n", - " if metrics.empty:\n", - " return\n", - " m = metrics.copy()\n", - " if train_genes is not None:\n", - " seen = set(train_genes)\n", - " m[\"cohort\"] = np.where(m[\"perturbation_gene\"].isin(seen), \"Seen at train\", \"Unseen at train\")\n", - " else:\n", - " m[\"cohort\"] = \"All\"\n", - "\n", - " fig = plt.figure(figsize=(7.2, 5.9), constrained_layout=True)\n", - " gs = fig.add_gridspec(2, 2, height_ratios=[1.05, 1.0], hspace=0.28, wspace=0.28)\n", - "\n", - " ax1 = fig.add_subplot(gs[0, 0])\n", - " if m[\"cohort\"].nunique() > 1:\n", - " palette = {\"Seen at train\": SCIENCE_COLORS[\"blue\"], \"Unseen at train\": SCIENCE_COLORS[\"orange\"]}\n", - " for lab, g in m.groupby(\"cohort\"):\n", - " s = 14 + np.clip(g[\"n_cells\"].to_numpy(dtype=float), 0, 120) * 0.12\n", - " ax1.scatter(\n", - " g[\"mse_mean\"],\n", - " g[\"pearson_mean\"],\n", - " s=s,\n", - " alpha=0.78,\n", - " edgecolors=\"none\",\n", - " c=palette.get(lab, SCIENCE_COLORS[\"gray\"]),\n", - " label=lab,\n", - " )\n", - " ax1.legend(loc=\"lower left\")\n", - " else:\n", - " ax1.scatter(\n", - " m[\"mse_mean\"],\n", - " m[\"pearson_mean\"],\n", - " s=22,\n", - " alpha=0.75,\n", - " edgecolors=\"none\",\n", - " c=SCIENCE_COLORS[\"blue\"],\n", - " )\n", - " ax1.set_xlabel(\"MSE (predicted vs true, gene means)\")\n", - " ax1.set_ylabel(\"Pearson r (gene means)\")\n", - " ax1.set_xscale(\"log\")\n", - " ax1.text(0.02, 0.98, \"a\", transform=ax1.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", - "\n", - " ax2 = fig.add_subplot(gs[0, 1])\n", - " m_sorted = m.sort_values(\"pearson_mean\", ascending=True)\n", - " max_rows = 36\n", - " if len(m_sorted) > max_rows:\n", - " head_n = max_rows // 2\n", - " plot_df = pd.concat([m_sorted.head(head_n), m_sorted.tail(head_n)], axis=0)\n", - " else:\n", - " plot_df = m_sorted\n", - " y = np.arange(len(plot_df))\n", - " ax2.barh(y, plot_df[\"pearson_mean\"], height=0.68, color=SCIENCE_COLORS[\"blue\"], alpha=0.88)\n", - " ax2.set_yticks(y)\n", - " ax2.set_yticklabels(plot_df[\"perturbation_gene\"], fontsize=5.8)\n", - " ax2.set_xlabel(\"Pearson r (gene means)\")\n", - " ax2.set_xlim(0, 1.02)\n", - " ax2.text(0.02, 0.98, \"b\", transform=ax2.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", - "\n", - " ax3 = fig.add_subplot(gs[1, 0])\n", - " d = m[\"delta_l2\"].dropna().to_numpy()\n", - " if d.size:\n", - " nbin = int(np.clip(len(d) // 2, 10, 32))\n", - " ax3.hist(\n", - " d,\n", - " bins=nbin,\n", - " color=SCIENCE_COLORS[\"green\"],\n", - " alpha=0.88,\n", - " edgecolor=\"white\",\n", - " linewidth=0.35,\n", - " )\n", - " ax3.set_xlabel(\"L2 distance (pred delta vs true delta)\")\n", - " ax3.set_ylabel(\"Perturbations\")\n", - " ax3.text(0.02, 0.98, \"c\", transform=ax3.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", - "\n", - " ax4 = fig.add_subplot(gs[1, 1])\n", - " ax4.scatter(\n", - " m[\"pearson_mean\"],\n", - " m[\"pearson_delta\"],\n", - " s=20,\n", - " c=SCIENCE_COLORS[\"orange\"],\n", - " alpha=0.72,\n", - " edgecolors=\"none\",\n", - " )\n", - " pm = m[\"pearson_mean\"].to_numpy(dtype=float)\n", - " pd_ = m[\"pearson_delta\"].to_numpy(dtype=float)\n", - " lo = float(np.nanmin([np.nanmin(pm), np.nanmin(pd_), 0.0]))\n", - " hi = float(np.nanmax([np.nanmax(pm), np.nanmax(pd_), 1.0]))\n", - " pad = 0.03 * (hi - lo + 1e-6)\n", - " ax4.plot([lo, hi], [lo, hi], ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.9, label=\"Identity\")\n", - " ax4.set_xlim(lo - pad, hi + pad)\n", - " ax4.set_ylim(lo - pad, hi + pad)\n", - " ax4.set_xlabel(\"Pearson r (means)\")\n", - " ax4.set_ylabel(\"Pearson r (delta vs control)\")\n", - " ax4.legend(loc=\"lower right\")\n", - " ax4.text(0.02, 0.98, \"d\", transform=ax4.transAxes, fontsize=11, fontweight=\"bold\", va=\"top\")\n", - "\n", - " fig.suptitle(\n", - " \"Held-out evaluation — per-perturbation metrics\",\n", - " fontsize=10.5,\n", - " color=SCIENCE_COLORS[\"dark\"],\n", - " )\n", - " for ext in (\"png\", \"pdf\"):\n", - " fig.savefig(FIGURE_DIR / f\"evaluation_metrics_{tag}.{ext}\", bbox_inches=\"tight\", facecolor=\"white\")\n", - " plt.show()\n" - ] + "name": "stdout", + "output_type": "stream", + "text": [ + "ConditionalFlowMatchingModel(\n", + " (encoder): MLPEncoder(\n", + " (net): Sequential(\n", + " (0): Linear(in_features=4334, out_features=1024, bias=True)\n", + " (1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", + " (2): GELU(approximate='none')\n", + " (3): Dropout(p=0.08, inplace=False)\n", + " (4): Linear(in_features=1024, out_features=1024, bias=True)\n", + " (5): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", + " (6): GELU(approximate='none')\n", + " (7): Dropout(p=0.08, inplace=False)\n", + " (8): Linear(in_features=1024, out_features=512, bias=True)\n", + " (9): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", + " (10): GELU(approximate='none')\n", + " (11): Dropout(p=0.08, inplace=False)\n", + " (12): Linear(in_features=512, out_features=128, bias=True)\n", + " )\n", + " )\n", + " (decoder): MLPDecoder(\n", + " (net): Sequential(\n", + " (0): Linear(in_features=128, out_features=512, bias=True)\n", + " (1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", + " (2): GELU(approximate='none')\n", + " (3): Dropout(p=0.08, inplace=False)\n", + " (4): Linear(in_features=512, out_features=1024, bias=True)\n", + " (5): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", + " (6): GELU(approximate='none')\n", + " (7): Dropout(p=0.08, inplace=False)\n", + " (8): Linear(in_features=1024, out_features=1024, bias=True)\n", + " (9): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", + " (10): GELU(approximate='none')\n", + " (11): Dropout(p=0.08, inplace=False)\n", + " (12): Linear(in_features=1024, out_features=4334, bias=True)\n", + " )\n", + " )\n", + " (condition_encoder): ConditionEncoder(\n", + " (gene_embedding): Embedding(2058, 128)\n", + " (batch_embedding): Embedding(49, 64)\n", + " (mlp): Sequential(\n", + " (0): Linear(in_features=192, out_features=256, bias=True)\n", + " (1): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n", + " (2): GELU(approximate='none')\n", + " (3): Dropout(p=0.08, inplace=False)\n", + " (4): Linear(in_features=256, out_features=256, bias=True)\n", + " )\n", + " )\n", + " (time_encoder): TimeEmbedding(\n", + " (proj): Sequential(\n", + " (0): Linear(in_features=128, out_features=128, bias=True)\n", + " (1): SiLU()\n", + " (2): Linear(in_features=128, out_features=128, bias=True)\n", + " )\n", + " )\n", + " (vector_field): ConditionalVectorField(\n", + " (input_proj): Linear(in_features=256, out_features=512, bias=True)\n", + " (blocks): ModuleList(\n", + " (0-5): 6 x FiLMResidualBlock(\n", + " (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", + " (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", + " (fc1): Linear(in_features=512, out_features=512, bias=True)\n", + " (fc2): Linear(in_features=512, out_features=512, bias=True)\n", + " (film1): Linear(in_features=256, out_features=1024, bias=True)\n", + " (film2): Linear(in_features=256, out_features=1024, bias=True)\n", + " (dropout): Dropout(p=0.08, inplace=False)\n", + " )\n", + " )\n", + " (out): Linear(in_features=512, out_features=128, bias=True)\n", + " )\n", + ")\n", + "LR scheduler: ReduceLROnPlateau | VAL_EVAL_SUBSET: 48 | flow_lr_mult: 1.5\n" + ] + } + ], + "source": [ + "class MLPEncoder(nn.Module):\n", + " def __init__(self, input_dim: int, latent_dim: int, hidden_dim: int, dropout: float):\n", + " super().__init__()\n", + " h2 = hidden_dim * 2\n", + " self.net = nn.Sequential(\n", + " nn.Linear(input_dim, h2),\n", + " nn.LayerNorm(h2),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(h2, h2),\n", + " nn.LayerNorm(h2),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(h2, hidden_dim),\n", + " nn.LayerNorm(hidden_dim),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(hidden_dim, latent_dim),\n", + " )\n", + "\n", + " def forward(self, x: torch.Tensor) -> torch.Tensor:\n", + " return self.net(x)\n", + "\n", + "\n", + "class MLPDecoder(nn.Module):\n", + " def __init__(self, latent_dim: int, output_dim: int, hidden_dim: int, dropout: float):\n", + " super().__init__()\n", + " h2 = hidden_dim * 2\n", + " self.net = nn.Sequential(\n", + " nn.Linear(latent_dim, hidden_dim),\n", + " nn.LayerNorm(hidden_dim),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(hidden_dim, h2),\n", + " nn.LayerNorm(h2),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(h2, h2),\n", + " nn.LayerNorm(h2),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(h2, output_dim),\n", + " )\n", + "\n", + " def forward(self, z: torch.Tensor) -> torch.Tensor:\n", + " return F.softplus(self.net(z))\n", + "\n", + "\n", + "# Condition h = MLP([gene_emb; batch_emb]) — \"基因 + batch\" only (no perturbation-type / cell-type).\n", + "class ConditionEncoder(nn.Module):\n", + " def __init__(self, n_genes: int, n_batches: int, condition_dim: int, dropout: float):\n", + " super().__init__()\n", + " gene_dim = max(32, condition_dim // 2)\n", + " batch_dim = max(16, condition_dim // 4)\n", + " self.gene_embedding = nn.Embedding(n_genes, gene_dim)\n", + " self.batch_embedding = nn.Embedding(n_batches, batch_dim)\n", + " self.mlp = nn.Sequential(\n", + " nn.Linear(gene_dim + batch_dim, condition_dim),\n", + " nn.LayerNorm(condition_dim),\n", + " nn.GELU(),\n", + " nn.Dropout(dropout),\n", + " nn.Linear(condition_dim, condition_dim),\n", + " )\n", + "\n", + " def forward(self, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", + " gene_emb = self.gene_embedding(gene_idx)\n", + " batch_emb = self.batch_embedding(batch_idx)\n", + " return self.mlp(torch.cat([gene_emb, batch_emb], dim=-1))\n", + "\n", + "\n", + "class TimeEmbedding(nn.Module):\n", + " def __init__(self, time_dim: int):\n", + " super().__init__()\n", + " self.time_dim = time_dim\n", + " self.proj = nn.Sequential(\n", + " nn.Linear(time_dim, time_dim),\n", + " nn.SiLU(),\n", + " nn.Linear(time_dim, time_dim),\n", + " )\n", + "\n", + " def forward(self, t: torch.Tensor) -> torch.Tensor:\n", + " half = self.time_dim // 2\n", + " freqs = torch.exp(\n", + " -math.log(10000.0) * torch.arange(half, device=t.device, dtype=t.dtype) / max(half, 1)\n", + " )\n", + " angles = t[:, None] * freqs[None, :]\n", + " emb = torch.cat([torch.cos(angles), torch.sin(angles)], dim=-1)\n", + " if emb.shape[1] < self.time_dim:\n", + " emb = F.pad(emb, (0, self.time_dim - emb.shape[1]))\n", + " return self.proj(emb)\n", + "\n", + "\n", + "class FiLMResidualBlock(nn.Module):\n", + " def __init__(self, hidden_dim: int, condition_dim: int, dropout: float):\n", + " super().__init__()\n", + " self.norm1 = nn.LayerNorm(hidden_dim)\n", + " self.norm2 = nn.LayerNorm(hidden_dim)\n", + " self.fc1 = nn.Linear(hidden_dim, hidden_dim)\n", + " self.fc2 = nn.Linear(hidden_dim, hidden_dim)\n", + " self.film1 = nn.Linear(condition_dim, hidden_dim * 2)\n", + " self.film2 = nn.Linear(condition_dim, hidden_dim * 2)\n", + " self.dropout = nn.Dropout(dropout)\n", + "\n", + " def forward(self, x: torch.Tensor, condition: torch.Tensor) -> torch.Tensor:\n", + " gamma1, beta1 = self.film1(condition).chunk(2, dim=-1)\n", + " h = self.norm1(x)\n", + " h = h * (1 + gamma1) + beta1\n", + " h = F.gelu(self.fc1(h))\n", + " h = self.dropout(h)\n", + "\n", + " gamma2, beta2 = self.film2(condition).chunk(2, dim=-1)\n", + " h = self.norm2(h)\n", + " h = h * (1 + gamma2) + beta2\n", + " h = self.fc2(h)\n", + " h = self.dropout(h)\n", + " return x + h\n", + "\n", + "\n", + "class ConditionalVectorField(nn.Module):\n", + " def __init__(self, latent_dim: int, time_dim: int, condition_dim: int, hidden_dim: int, num_blocks: int, dropout: float):\n", + " super().__init__()\n", + " self.input_proj = nn.Linear(latent_dim + time_dim, hidden_dim)\n", + " self.blocks = nn.ModuleList(\n", + " [FiLMResidualBlock(hidden_dim, condition_dim, dropout) for _ in range(num_blocks)]\n", + " )\n", + " self.out = nn.Linear(hidden_dim, latent_dim)\n", + "\n", + " def forward(self, z_t: torch.Tensor, t_emb: torch.Tensor, condition: torch.Tensor) -> torch.Tensor:\n", + " h = self.input_proj(torch.cat([z_t, t_emb], dim=-1))\n", + " for block in self.blocks:\n", + " h = block(h, condition)\n", + " return self.out(h)\n", + "\n", + "\n", + "class ConditionalFlowMatchingModel(nn.Module):\n", + " def __init__(self, input_dim: int, latent_dim: int, hidden_dim: int, condition_dim: int, time_dim: int, n_genes: int, n_batches: int, num_blocks: int, dropout: float):\n", + " super().__init__()\n", + " self.encoder = MLPEncoder(input_dim=input_dim, latent_dim=latent_dim, hidden_dim=hidden_dim, dropout=dropout)\n", + " self.decoder = MLPDecoder(latent_dim=latent_dim, output_dim=input_dim, hidden_dim=hidden_dim, dropout=dropout)\n", + " self.condition_encoder = ConditionEncoder(n_genes=n_genes, n_batches=n_batches, condition_dim=condition_dim, dropout=dropout)\n", + " self.time_encoder = TimeEmbedding(time_dim=time_dim)\n", + " self.vector_field = ConditionalVectorField(\n", + " latent_dim=latent_dim,\n", + " time_dim=time_dim,\n", + " condition_dim=condition_dim,\n", + " hidden_dim=hidden_dim,\n", + " num_blocks=num_blocks,\n", + " dropout=dropout,\n", + " )\n", + "\n", + " def encode_x(self, x: torch.Tensor) -> torch.Tensor:\n", + " return self.encoder(x)\n", + "\n", + " def decode_z(self, z: torch.Tensor) -> torch.Tensor:\n", + " return self.decoder(z)\n", + "\n", + " def reconstruct(self, x: torch.Tensor) -> torch.Tensor:\n", + " return self.decode_z(self.encode_x(x))\n", + "\n", + " def condition(self, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", + " return self.condition_encoder(gene_idx, batch_idx)\n", + "\n", + " def velocity(self, z_t: torch.Tensor, t: torch.Tensor, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", + " condition = self.condition(gene_idx, batch_idx)\n", + " t_emb = self.time_encoder(t)\n", + " return self.vector_field(z_t, t_emb, condition)\n", + "\n", + "\n", + "model = ConditionalFlowMatchingModel(\n", + " input_dim=adata.n_vars,\n", + " latent_dim=LATENT_DIM,\n", + " hidden_dim=HIDDEN_DIM,\n", + " condition_dim=CONDITION_DIM,\n", + " time_dim=TIME_DIM,\n", + " n_genes=len(gene_vocab),\n", + " n_batches=len(batch_vocab),\n", + " num_blocks=NUM_FLOW_BLOCKS,\n", + " dropout=DROPOUT,\n", + ").to(DEVICE)\n", + "\n", + "ae_params = list(model.encoder.parameters()) + list(model.decoder.parameters())\n", + "flow_params = (\n", + " list(model.vector_field.parameters())\n", + " + list(model.time_encoder.parameters())\n", + " + list(model.condition_encoder.parameters())\n", + ")\n", + "optimizer = torch.optim.AdamW(\n", + " [\n", + " {\"params\": ae_params, \"lr\": LEARNING_RATE},\n", + " {\"params\": flow_params, \"lr\": LEARNING_RATE * FLOW_LR_MULT},\n", + " ],\n", + " weight_decay=WEIGHT_DECAY,\n", + ")\n", + "\n", + "if val_conditions:\n", + " _n = min(len(val_conditions), VALIDATION_STEPS)\n", + " VAL_EVAL_SUBSET = sorted(val_conditions)[:_n]\n", + "else:\n", + " VAL_EVAL_SUBSET = []\n", + "\n", + "lr_scheduler = None\n", + "if LR_SCHEDULER == \"plateau\":\n", + " lr_scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(\n", + " optimizer, mode=\"min\", factor=PLATEAU_FACTOR, patience=PLATEAU_PATIENCE, min_lr=PLATEAU_MIN_LR\n", + " )\n", + "elif LR_SCHEDULER == \"cosine\":\n", + " lr_scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=EPOCHS, eta_min=COSINE_MIN_LR)\n", + "elif LR_SCHEDULER == \"onecycle\":\n", + " total_steps = EPOCHS * STEPS_PER_EPOCH\n", + " lr_scheduler = torch.optim.lr_scheduler.OneCycleLR(\n", + " optimizer,\n", + " max_lr=[LEARNING_RATE, LEARNING_RATE * FLOW_LR_MULT],\n", + " total_steps=total_steps,\n", + " pct_start=0.1,\n", + " )\n", + "\n", + "print(model)\n", + "print(\n", + " \"LR scheduler:\",\n", + " type(lr_scheduler).__name__ if lr_scheduler is not None else \"none\",\n", + " \"| VAL_EVAL_SUBSET:\",\n", + " len(VAL_EVAL_SUBSET),\n", + " \"| flow_lr_mult:\",\n", + " FLOW_LR_MULT,\n", + ")\n" + ] + }, + { + "cell_type": "code", + "execution_count": 6, + "id": "91749254", + "metadata": {}, + "outputs": [], + "source": [ + "def sample_controls_like_targets(target_idx: np.ndarray, rng: np.random.Generator) -> np.ndarray:\n", + " sampled = []\n", + " for idx in target_idx:\n", + " batch_name = batches[idx]\n", + " pool = control_idx_by_batch.get(batch_name, control_train_idx)\n", + " sampled.append(rng.choice(pool))\n", + " return np.asarray(sampled, dtype=np.int64)\n", + "\n", + "\n", + "def resample_unpaired_controls(\n", + " source_idx: np.ndarray, target_idx: np.ndarray, rng: np.random.Generator\n", + ") -> np.ndarray:\n", + " \"\"\"Ensure source (control) indices never equal target (perturbed) indices.\"\"\"\n", + " source_idx = source_idx.copy()\n", + " for _ in range(64):\n", + " clash = source_idx == target_idx\n", + " if not np.any(clash):\n", + " break\n", + " for i in np.flatnonzero(clash):\n", + " batch_name = batches[target_idx[i]]\n", + " pool = control_idx_by_batch.get(batch_name, control_train_idx)\n", + " source_idx[i] = rng.choice(pool)\n", + " return source_idx\n", + "\n", + "\n", + "def make_condition_batch(condition: str, split_name: str, rng: np.random.Generator) -> dict:\n", + " target_pool = indices_by_split_gene[split_name][condition]\n", + " size = min(BATCH_SIZE, len(target_pool))\n", + " replace = len(target_pool) < size\n", + " target_idx = rng.choice(target_pool, size=size, replace=replace)\n", + " source_idx = sample_controls_like_targets(target_idx, rng)\n", + " source_idx = resample_unpaired_controls(source_idx, target_idx, rng)\n", + "\n", + " if MAX_EVAL_CELLS_PER_GENE is not None and len(target_idx) > MAX_EVAL_CELLS_PER_GENE:\n", + " target_idx = target_idx[:MAX_EVAL_CELLS_PER_GENE]\n", + " source_idx = source_idx[:MAX_EVAL_CELLS_PER_GENE]\n", + "\n", + " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", + " target_x = torch.from_numpy(take_dense_rows(adata.X, target_idx)).to(DEVICE)\n", + " gene_idx = torch.full(\n", + " (len(target_idx),),\n", + " fill_value=gene_to_idx.get(condition, gene_to_idx[\"__unk__\"]),\n", + " device=DEVICE,\n", + " dtype=torch.long,\n", + " )\n", + " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", + "\n", + " return {\n", + " \"condition\": condition,\n", + " \"source_idx\": source_idx,\n", + " \"target_idx\": target_idx,\n", + " \"source_x\": source_x,\n", + " \"target_x\": target_x,\n", + " \"gene_idx\": gene_idx,\n", + " \"batch_idx\": batch_idx,\n", + " }\n", + "\n", + "\n", + "def compute_loss_terms(batch: dict) -> dict[str, torch.Tensor]:\n", + " source_x = batch[\"source_x\"]\n", + " target_x = batch[\"target_x\"]\n", + " gene_idx = batch[\"gene_idx\"]\n", + " batch_idx = batch[\"batch_idx\"]\n", + "\n", + " z0 = model.encode_x(source_x)\n", + " z1 = model.encode_x(target_x)\n", + " z1_bar = sinkhorn_barycentric_targets(\n", + " x0=z0,\n", + " x1=z1,\n", + " epsilon=SINKHORN_EPSILON,\n", + " n_iters=SINKHORN_ITERS,\n", + " )\n", + "\n", + " t = torch.rand(z0.shape[0], device=DEVICE)\n", + " sigma = NOISE_STD * torch.sqrt((t * (1.0 - t)).clamp_min(1e-6)).unsqueeze(-1)\n", + " z_t = (1.0 - t).unsqueeze(-1) * z0 + t.unsqueeze(-1) * z1_bar + sigma * torch.randn_like(z0)\n", + "\n", + " target_velocity = z1_bar - z0\n", + " pred_velocity = model.velocity(z_t, t, gene_idx, batch_idx)\n", + " pred_z1 = z_t + (1.0 - t).unsqueeze(-1) * pred_velocity\n", + " pred_x1 = model.decode_z(pred_z1)\n", + " x_ot = model.decode_z(z1_bar)\n", + "\n", + " source_recon = model.reconstruct(source_x)\n", + " target_recon = model.reconstruct(target_x)\n", + "\n", + " recon_loss = F.mse_loss(source_recon, source_x) + F.mse_loss(target_recon, target_x)\n", + " flow_loss = F.mse_loss(pred_velocity, target_velocity)\n", + " endpoint_loss = F.mse_loss(pred_z1, z1_bar)\n", + " mmd_loss = rbf_mmd_loss(pred_x1, target_x)\n", + " mean_loss = F.mse_loss(pred_x1.mean(dim=0), target_x.mean(dim=0))\n", + " ot_decode_mmd = rbf_mmd_loss(x_ot, target_x)\n", + " ot_decode_mean = F.mse_loss(x_ot.mean(dim=0), target_x.mean(dim=0))\n", + " disp_reg = torch.mean((pred_z1 - z0) ** 2)\n", + " vel_l2 = torch.mean(pred_velocity ** 2)\n", + "\n", + " total_loss = (\n", + " RECON_WEIGHT * recon_loss\n", + " + FLOW_WEIGHT * flow_loss\n", + " + ENDPOINT_WEIGHT * endpoint_loss\n", + " + MMD_WEIGHT * mmd_loss\n", + " + MEAN_WEIGHT * mean_loss\n", + " + OT_DECODE_MMD_WEIGHT * ot_decode_mmd\n", + " + OT_DECODE_MEAN_WEIGHT * ot_decode_mean\n", + " + DISPLACEMENT_REG_WEIGHT * disp_reg\n", + " + VELOCITY_L2_WEIGHT * vel_l2\n", + " )\n", + "\n", + " return {\n", + " \"total\": total_loss,\n", + " \"recon\": recon_loss.detach(),\n", + " \"flow\": flow_loss.detach(),\n", + " \"endpoint\": endpoint_loss.detach(),\n", + " \"mmd\": mmd_loss.detach(),\n", + " \"mean\": mean_loss.detach(),\n", + " \"ot_decode_mmd\": ot_decode_mmd.detach(),\n", + " \"ot_decode_mean\": ot_decode_mean.detach(),\n", + " \"disp_reg\": disp_reg.detach(),\n", + " \"vel_l2\": vel_l2.detach(),\n", + " }\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def evaluate_split_loss(\n", + " split_name: str,\n", + " conditions: list[str],\n", + " seed: int,\n", + " steps: int,\n", + " fixed_condition_subset: list[str] | None = None,\n", + " batch_rng_seed: int | None = None,\n", + ") -> dict[str, float]:\n", + " if fixed_condition_subset is not None:\n", + " if len(fixed_condition_subset) == 0:\n", + " return {}\n", + " elif not conditions:\n", + " return {}\n", + "\n", + " model.eval()\n", + " if fixed_condition_subset is not None:\n", + " subset = fixed_condition_subset[: min(len(fixed_condition_subset), steps)]\n", + " else:\n", + " rng = np.random.default_rng(seed)\n", + " subset = conditions.copy()\n", + " rng.shuffle(subset)\n", + " subset = subset[: min(len(subset), steps)]\n", + "\n", + " brng = batch_rng_seed if batch_rng_seed is not None else seed\n", + " rng = np.random.default_rng(brng)\n", + "\n", + " rows = []\n", + " for condition in subset:\n", + " batch = make_condition_batch(condition, split_name=split_name, rng=rng)\n", + " losses = compute_loss_terms(batch)\n", + " rows.append({key: float(value.item()) for key, value in losses.items()})\n", + "\n", + " return pd.DataFrame(rows).mean().to_dict()\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def predict_gene_expression(\n", + " condition: str,\n", + " split_name: str,\n", + " rng: np.random.Generator,\n", + " n_stochastic: int = 1,\n", + ") -> tuple[np.ndarray, np.ndarray, pd.DataFrame, np.ndarray]:\n", + " target_idx_all = indices_by_split_gene[split_name][condition]\n", + " if MAX_EVAL_CELLS_PER_GENE is not None and len(target_idx_all) > MAX_EVAL_CELLS_PER_GENE:\n", + " target_idx_all = target_idx_all[:MAX_EVAL_CELLS_PER_GENE]\n", + "\n", + " pred_blocks = []\n", + " pred_std_blocks = []\n", + " real_blocks = []\n", + " obs_blocks = []\n", + " gene_id = gene_to_idx.get(condition, gene_to_idx[\"__unk__\"])\n", + " condition_source = \"trained_gene_embedding\" if condition in gene_to_idx else \"unk_gene_embedding\"\n", + "\n", + " for start in range(0, len(target_idx_all), PRED_BATCH_SIZE):\n", + " target_idx = target_idx_all[start : start + PRED_BATCH_SIZE]\n", + " if n_stochastic <= 1:\n", + " source_idx = sample_controls_like_targets(target_idx, rng)\n", + " source_idx = resample_unpaired_controls(source_idx, target_idx, rng)\n", + " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", + " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", + " gene_idx = torch.full((len(target_idx),), gene_id, device=DEVICE, dtype=torch.long)\n", + " z0 = model.encode_x(source_x)\n", + " z1_hat = rk2_integrate_latent(model, z0, gene_idx, batch_idx, steps=INFERENCE_STEPS)\n", + " pred_x = model.decode_z(z1_hat).cpu().numpy().astype(np.float32)\n", + " pred_std_blocks.append(np.zeros_like(pred_x, dtype=np.float32))\n", + " else:\n", + " runs = []\n", + " for s in range(n_stochastic):\n", + " sub_rng = np.random.default_rng(int(rng.integers(0, 2**31 - 1)) + s * 1000003)\n", + " source_idx = sample_controls_like_targets(target_idx, sub_rng)\n", + " source_idx = resample_unpaired_controls(source_idx, target_idx, sub_rng)\n", + " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", + " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", + " gene_idx = torch.full((len(target_idx),), gene_id, device=DEVICE, dtype=torch.long)\n", + " z0 = model.encode_x(source_x)\n", + " z1_hat = rk2_integrate_latent(model, z0, gene_idx, batch_idx, steps=INFERENCE_STEPS)\n", + " runs.append(model.decode_z(z1_hat).cpu().numpy().astype(np.float32))\n", + " stacked = np.stack(runs, axis=0)\n", + " pred_x = stacked.mean(axis=0)\n", + " pred_std_blocks.append(stacked.std(axis=0))\n", + "\n", + " real_x = take_dense_rows(adata.X, target_idx)\n", + " pred_blocks.append(pred_x)\n", + " real_blocks.append(real_x)\n", + "\n", + " obs_chunk = adata.obs.iloc[target_idx][[\"perturbation_gene\", \"condition\", \"split\"]].copy()\n", + " obs_chunk[\"target_cell_id\"] = adata.obs_names[target_idx].astype(str)\n", + " if n_stochastic <= 1:\n", + " obs_chunk[\"sampled_control_cell_id\"] = adata.obs_names[source_idx].astype(str)\n", + " else:\n", + " obs_chunk[\"sampled_control_cell_id\"] = \"stochastic_mean\"\n", + " obs_chunk[\"condition_source\"] = condition_source\n", + " obs_blocks.append(obs_chunk)\n", + "\n", + " pred_gene = np.vstack(pred_blocks).astype(np.float32)\n", + " pred_std_gene = np.vstack(pred_std_blocks).astype(np.float32)\n", + " real_gene = np.vstack(real_blocks).astype(np.float32)\n", + " obs_gene = pd.concat(obs_blocks, axis=0)\n", + " return pred_gene, real_gene, obs_gene, pred_std_gene\n", + "\n", + "\n", + "@torch.no_grad()\n", + "def evaluate_gene_metrics(split_name: str, conditions: list[str], seed: int) -> tuple[pd.DataFrame, ad.AnnData, ad.AnnData]:\n", + " rng = np.random.default_rng(seed)\n", + "\n", + " pred_blocks = []\n", + " real_blocks = []\n", + " obs_blocks = []\n", + " metrics_rows = []\n", + "\n", + " for condition in conditions:\n", + " pred_gene, real_gene, obs_gene, pred_std_gene = predict_gene_expression(\n", + " condition=condition, split_name=split_name, rng=rng, n_stochastic=UNCERTAINTY_SAMPLES\n", + " )\n", + "\n", + " pred_blocks.append(pred_gene)\n", + " real_blocks.append(real_gene)\n", + " obs_blocks.append(obs_gene)\n", + "\n", + " pred_mean = pred_gene.mean(axis=0)\n", + " real_mean = real_gene.mean(axis=0)\n", + " pred_delta = pred_mean - control_mean\n", + " real_delta = real_mean - control_mean\n", + "\n", + " row = {\n", + " \"perturbation_gene\": condition,\n", + " \"n_cells\": int(real_gene.shape[0]),\n", + " \"mse_mean\": float(np.mean((pred_mean - real_mean) ** 2)),\n", + " \"mae_mean\": float(np.mean(np.abs(pred_mean - real_mean))),\n", + " \"pearson_mean\": safe_pearson(pred_mean, real_mean),\n", + " \"pearson_delta\": safe_pearson(pred_delta, real_delta),\n", + " \"delta_l2\": float(np.linalg.norm(pred_delta - real_delta)),\n", + " }\n", + " if UNCERTAINTY_SAMPLES > 1:\n", + " row[\"pred_stochastic_std_mean\"] = float(np.mean(pred_std_gene))\n", + " metrics_rows.append(row)\n", + "\n", + " pred = ad.AnnData(\n", + " X=np.vstack(pred_blocks).astype(np.float32),\n", + " obs=pd.concat(obs_blocks, axis=0),\n", + " var=adata.var.copy(),\n", + " )\n", + " real = ad.AnnData(\n", + " X=np.vstack(real_blocks).astype(np.float32),\n", + " obs=pd.concat(obs_blocks, axis=0),\n", + " var=adata.var.copy(),\n", + " )\n", + " metrics = pd.DataFrame(metrics_rows).sort_values(\"perturbation_gene\").reset_index(drop=True)\n", + " return metrics, pred, real\n" + ] + }, + { + "cell_type": "markdown", + "id": "e2bf6516", + "metadata": {}, + "source": [ + "## Train The Conditional Flow Matching Model\n" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "id": "27f82aeb", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "{'epoch': 1, 'train_total': 0.7261, 'train_flow': 0.2356, 'train_recon': 0.2876, 'val_total': 0.3572, 'lr': 0.001}\n", + "{'epoch': 2, 'train_total': 0.322, 'train_flow': 0.0056, 'train_recon': 0.2768, 'val_total': 0.3224, 'lr': 0.001}\n", + "{'epoch': 3, 'train_total': 0.3096, 'train_flow': 0.0009, 'train_recon': 0.2731, 'val_total': 0.3219, 'lr': 0.001}\n", + "{'epoch': 4, 'train_total': 0.3087, 'train_flow': 0.0006, 'train_recon': 0.2731, 'val_total': 0.3197, 'lr': 0.001}\n", + "{'epoch': 5, 'train_total': 0.3108, 'train_flow': 0.0004, 'train_recon': 0.2751, 'val_total': 0.3196, 'lr': 0.001}\n", + "{'epoch': 6, 'train_total': 0.3125, 'train_flow': 0.0003, 'train_recon': 0.2764, 'val_total': 0.3186, 'lr': 0.001}\n", + "{'epoch': 7, 'train_total': 0.3083, 'train_flow': 0.0002, 'train_recon': 0.2729, 'val_total': 0.3191, 'lr': 0.001}\n", + "{'epoch': 8, 'train_total': 0.3069, 'train_flow': 0.0002, 'train_recon': 0.272, 'val_total': 0.3208, 'lr': 0.001}\n", + "{'epoch': 9, 'train_total': 0.3083, 'train_flow': 0.0001, 'train_recon': 0.2734, 'val_total': 0.3183, 'lr': 0.001}\n", + "{'epoch': 10, 'train_total': 0.312, 'train_flow': 0.0001, 'train_recon': 0.2762, 'val_total': 0.3204, 'lr': 0.001}\n", + "{'epoch': 11, 'train_total': 0.3084, 'train_flow': 0.0001, 'train_recon': 0.2737, 'val_total': 0.3186, 'lr': 0.001}\n", + "{'epoch': 12, 'train_total': 0.3104, 'train_flow': 0.0001, 'train_recon': 0.2757, 'val_total': 0.3184, 'lr': 0.001}\n", + "{'epoch': 13, 'train_total': 0.3096, 'train_flow': 0.0001, 'train_recon': 0.2749, 'val_total': 0.319, 'lr': 0.0005}\n", + "{'epoch': 14, 'train_total': 0.3072, 'train_flow': 0.0, 'train_recon': 0.2732, 'val_total': 0.3175, 'lr': 0.0005}\n", + "{'epoch': 15, 'train_total': 0.3046, 'train_flow': 0.0, 'train_recon': 0.2713, 'val_total': 0.3193, 'lr': 0.0005}\n", + "{'epoch': 16, 'train_total': 0.3084, 'train_flow': 0.0, 'train_recon': 0.2745, 'val_total': 0.3177, 'lr': 0.0005}\n", + "{'epoch': 17, 'train_total': 0.3057, 'train_flow': 0.0, 'train_recon': 0.2722, 'val_total': 0.3182, 'lr': 0.0005}\n", + "{'epoch': 18, 'train_total': 0.3075, 'train_flow': 0.0, 'train_recon': 0.2735, 'val_total': 0.3174, 'lr': 0.0005}\n", + "{'epoch': 19, 'train_total': 0.3051, 'train_flow': 0.0, 'train_recon': 0.2718, 'val_total': 0.318, 'lr': 0.0005}\n", + "{'epoch': 20, 'train_total': 0.3029, 'train_flow': 0.0, 'train_recon': 0.2694, 'val_total': 0.3179, 'lr': 0.0005}\n" + ] }, { - "cell_type": "code", - "execution_count": 23, - "id": "b3c3b561", - "metadata": {}, - "outputs": [], - "source": [ - "def normalize_bool(series: pd.Series) -> pd.Series:\n", - " if pd.api.types.is_bool_dtype(series):\n", - " return series.fillna(False).astype(bool)\n", - " text = series.astype(str).str.lower().str.strip()\n", - " return text.isin([\"true\", \"1\", \"yes\", \"y\"])\n", - "\n", - "\n", - "def require_columns(obs: pd.DataFrame, columns: list[str]) -> None:\n", - " missing = [col for col in columns if col not in obs.columns]\n", - " if missing:\n", - " raise KeyError(f\"Missing required obs columns: {missing}\")\n", - "\n", - "\n", - "def take_dense_rows(X, idx: np.ndarray) -> np.ndarray:\n", - " out = X[idx]\n", - " if sp.issparse(out):\n", - " out = out.toarray()\n", - " return np.asarray(out, dtype=np.float32)\n", - "\n", - "\n", - "def mean_vector(X) -> np.ndarray:\n", - " if sp.issparse(X):\n", - " return np.asarray(X.mean(axis=0)).ravel().astype(np.float32)\n", - " return np.asarray(X, dtype=np.float32).mean(axis=0)\n", - "\n", - "\n", - "def safe_pearson(x: np.ndarray, y: np.ndarray) -> float:\n", - " x = np.asarray(x).ravel()\n", - " y = np.asarray(y).ravel()\n", - " if x.size < 2 or np.std(x) == 0 or np.std(y) == 0:\n", - " return np.nan\n", - " return float(np.corrcoef(x, y)[0, 1])\n", - "\n", - "\n", - "def split_array_indices(indices: np.ndarray, rng: np.random.Generator) -> tuple[np.ndarray, np.ndarray, np.ndarray]:\n", - " indices = np.asarray(indices)\n", - " if len(indices) == 0:\n", - " return indices, indices, indices\n", - " shuffled = indices.copy()\n", - " rng.shuffle(shuffled)\n", - " if len(shuffled) < 3:\n", - " return np.sort(shuffled), np.array([], dtype=indices.dtype), np.array([], dtype=indices.dtype)\n", - " n_train = max(1, int(round(len(shuffled) * TRAIN_FRAC)))\n", - " n_val = int(round(len(shuffled) * VAL_FRAC))\n", - " n_train = min(n_train, len(shuffled) - 2)\n", - " n_val = max(1, min(n_val, len(shuffled) - n_train - 1))\n", - " train = np.sort(shuffled[:n_train])\n", - " val = np.sort(shuffled[n_train : n_train + n_val])\n", - " test = np.sort(shuffled[n_train + n_val :])\n", - " if len(test) == 0:\n", - " test = val[-1:].copy()\n", - " val = val[:-1]\n", - " if len(val) == 0:\n", - " val = train[-1:].copy()\n", - " train = train[:-1]\n", - " return train, val, test\n", - "\n", - "\n", - "def assign_seen_cell_split(obs: pd.DataFrame, seed: int) -> pd.Categorical:\n", - " rng = np.random.default_rng(seed)\n", - " split = np.full(len(obs), \"train\", dtype=object)\n", - " is_control = normalize_bool(obs[\"is_control\"]).to_numpy(dtype=bool)\n", - " genes = obs[\"perturbation_gene\"].astype(str).to_numpy()\n", - "\n", - " control_idx = np.where(is_control)[0]\n", - " train_idx, val_idx, test_idx = split_array_indices(control_idx, rng)\n", - " split[train_idx] = \"train\"\n", - " split[val_idx] = \"val\"\n", - " split[test_idx] = \"test\"\n", - "\n", - " for gene in sorted(set(genes[~is_control])):\n", - " idx = np.where((~is_control) & (genes == gene))[0]\n", - " train_idx, val_idx, test_idx = split_array_indices(idx, rng)\n", - " split[train_idx] = \"train\"\n", - " split[val_idx] = \"val\"\n", - " split[test_idx] = \"test\"\n", - "\n", - " return pd.Categorical(split, categories=[\"train\", \"val\", \"test\"], ordered=True)\n", - "\n", - "\n", - "def choose_feature_mask(var: pd.DataFrame) -> np.ndarray:\n", - " if not USE_FEATURE_GENES:\n", - " return np.ones(var.shape[0], dtype=bool)\n", - " for column in FEATURE_GENE_COLUMNS:\n", - " if column in var.columns:\n", - " mask = var[column].astype(bool).to_numpy()\n", - " if mask.any():\n", - " print(f\"Using feature mask from var['{column}']: {int(mask.sum())} genes\")\n", - " return mask\n", - " print(\"Feature-gene columns not found or empty; using all genes.\")\n", - " return np.ones(var.shape[0], dtype=bool)\n", - "\n", - "\n", - "def median_heuristic_sigma(x: torch.Tensor, y: torch.Tensor, max_points: int = 256) -> float:\n", - " z = torch.cat([x[:max_points], y[:max_points]], dim=0)\n", - " if z.shape[0] < 2:\n", - " return 1.0\n", - " d2 = torch.cdist(z, z, p=2).pow(2)\n", - " mask = ~torch.eye(d2.shape[0], dtype=torch.bool, device=d2.device)\n", - " valid = d2[mask]\n", - " valid = valid[valid > 0]\n", - " if valid.numel() == 0:\n", - " return 1.0\n", - " return float(torch.sqrt(valid.median()).item())\n", - "\n", - "\n", - "def rbf_mmd_loss(x: torch.Tensor, y: torch.Tensor, sigma: float | None = None) -> torch.Tensor:\n", - " if x.shape[0] < 2 or y.shape[0] < 2:\n", - " return x.new_tensor(0.0)\n", - " sigma = sigma or median_heuristic_sigma(x, y)\n", - " gamma = 1.0 / (2.0 * sigma * sigma + 1e-8)\n", - " d_xx = torch.cdist(x, x, p=2).pow(2)\n", - " d_yy = torch.cdist(y, y, p=2).pow(2)\n", - " d_xy = torch.cdist(x, y, p=2).pow(2)\n", - " k_xx = torch.exp(-gamma * d_xx)\n", - " k_yy = torch.exp(-gamma * d_yy)\n", - " k_xy = torch.exp(-gamma * d_xy)\n", - " return k_xx.mean() + k_yy.mean() - 2.0 * k_xy.mean()\n", - "\n", - "\n", - "def sinkhorn_barycentric_targets(x0: torch.Tensor, x1: torch.Tensor, epsilon: float, n_iters: int) -> torch.Tensor:\n", - " cost = torch.cdist(x0, x1, p=2).pow(2)\n", - " kernel = torch.exp(-cost / max(epsilon, 1e-4)).clamp_min(1e-8)\n", - "\n", - " a = torch.full((x0.shape[0],), 1.0 / x0.shape[0], device=x0.device, dtype=x0.dtype)\n", - " b = torch.full((x1.shape[0],), 1.0 / x1.shape[0], device=x1.device, dtype=x1.dtype)\n", - " u = torch.ones_like(a)\n", - " v = torch.ones_like(b)\n", - "\n", - " for _ in range(n_iters):\n", - " u = a / (kernel @ v + 1e-8)\n", - " v = b / (kernel.t() @ u + 1e-8)\n", - "\n", - " plan = u[:, None] * kernel * v[None, :]\n", - " row_mass = plan.sum(dim=1, keepdim=True).clamp_min(1e-8)\n", - " return (plan / row_mass) @ x1\n", - "\n", - "\n", - "def build_aggregate_frame(metrics: pd.DataFrame) -> pd.DataFrame:\n", - " numeric = metrics.select_dtypes(include=[np.number])\n", - " if numeric.empty:\n", - " return pd.DataFrame()\n", - " return numeric.agg([\"mean\", \"median\"]).T\n", - "\n", - "\n", - "@torch.no_grad()\n", - "def rk2_integrate_latent(model, z0, gene_idx, batch_idx, steps: int) -> torch.Tensor:\n", - " z = z0\n", - " dt = 1.0 / steps\n", - " for step in range(steps):\n", - " t0 = torch.full((z.shape[0],), step / steps, device=z.device, dtype=z.dtype)\n", - " k1 = model.velocity(z, t0, gene_idx, batch_idx)\n", - " z_mid = z + 0.5 * dt * k1\n", - " t_mid = torch.full((z.shape[0],), (step + 0.5) / steps, device=z.device, dtype=z.dtype)\n", - " k2 = model.velocity(z_mid, t_mid, gene_idx, batch_idx)\n", - " z = z + dt * k2\n", - " return z\n" + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
epochtrain_totaltrain_recontrain_flowtrain_endpointtrain_mmdtrain_meantrain_ot_decode_mmdtrain_ot_decode_meantrain_disp_reg...val_totalval_reconval_flowval_endpointval_mmdval_meanval_ot_decode_mmdval_ot_decode_meanval_disp_regval_vel_l2
010.7260860.2875970.2356180.0788300.1904310.0130610.1907110.0128740.078863...0.3572330.2700250.0192850.0063760.2326060.0165780.2208820.0165810.0063760.019285
120.3219510.2768320.0055640.0022450.1822670.0066300.1822600.0066320.002248...0.3223580.2681410.0006240.0006240.2227930.0156180.2232240.0156180.0006240.000624
230.3096130.2731090.0009280.0007220.1819450.0059600.1819870.0059570.000724...0.3218670.2685120.0002470.0005000.2218050.0158970.2135800.0158970.0005000.000247
340.3086550.2730630.0005540.0006010.1811110.0058320.1810940.0058270.000600...0.3196910.2672350.0001090.0004520.2199890.0152240.2206510.0152240.0004520.000109
450.3108050.2751490.0003630.0005340.1819210.0060410.1819200.0060420.000534...0.3195580.2676340.0001080.0004580.2182500.0153240.2109940.0153240.0004580.000108
560.3125280.2763590.0002810.0005050.1828270.0064530.1828230.0064560.000505...0.3185670.2669990.0000800.0004460.2173840.0150670.2129760.0150670.0004460.000080
670.3083390.2728670.0002110.0004830.1833400.0059280.1833330.0059260.000483...0.3191220.2673650.0000570.0004270.2186000.0152630.2103170.0152630.0004270.000057
780.3068900.2720270.0001590.0004650.1818420.0056900.1818230.0056900.000465...0.3208150.2681770.0000360.0004210.2186800.0155800.2205390.0155800.0004210.000036
890.3083170.2733630.0001320.0004600.1823200.0057370.1823300.0057410.000460...0.3183220.2667490.0000340.0004150.2173380.0149700.2172910.0149700.0004150.000034
9100.3119560.2761950.0001190.0004490.1828010.0063460.1828000.0063490.000449...0.3203590.2680120.0000320.0004310.2180440.0154700.2187470.0154700.0004310.000032
10110.3083870.2736840.0000940.0004420.1803440.0058280.1803410.0058280.000442...0.3185820.2668600.0000240.0004080.2165530.0150410.2204230.0150410.0004080.000024
11120.3104450.2756630.0000770.0004380.1808430.0058520.1808870.0058530.000438...0.3183730.2669540.0000140.0004220.2162610.0150090.2159130.0150090.0004220.000014
12130.3095720.2748600.0000640.0004310.1807770.0058230.1807750.0058240.000431...0.3190460.2669580.0000130.0004080.2166010.0150540.2276460.0150540.0004080.000013
13140.3071900.2732070.0000480.0004270.1806400.0052750.1806900.0052750.000427...0.3175300.2665520.0000060.0004090.2155010.0148520.2129160.0148520.0004090.000006
14150.3046040.2713300.0000430.0004240.1790070.0049140.1790170.0049120.000424...0.3192890.2673790.0000050.0004150.2174760.0152190.2183940.0152190.0004150.000005
15160.3084000.2745480.0000400.0004240.1796470.0053040.1796470.0053050.000424...0.3177500.2666000.0000050.0004090.2160360.0148510.2153160.0148510.0004090.000005
16170.3056990.2722130.0000370.0004230.1785510.0051460.1785490.0051460.000423...0.3182280.2668570.0000050.0004080.2162360.0149490.2169270.0149490.0004080.000005
17180.3075270.2734610.0000340.0004190.1795870.0054910.1795810.0054890.000419...0.3173830.2667660.0000050.0004110.2157170.0148970.2041300.0148970.0004110.000005
18190.3050710.2717680.0000320.0004210.1784010.0050250.1784070.0050230.000421...0.3179690.2670770.0000050.0004070.2162720.0150690.2042900.0150690.0004070.000005
19200.3028730.2694020.0000300.0004210.1794070.0050410.1794030.0050390.000421...0.3178810.2667010.0000070.0004120.2159590.0149030.2146770.0149030.0004120.000007
\n", + "

20 rows × 21 columns

\n", + "
" + ], + "text/plain": [ + " epoch train_total train_recon train_flow train_endpoint train_mmd \\\n", + "0 1 0.726086 0.287597 0.235618 0.078830 0.190431 \n", + "1 2 0.321951 0.276832 0.005564 0.002245 0.182267 \n", + "2 3 0.309613 0.273109 0.000928 0.000722 0.181945 \n", + "3 4 0.308655 0.273063 0.000554 0.000601 0.181111 \n", + "4 5 0.310805 0.275149 0.000363 0.000534 0.181921 \n", + "5 6 0.312528 0.276359 0.000281 0.000505 0.182827 \n", + "6 7 0.308339 0.272867 0.000211 0.000483 0.183340 \n", + "7 8 0.306890 0.272027 0.000159 0.000465 0.181842 \n", + "8 9 0.308317 0.273363 0.000132 0.000460 0.182320 \n", + "9 10 0.311956 0.276195 0.000119 0.000449 0.182801 \n", + "10 11 0.308387 0.273684 0.000094 0.000442 0.180344 \n", + "11 12 0.310445 0.275663 0.000077 0.000438 0.180843 \n", + "12 13 0.309572 0.274860 0.000064 0.000431 0.180777 \n", + "13 14 0.307190 0.273207 0.000048 0.000427 0.180640 \n", + "14 15 0.304604 0.271330 0.000043 0.000424 0.179007 \n", + "15 16 0.308400 0.274548 0.000040 0.000424 0.179647 \n", + "16 17 0.305699 0.272213 0.000037 0.000423 0.178551 \n", + "17 18 0.307527 0.273461 0.000034 0.000419 0.179587 \n", + "18 19 0.305071 0.271768 0.000032 0.000421 0.178401 \n", + "19 20 0.302873 0.269402 0.000030 0.000421 0.179407 \n", + "\n", + " train_mean train_ot_decode_mmd train_ot_decode_mean train_disp_reg \\\n", + "0 0.013061 0.190711 0.012874 0.078863 \n", + "1 0.006630 0.182260 0.006632 0.002248 \n", + "2 0.005960 0.181987 0.005957 0.000724 \n", + "3 0.005832 0.181094 0.005827 0.000600 \n", + "4 0.006041 0.181920 0.006042 0.000534 \n", + "5 0.006453 0.182823 0.006456 0.000505 \n", + "6 0.005928 0.183333 0.005926 0.000483 \n", + "7 0.005690 0.181823 0.005690 0.000465 \n", + "8 0.005737 0.182330 0.005741 0.000460 \n", + "9 0.006346 0.182800 0.006349 0.000449 \n", + "10 0.005828 0.180341 0.005828 0.000442 \n", + "11 0.005852 0.180887 0.005853 0.000438 \n", + "12 0.005823 0.180775 0.005824 0.000431 \n", + "13 0.005275 0.180690 0.005275 0.000427 \n", + "14 0.004914 0.179017 0.004912 0.000424 \n", + "15 0.005304 0.179647 0.005305 0.000424 \n", + "16 0.005146 0.178549 0.005146 0.000423 \n", + "17 0.005491 0.179581 0.005489 0.000419 \n", + "18 0.005025 0.178407 0.005023 0.000421 \n", + "19 0.005041 0.179403 0.005039 0.000421 \n", + "\n", + " ... val_total val_recon val_flow val_endpoint val_mmd val_mean \\\n", + "0 ... 0.357233 0.270025 0.019285 0.006376 0.232606 0.016578 \n", + "1 ... 0.322358 0.268141 0.000624 0.000624 0.222793 0.015618 \n", + "2 ... 0.321867 0.268512 0.000247 0.000500 0.221805 0.015897 \n", + "3 ... 0.319691 0.267235 0.000109 0.000452 0.219989 0.015224 \n", + "4 ... 0.319558 0.267634 0.000108 0.000458 0.218250 0.015324 \n", + "5 ... 0.318567 0.266999 0.000080 0.000446 0.217384 0.015067 \n", + "6 ... 0.319122 0.267365 0.000057 0.000427 0.218600 0.015263 \n", + "7 ... 0.320815 0.268177 0.000036 0.000421 0.218680 0.015580 \n", + "8 ... 0.318322 0.266749 0.000034 0.000415 0.217338 0.014970 \n", + "9 ... 0.320359 0.268012 0.000032 0.000431 0.218044 0.015470 \n", + "10 ... 0.318582 0.266860 0.000024 0.000408 0.216553 0.015041 \n", + "11 ... 0.318373 0.266954 0.000014 0.000422 0.216261 0.015009 \n", + "12 ... 0.319046 0.266958 0.000013 0.000408 0.216601 0.015054 \n", + "13 ... 0.317530 0.266552 0.000006 0.000409 0.215501 0.014852 \n", + "14 ... 0.319289 0.267379 0.000005 0.000415 0.217476 0.015219 \n", + "15 ... 0.317750 0.266600 0.000005 0.000409 0.216036 0.014851 \n", + "16 ... 0.318228 0.266857 0.000005 0.000408 0.216236 0.014949 \n", + "17 ... 0.317383 0.266766 0.000005 0.000411 0.215717 0.014897 \n", + "18 ... 0.317969 0.267077 0.000005 0.000407 0.216272 0.015069 \n", + "19 ... 0.317881 0.266701 0.000007 0.000412 0.215959 0.014903 \n", + "\n", + " val_ot_decode_mmd val_ot_decode_mean val_disp_reg val_vel_l2 \n", + "0 0.220882 0.016581 0.006376 0.019285 \n", + "1 0.223224 0.015618 0.000624 0.000624 \n", + "2 0.213580 0.015897 0.000500 0.000247 \n", + "3 0.220651 0.015224 0.000452 0.000109 \n", + "4 0.210994 0.015324 0.000458 0.000108 \n", + "5 0.212976 0.015067 0.000446 0.000080 \n", + "6 0.210317 0.015263 0.000427 0.000057 \n", + "7 0.220539 0.015580 0.000421 0.000036 \n", + "8 0.217291 0.014970 0.000415 0.000034 \n", + "9 0.218747 0.015470 0.000431 0.000032 \n", + "10 0.220423 0.015041 0.000408 0.000024 \n", + "11 0.215913 0.015009 0.000422 0.000014 \n", + "12 0.227646 0.015054 0.000408 0.000013 \n", + "13 0.212916 0.014852 0.000409 0.000006 \n", + "14 0.218394 0.015219 0.000415 0.000005 \n", + "15 0.215316 0.014851 0.000409 0.000005 \n", + "16 0.216927 0.014949 0.000408 0.000005 \n", + "17 0.204130 0.014897 0.000411 0.000005 \n", + "18 0.204290 0.015069 0.000407 0.000005 \n", + "19 0.214677 0.014903 0.000412 0.000007 \n", + "\n", + "[20 rows x 21 columns]" ] - }, + }, + "execution_count": 7, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "history_rows = []\n", + "best_state = None\n", + "best_val_total = float(\"inf\")\n", + "rng = np.random.default_rng(RANDOM_SEED)\n", + "epochs_no_improve = 0\n", + "\n", + "train_sampling_conditions = [g for g in train_conditions if g in indices_by_split_gene[TRAIN_SPLIT]]\n", + "if not train_sampling_conditions:\n", + " raise ValueError(\"No train perturbation conditions are available.\")\n", + "\n", + "for epoch in range(1, EPOCHS + 1):\n", + " model.train()\n", + " epoch_rows = []\n", + "\n", + " for _ in range(STEPS_PER_EPOCH):\n", + " condition = rng.choice(train_sampling_conditions)\n", + " batch = make_condition_batch(condition=condition, split_name=TRAIN_SPLIT, rng=rng)\n", + "\n", + " optimizer.zero_grad(set_to_none=True)\n", + " losses = compute_loss_terms(batch)\n", + " losses[\"total\"].backward()\n", + " nn.utils.clip_grad_norm_(model.parameters(), max_norm=GRAD_CLIP_NORM)\n", + " optimizer.step()\n", + "\n", + " epoch_rows.append({key: float(value.item()) for key, value in losses.items()})\n", + "\n", + " if lr_scheduler is not None and LR_SCHEDULER == \"onecycle\":\n", + " lr_scheduler.step()\n", + "\n", + " train_summary = pd.DataFrame(epoch_rows).mean().to_dict()\n", + " val_seed = VAL_BATCH_SEED if FIXED_VAL_BATCH_SEED else (RANDOM_SEED + epoch)\n", + " val_sub = VAL_EVAL_SUBSET if (FIXED_VAL_SUBSET and VAL_EVAL_SUBSET) else None\n", + " val_summary = evaluate_split_loss(\n", + " split_name=VAL_SPLIT,\n", + " conditions=val_conditions,\n", + " seed=RANDOM_SEED + epoch,\n", + " steps=VALIDATION_STEPS,\n", + " fixed_condition_subset=val_sub,\n", + " batch_rng_seed=val_seed,\n", + " ) if val_conditions else {}\n", + "\n", + " row = {\"epoch\": epoch, **{f\"train_{k}\": v for k, v in train_summary.items()}}\n", + " row.update({f\"val_{k}\": v for k, v in val_summary.items()})\n", + " history_rows.append(row)\n", + "\n", + " current_val_total = val_summary.get(\"total\", train_summary[\"total\"])\n", + " if current_val_total < best_val_total:\n", + " best_val_total = current_val_total\n", + " best_state = deepcopy(model.state_dict())\n", + " epochs_no_improve = 0\n", + " else:\n", + " epochs_no_improve += 1\n", + "\n", + " if lr_scheduler is not None and LR_SCHEDULER == \"plateau\":\n", + " lr_scheduler.step(current_val_total)\n", + " elif lr_scheduler is not None and LR_SCHEDULER == \"cosine\":\n", + " lr_scheduler.step()\n", + "\n", + " display_row = {\n", + " \"epoch\": epoch,\n", + " \"train_total\": round(train_summary[\"total\"], 4),\n", + " \"train_flow\": round(train_summary[\"flow\"], 4),\n", + " \"train_recon\": round(train_summary[\"recon\"], 4),\n", + " \"val_total\": round(current_val_total, 4),\n", + " \"lr\": round(optimizer.param_groups[0][\"lr\"], 8),\n", + " }\n", + " print(display_row)\n", + "\n", + " if EARLY_STOP_PATIENCE and epochs_no_improve >= EARLY_STOP_PATIENCE:\n", + " print(f\"Early stopping at epoch {epoch} (no val improvement for {EARLY_STOP_PATIENCE} epochs).\")\n", + " break\n", + "\n", + "if best_state is not None:\n", + " model.load_state_dict(best_state)\n", + "\n", + "history = pd.DataFrame(history_rows)\n", + "history\n" + ] + }, + { + "cell_type": "markdown", + "id": "fc2eba64", + "metadata": {}, + "source": [ + "## Figures — training dynamics\n", + "\n", + "Loss curves (total / validation) and training component losses. Files: `OUTPUT_DIR / \"figures\"` as PNG and PDF.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 8, + "id": "a01ce791", + "metadata": {}, + "outputs": [ { - "cell_type": "markdown", - "id": "1296b782", - "metadata": {}, - "source": [ - "## Load And Prepare Data\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "_fig_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "plot_training_history(history, tag=_fig_tag)\n" + ] + }, + { + "cell_type": "markdown", + "id": "01fea9eb", + "metadata": {}, + "source": [ + "## Predict the evaluation split and save outputs\n", + "\n", + "Run inference on the held-out split, compute per-gene metrics, and persist `pred` / `real` AnnData plus CSV summaries.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "de3913d9", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Aggregate summary:\n", + "method_name conditional_flow_matching_method\n", + "input_path /root/final/data/processed/scPerturb/predictio...\n", + "eval_split test\n", + "train_split train\n", + "val_split val\n", + "n_control_cells_train 8653\n", + "n_eval_cells 29965\n", + "n_train_genes 2056\n", + "n_eval_genes 2056\n", + "n_eval_genes_missing_from_train 0\n", + "mse_mean_avg 0.019112\n", + "mae_mean_avg 0.097725\n", + "pearson_mean_avg 0.975635\n", + "pearson_delta_avg 0.111586\n", + "delta_l2_avg 8.491366\n", + "dtype: object\n", + "\n", + "Saved:\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/conditional_flow_matching_model.pt\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/pred_conditional_flow_matching_method_seen_cell_split_test.h5ad\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/real_conditional_flow_matching_method_seen_cell_split_test.h5ad\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/metrics_by_gene_conditional_flow_matching_method_seen_cell_split_test.csv\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/aggregate_conditional_flow_matching_method_seen_cell_split_test.csv\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/summary_conditional_flow_matching_method_seen_cell_split_test.csv\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/training_history_conditional_flow_matching_method_seen_cell_split_test.csv\n", + "- /root/final/baseline/outputs/conditional_flow_matching_method/config_conditional_flow_matching_method_seen_cell_split_test.json\n" + ] }, { - "cell_type": "code", - "execution_count": 24, - "id": "b1ead7e2", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Using feature mask from var['is_feature_gene']: 4334 genes\n", - "adata: AnnData object with n_obs × n_vars = 310385 × 4334\n", - " obs: 'batch', 'gene', 'gene_id', 'transcript', 'gene_transcript', 'guide_id', 'percent_mito', 'UMI_count', 'z_gemgroup_UMI', 'core_scale_factor', 'core_adjusted_UMI_count', 'disease', 'cancer', 'cell_line', 'sex', 'age', 'perturbation', 'organism', 'perturbation_type', 'tissue_type', 'ncounts', 'ngenes', 'nperts', 'percent_ribo', 'perturbation_label', 'perturbation_gene', 'is_control', 'is_single_perturbation', 'split', 'condition', 'batch_model', 'source_split'\n", - " var: 'chr', 'start', 'end', 'class', 'strand', 'length', 'in_matrix', 'mean', 'std', 'cv', 'fano', 'ensembl_id', 'ncounts', 'ncells', 'highly_variable', 'highly_variable_rank', 'means', 'variances', 'variances_norm', 'is_target_gene', 'is_feature_gene'\n", - " uns: 'hvg', 'log1p', 'prediction_ready'\n", - " layers: 'counts'\n", - "split counts:\n", - "split\n", - "train 248311\n", - "test 31046\n", - "val 31028\n", - "Name: count, dtype: int64\n", - "train conditions: 2056\n", - "val conditions: 2056\n", - "eval conditions: 2056\n", - "eval genes missing from training: 0\n" - ] - } + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
perturbation_genen_cellsmse_meanmae_meanpearson_meanpearson_deltadelta_l2
0AAAS210.0071280.0640980.990898-0.0841285.558319
1AAMP100.0187540.1031770.9748460.3199239.015438
2AARS40.0434230.1586150.9451730.38464413.718369
3AARS2180.0077620.0678040.989657-0.0318095.800062
4AASDHPPT50.0299580.1336160.9644450.05261111.394714
\n", + "
" ], - "source": [ - "adata = ad.read_h5ad(INPUT_PATH)\n", - "require_columns(adata.obs, [\"perturbation_gene\", \"is_control\", \"split\"])\n", - "\n", - "adata.obs[\"is_control\"] = normalize_bool(adata.obs[\"is_control\"])\n", - "adata.obs[\"perturbation_gene\"] = adata.obs[\"perturbation_gene\"].astype(str).str.strip()\n", - "adata.obs[\"perturbation_gene\"] = adata.obs[\"perturbation_gene\"].mask(adata.obs[\"is_control\"], \"__ctrl__\")\n", - "adata.obs[\"condition\"] = adata.obs[\"perturbation_gene\"].astype(str)\n", - "adata.obs[\"batch_model\"] = adata.obs[\"batch\"].astype(str) if \"batch\" in adata.obs.columns else \"global\"\n", - "\n", - "feature_mask = choose_feature_mask(adata.var)\n", - "adata = adata[:, feature_mask].copy()\n", - "\n", - "if SPLIT_STRATEGY == \"seen_cell_split\":\n", - " adata.obs[\"source_split\"] = adata.obs[\"split\"].astype(str)\n", - " adata.obs[\"split\"] = assign_seen_cell_split(adata.obs, RANDOM_SEED)\n", - "elif SPLIT_STRATEGY == \"use_existing\":\n", - " adata.obs[\"split\"] = pd.Categorical(\n", - " adata.obs[\"split\"].astype(str),\n", - " categories=[\"train\", \"val\", \"test\"],\n", - " ordered=True,\n", - " )\n", - "else:\n", - " raise ValueError(f\"Unknown SPLIT_STRATEGY: {SPLIT_STRATEGY}\")\n", - "\n", - "is_control = adata.obs[\"is_control\"].to_numpy(dtype=bool)\n", - "split = adata.obs[\"split\"].astype(str).to_numpy()\n", - "genes = adata.obs[\"perturbation_gene\"].astype(str).to_numpy()\n", - "batches = adata.obs[\"batch_model\"].astype(str).to_numpy()\n", - "\n", - "control_train_idx = np.where(is_control & (split == CONTROL_SPLIT))[0]\n", - "if len(control_train_idx) == 0:\n", - " print(f\"No controls found in CONTROL_SPLIT={CONTROL_SPLIT}; falling back to all controls.\")\n", - " control_train_idx = np.where(is_control)[0]\n", - "if len(control_train_idx) == 0:\n", - " raise ValueError(\"No control cells found.\")\n", - "\n", - "train_conditions = sorted(set(genes[(~is_control) & (split == TRAIN_SPLIT)]))\n", - "val_conditions = sorted(set(genes[(~is_control) & (split == VAL_SPLIT)]))\n", - "eval_conditions = sorted(set(genes[(~is_control) & (split == EVAL_SPLIT)]))\n", - "\n", - "if MAX_TRAIN_CONDITIONS is not None:\n", - " train_conditions = train_conditions[:MAX_TRAIN_CONDITIONS]\n", - " train_condition_set = set(train_conditions)\n", - " keep_mask = is_control | np.isin(genes, list(train_condition_set)) | np.isin(genes, eval_conditions) | np.isin(genes, val_conditions)\n", - " adata = adata[keep_mask].copy()\n", - " is_control = adata.obs[\"is_control\"].to_numpy(dtype=bool)\n", - " split = adata.obs[\"split\"].astype(str).to_numpy()\n", - " genes = adata.obs[\"perturbation_gene\"].astype(str).to_numpy()\n", - " batches = adata.obs[\"batch_model\"].astype(str).to_numpy()\n", - " control_train_idx = np.where(is_control & (split == CONTROL_SPLIT))[0]\n", - " train_conditions = sorted(set(genes[(~is_control) & (split == TRAIN_SPLIT)]))\n", - " val_conditions = sorted(set(genes[(~is_control) & (split == VAL_SPLIT)]))\n", - " eval_conditions = sorted(set(genes[(~is_control) & (split == EVAL_SPLIT)]))\n", - "\n", - "if MAX_EVAL_GENES is not None:\n", - " eval_conditions = eval_conditions[:MAX_EVAL_GENES]\n", - "\n", - "indices_by_split_gene = {}\n", - "for split_name in [TRAIN_SPLIT, VAL_SPLIT, EVAL_SPLIT]:\n", - " indices_by_split_gene[split_name] = {}\n", - " split_mask = (~is_control) & (split == split_name)\n", - " for gene in sorted(set(genes[split_mask])):\n", - " indices = np.where(split_mask & (genes == gene))[0]\n", - " if len(indices) > 0:\n", - " indices_by_split_gene[split_name][gene] = indices\n", - "\n", - "control_idx_by_batch = {}\n", - "for batch_name in sorted(set(batches[control_train_idx])):\n", - " batch_idx = control_train_idx[batches[control_train_idx] == batch_name]\n", - " if len(batch_idx) > 0:\n", - " control_idx_by_batch[batch_name] = batch_idx\n", - "\n", - "gene_vocab = [\"__ctrl__\", \"__unk__\", *train_conditions]\n", - "gene_to_idx = {gene: idx for idx, gene in enumerate(gene_vocab)}\n", - "batch_vocab = [\"__unk__\", *sorted(set(batches))]\n", - "batch_to_idx = {batch: idx for idx, batch in enumerate(batch_vocab)}\n", - "\n", - "gene_index_per_obs = np.array([gene_to_idx.get(gene, gene_to_idx[\"__unk__\"]) for gene in genes], dtype=np.int64)\n", - "batch_index_per_obs = np.array([batch_to_idx.get(batch, batch_to_idx[\"__unk__\"]) for batch in batches], dtype=np.int64)\n", - "\n", - "control_mean = mean_vector(adata.X[control_train_idx])\n", - "unseen_eval_genes = sorted(set(eval_conditions) - set(train_conditions))\n", - "\n", - "print(\"adata:\", adata)\n", - "print(\"split counts:\")\n", - "print(adata.obs[\"split\"].astype(str).value_counts())\n", - "print(\"train conditions:\", len(train_conditions))\n", - "print(\"val conditions:\", len(val_conditions))\n", - "print(\"eval conditions:\", len(eval_conditions))\n", - "print(\"eval genes missing from training:\", len(unseen_eval_genes))\n", - "if unseen_eval_genes[:10]:\n", - " print(\"first unseen eval genes:\", unseen_eval_genes[:10])\n" + "text/plain": [ + " perturbation_gene n_cells mse_mean mae_mean pearson_mean pearson_delta \\\n", + "0 AAAS 21 0.007128 0.064098 0.990898 -0.084128 \n", + "1 AAMP 10 0.018754 0.103177 0.974846 0.319923 \n", + "2 AARS 4 0.043423 0.158615 0.945173 0.384644 \n", + "3 AARS2 18 0.007762 0.067804 0.989657 -0.031809 \n", + "4 AASDHPPT 5 0.029958 0.133616 0.964445 0.052611 \n", + "\n", + " delta_l2 \n", + "0 5.558319 \n", + "1 9.015438 \n", + "2 13.718369 \n", + "3 5.800062 \n", + "4 11.394714 " ] - }, + }, + "execution_count": 9, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "eval_conditions_to_run = [condition for condition in eval_conditions if condition in indices_by_split_gene[EVAL_SPLIT]]\n", + "if not eval_conditions_to_run:\n", + " raise ValueError(f\"No conditions found for EVAL_SPLIT={EVAL_SPLIT}.\")\n", + "\n", + "metrics, pred, real = evaluate_gene_metrics(\n", + " split_name=EVAL_SPLIT,\n", + " conditions=eval_conditions_to_run,\n", + " seed=RANDOM_SEED + 999,\n", + ")\n", + "aggregate = build_aggregate_frame(metrics)\n", + "aggregate_summary = pd.Series(\n", + " {\n", + " \"method_name\": METHOD_NAME,\n", + " \"input_path\": str(INPUT_PATH),\n", + " \"eval_split\": EVAL_SPLIT,\n", + " \"train_split\": TRAIN_SPLIT,\n", + " \"val_split\": VAL_SPLIT,\n", + " \"n_control_cells_train\": int(len(control_train_idx)),\n", + " \"n_eval_cells\": int(pred.n_obs),\n", + " \"n_train_genes\": int(len(train_conditions)),\n", + " \"n_eval_genes\": int(len(eval_conditions_to_run)),\n", + " \"n_eval_genes_missing_from_train\": int(len(set(eval_conditions_to_run) - set(train_conditions))),\n", + " \"mse_mean_avg\": float(metrics[\"mse_mean\"].mean()),\n", + " \"mae_mean_avg\": float(metrics[\"mae_mean\"].mean()),\n", + " \"pearson_mean_avg\": float(metrics[\"pearson_mean\"].mean(skipna=True)),\n", + " \"pearson_delta_avg\": float(metrics[\"pearson_delta\"].mean(skipna=True)),\n", + " \"delta_l2_avg\": float(metrics[\"delta_l2\"].mean()),\n", + " }\n", + ")\n", + "\n", + "config = {\n", + " \"name\": METHOD_NAME,\n", + " \"input_path\": str(INPUT_PATH),\n", + " \"split_strategy\": SPLIT_STRATEGY,\n", + " \"train_frac\": float(TRAIN_FRAC),\n", + " \"val_frac\": float(VAL_FRAC),\n", + " \"eval_split\": EVAL_SPLIT,\n", + " \"random_seed\": int(RANDOM_SEED),\n", + " \"latent_dim\": int(LATENT_DIM),\n", + " \"hidden_dim\": int(HIDDEN_DIM),\n", + " \"condition_dim\": int(CONDITION_DIM),\n", + " \"time_dim\": int(TIME_DIM),\n", + " \"dropout\": float(DROPOUT),\n", + " \"num_flow_blocks\": int(NUM_FLOW_BLOCKS),\n", + " \"batch_size\": int(BATCH_SIZE),\n", + " \"epochs\": int(EPOCHS),\n", + " \"steps_per_epoch\": int(STEPS_PER_EPOCH),\n", + " \"validation_steps\": int(VALIDATION_STEPS),\n", + " \"learning_rate\": float(LEARNING_RATE),\n", + " \"lr_scheduler\": LR_SCHEDULER,\n", + " \"plateau_factor\": float(PLATEAU_FACTOR),\n", + " \"plateau_patience\": int(PLATEAU_PATIENCE),\n", + " \"early_stop_patience\": int(EARLY_STOP_PATIENCE),\n", + " \"fixed_val_subset\": FIXED_VAL_SUBSET,\n", + " \"flow_lr_mult\": float(FLOW_LR_MULT),\n", + " \"uncertainty_samples\": int(UNCERTAINTY_SAMPLES),\n", + " \"weight_decay\": float(WEIGHT_DECAY),\n", + " \"recon_weight\": float(RECON_WEIGHT),\n", + " \"flow_weight\": float(FLOW_WEIGHT),\n", + " \"endpoint_weight\": float(ENDPOINT_WEIGHT),\n", + " \"mmd_weight\": float(MMD_WEIGHT),\n", + " \"mean_weight\": float(MEAN_WEIGHT),\n", + " \"delta_mse_weight\": float(DELTA_MSE_WEIGHT),\n", + " \"delta_cos_weight\": float(DELTA_COS_WEIGHT),\n", + " \"delta_mmd_weight\": float(DELTA_MMD_WEIGHT),\n", + " \"delta_bulk_weight\": float(DELTA_BULK_WEIGHT),\n", + " \"ot_decode_mmd_weight\": float(OT_DECODE_MMD_WEIGHT),\n", + " \"ot_decode_mean_weight\": float(OT_DECODE_MEAN_WEIGHT),\n", + " \"displacement_reg_weight\": float(DISPLACEMENT_REG_WEIGHT),\n", + " \"velocity_l2_weight\": float(VELOCITY_L2_WEIGHT),\n", + " \"noise_std\": float(NOISE_STD),\n", + " \"sinkhorn_epsilon\": float(SINKHORN_EPSILON),\n", + " \"sinkhorn_iters\": int(SINKHORN_ITERS),\n", + " \"inference_steps\": int(INFERENCE_STEPS),\n", + " \"n_train_conditions\": int(len(train_conditions)),\n", + " \"n_eval_conditions\": int(len(eval_conditions_to_run)),\n", + " \"n_unseen_eval_conditions\": int(len(set(eval_conditions_to_run) - set(train_conditions))),\n", + " \"feature_gene_columns\": FEATURE_GENE_COLUMNS,\n", + " \"model_path\": str(MODEL_PATH),\n", + "}\n", + "\n", + "pred.uns[\"baseline\"] = config\n", + "real.uns[\"baseline\"] = config\n", + "\n", + "tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "pred_path = OUTPUT_DIR / f\"pred_{METHOD_NAME}_{tag}.h5ad\"\n", + "real_path = OUTPUT_DIR / f\"real_{METHOD_NAME}_{tag}.h5ad\"\n", + "metrics_path = OUTPUT_DIR / f\"metrics_by_gene_{METHOD_NAME}_{tag}.csv\"\n", + "aggregate_path = OUTPUT_DIR / f\"aggregate_{METHOD_NAME}_{tag}.csv\"\n", + "summary_path = OUTPUT_DIR / f\"summary_{METHOD_NAME}_{tag}.csv\"\n", + "history_path = OUTPUT_DIR / f\"training_history_{METHOD_NAME}_{tag}.csv\"\n", + "config_path = OUTPUT_DIR / f\"config_{METHOD_NAME}_{tag}.json\"\n", + "\n", + "torch.save(model.state_dict(), MODEL_PATH)\n", + "pred.write_h5ad(pred_path, compression=\"gzip\")\n", + "real.write_h5ad(real_path, compression=\"gzip\")\n", + "metrics.to_csv(metrics_path, index=False)\n", + "aggregate.to_csv(aggregate_path)\n", + "aggregate_summary.to_frame(name=\"value\").to_csv(summary_path)\n", + "history.to_csv(history_path, index=False)\n", + "with config_path.open(\"w\") as f:\n", + " json.dump(config, f, indent=2)\n", + "\n", + "print(\"Aggregate summary:\")\n", + "print(aggregate_summary)\n", + "print(\"\\nSaved:\")\n", + "for path in [MODEL_PATH, pred_path, real_path, metrics_path, aggregate_path, summary_path, history_path, config_path]:\n", + " print(\"-\", path)\n", + "\n", + "metrics.head()\n" + ] + }, + { + "cell_type": "markdown", + "id": "viz-eval-metrics-md", + "metadata": {}, + "source": [ + "## Figures — held-out evaluation\n", + "\n", + "Per-perturbation metrics (MSE vs Pearson, ranked Pearson, delta L2, mean vs delta correlation). Seen vs unseen training genes are colored when both cohorts exist.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "97a06a5d", + "metadata": {}, + "outputs": [ { - "cell_type": "markdown", - "id": "3fa220a2", - "metadata": {}, - "source": [ - "## Model Definition\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] - }, + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "_fig_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "plot_evaluation_metrics(metrics, tag=_fig_tag, train_genes=train_conditions)\n" + ] + }, + { + "cell_type": "markdown", + "id": "bio-extended-md", + "metadata": {}, + "source": [ + "## Extended biological analysis figures\n", + "\n", + "Additional plots for manuscript-style interpretation: gene-level calibration, MA-style view, perturbation–gene heatmaps, perturbation similarity, cell-space PCA, and error distributions. Requires upstream cells that define `pred`, `real`, `adata`, `metrics`, `control_mean`, and `eval_conditions_to_run`. Figures append to `FIGURE_DIR` with the `bio_` prefix.\n" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "bio-extended-code", + "metadata": {}, + "outputs": [ { - "cell_type": "code", - "execution_count": 25, - "id": "ec5b92e5", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "ConditionalFlowMatchingModel(\n", - " (encoder): MLPEncoder(\n", - " (net): Sequential(\n", - " (0): Linear(in_features=4334, out_features=1024, bias=True)\n", - " (1): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", - " (2): GELU(approximate='none')\n", - " (3): Dropout(p=0.08, inplace=False)\n", - " (4): Linear(in_features=1024, out_features=1024, bias=True)\n", - " (5): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", - " (6): GELU(approximate='none')\n", - " (7): Dropout(p=0.08, inplace=False)\n", - " (8): Linear(in_features=1024, out_features=512, bias=True)\n", - " (9): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", - " (10): GELU(approximate='none')\n", - " (11): Dropout(p=0.08, inplace=False)\n", - " (12): Linear(in_features=512, out_features=128, bias=True)\n", - " )\n", - " )\n", - " (decoder): MLPDecoder(\n", - " (net): Sequential(\n", - " (0): Linear(in_features=128, out_features=512, bias=True)\n", - " (1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", - " (2): GELU(approximate='none')\n", - " (3): Dropout(p=0.08, inplace=False)\n", - " (4): Linear(in_features=512, out_features=1024, bias=True)\n", - " (5): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", - " (6): GELU(approximate='none')\n", - " (7): Dropout(p=0.08, inplace=False)\n", - " (8): Linear(in_features=1024, out_features=1024, bias=True)\n", - " (9): LayerNorm((1024,), eps=1e-05, elementwise_affine=True)\n", - " (10): GELU(approximate='none')\n", - " (11): Dropout(p=0.08, inplace=False)\n", - " (12): Linear(in_features=1024, out_features=4334, bias=True)\n", - " )\n", - " )\n", - " (condition_encoder): ConditionEncoder(\n", - " (gene_embedding): Embedding(2058, 128)\n", - " (batch_embedding): Embedding(49, 64)\n", - " (mlp): Sequential(\n", - " (0): Linear(in_features=192, out_features=256, bias=True)\n", - " (1): LayerNorm((256,), eps=1e-05, elementwise_affine=True)\n", - " (2): GELU(approximate='none')\n", - " (3): Dropout(p=0.08, inplace=False)\n", - " (4): Linear(in_features=256, out_features=256, bias=True)\n", - " )\n", - " )\n", - " (time_encoder): TimeEmbedding(\n", - " (proj): Sequential(\n", - " (0): Linear(in_features=128, out_features=128, bias=True)\n", - " (1): SiLU()\n", - " (2): Linear(in_features=128, out_features=128, bias=True)\n", - " )\n", - " )\n", - " (vector_field): ConditionalVectorField(\n", - " (input_proj): Linear(in_features=256, out_features=512, bias=True)\n", - " (blocks): ModuleList(\n", - " (0-5): 6 x FiLMResidualBlock(\n", - " (norm1): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", - " (norm2): LayerNorm((512,), eps=1e-05, elementwise_affine=True)\n", - " (fc1): Linear(in_features=512, out_features=512, bias=True)\n", - " (fc2): Linear(in_features=512, out_features=512, bias=True)\n", - " (film1): Linear(in_features=256, out_features=1024, bias=True)\n", - " (film2): Linear(in_features=256, out_features=1024, bias=True)\n", - " (dropout): Dropout(p=0.08, inplace=False)\n", - " )\n", - " )\n", - " (out): Linear(in_features=512, out_features=128, bias=True)\n", - " )\n", - ")\n", - "LR scheduler: ReduceLROnPlateau | VAL_EVAL_SUBSET: 48 | flow_lr_mult: 1.5\n" - ] - } - ], - "source": [ - "class MLPEncoder(nn.Module):\n", - " def __init__(self, input_dim: int, latent_dim: int, hidden_dim: int, dropout: float):\n", - " super().__init__()\n", - " h2 = hidden_dim * 2\n", - " self.net = nn.Sequential(\n", - " nn.Linear(input_dim, h2),\n", - " nn.LayerNorm(h2),\n", - " nn.GELU(),\n", - " nn.Dropout(dropout),\n", - " nn.Linear(h2, h2),\n", - " nn.LayerNorm(h2),\n", - " nn.GELU(),\n", - " nn.Dropout(dropout),\n", - " nn.Linear(h2, hidden_dim),\n", - " nn.LayerNorm(hidden_dim),\n", - " nn.GELU(),\n", - " nn.Dropout(dropout),\n", - " nn.Linear(hidden_dim, latent_dim),\n", - " )\n", - "\n", - " def forward(self, x: torch.Tensor) -> torch.Tensor:\n", - " return self.net(x)\n", - "\n", - "\n", - "class MLPDecoder(nn.Module):\n", - " def __init__(self, latent_dim: int, output_dim: int, hidden_dim: int, dropout: float):\n", - " super().__init__()\n", - " h2 = hidden_dim * 2\n", - " self.net = nn.Sequential(\n", - " nn.Linear(latent_dim, hidden_dim),\n", - " nn.LayerNorm(hidden_dim),\n", - " nn.GELU(),\n", - " nn.Dropout(dropout),\n", - " nn.Linear(hidden_dim, h2),\n", - " nn.LayerNorm(h2),\n", - " nn.GELU(),\n", - " nn.Dropout(dropout),\n", - " nn.Linear(h2, h2),\n", - " nn.LayerNorm(h2),\n", - " nn.GELU(),\n", - " nn.Dropout(dropout),\n", - " nn.Linear(h2, output_dim),\n", - " )\n", - "\n", - " def forward(self, z: torch.Tensor) -> torch.Tensor:\n", - " return F.softplus(self.net(z))\n", - "\n", - "\n", - "# Condition h = MLP([gene_emb; batch_emb]) — \"基因 + batch\" only (no perturbation-type / cell-type).\n", - "class ConditionEncoder(nn.Module):\n", - " def __init__(self, n_genes: int, n_batches: int, condition_dim: int, dropout: float):\n", - " super().__init__()\n", - " gene_dim = max(32, condition_dim // 2)\n", - " batch_dim = max(16, condition_dim // 4)\n", - " self.gene_embedding = nn.Embedding(n_genes, gene_dim)\n", - " self.batch_embedding = nn.Embedding(n_batches, batch_dim)\n", - " self.mlp = nn.Sequential(\n", - " nn.Linear(gene_dim + batch_dim, condition_dim),\n", - " nn.LayerNorm(condition_dim),\n", - " nn.GELU(),\n", - " nn.Dropout(dropout),\n", - " nn.Linear(condition_dim, condition_dim),\n", - " )\n", - "\n", - " def forward(self, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", - " gene_emb = self.gene_embedding(gene_idx)\n", - " batch_emb = self.batch_embedding(batch_idx)\n", - " return self.mlp(torch.cat([gene_emb, batch_emb], dim=-1))\n", - "\n", - "\n", - "class TimeEmbedding(nn.Module):\n", - " def __init__(self, time_dim: int):\n", - " super().__init__()\n", - " self.time_dim = time_dim\n", - " self.proj = nn.Sequential(\n", - " nn.Linear(time_dim, time_dim),\n", - " nn.SiLU(),\n", - " nn.Linear(time_dim, time_dim),\n", - " )\n", - "\n", - " def forward(self, t: torch.Tensor) -> torch.Tensor:\n", - " half = self.time_dim // 2\n", - " freqs = torch.exp(\n", - " -math.log(10000.0) * torch.arange(half, device=t.device, dtype=t.dtype) / max(half, 1)\n", - " )\n", - " angles = t[:, None] * freqs[None, :]\n", - " emb = torch.cat([torch.cos(angles), torch.sin(angles)], dim=-1)\n", - " if emb.shape[1] < self.time_dim:\n", - " emb = F.pad(emb, (0, self.time_dim - emb.shape[1]))\n", - " return self.proj(emb)\n", - "\n", - "\n", - "class FiLMResidualBlock(nn.Module):\n", - " def __init__(self, hidden_dim: int, condition_dim: int, dropout: float):\n", - " super().__init__()\n", - " self.norm1 = nn.LayerNorm(hidden_dim)\n", - " self.norm2 = nn.LayerNorm(hidden_dim)\n", - " self.fc1 = nn.Linear(hidden_dim, hidden_dim)\n", - " self.fc2 = nn.Linear(hidden_dim, hidden_dim)\n", - " self.film1 = nn.Linear(condition_dim, hidden_dim * 2)\n", - " self.film2 = nn.Linear(condition_dim, hidden_dim * 2)\n", - " self.dropout = nn.Dropout(dropout)\n", - "\n", - " def forward(self, x: torch.Tensor, condition: torch.Tensor) -> torch.Tensor:\n", - " gamma1, beta1 = self.film1(condition).chunk(2, dim=-1)\n", - " h = self.norm1(x)\n", - " h = h * (1 + gamma1) + beta1\n", - " h = F.gelu(self.fc1(h))\n", - " h = self.dropout(h)\n", - "\n", - " gamma2, beta2 = self.film2(condition).chunk(2, dim=-1)\n", - " h = self.norm2(h)\n", - " h = h * (1 + gamma2) + beta2\n", - " h = self.fc2(h)\n", - " h = self.dropout(h)\n", - " return x + h\n", - "\n", - "\n", - "class ConditionalVectorField(nn.Module):\n", - " def __init__(self, latent_dim: int, time_dim: int, condition_dim: int, hidden_dim: int, num_blocks: int, dropout: float):\n", - " super().__init__()\n", - " self.input_proj = nn.Linear(latent_dim + time_dim, hidden_dim)\n", - " self.blocks = nn.ModuleList(\n", - " [FiLMResidualBlock(hidden_dim, condition_dim, dropout) for _ in range(num_blocks)]\n", - " )\n", - " self.out = nn.Linear(hidden_dim, latent_dim)\n", - "\n", - " def forward(self, z_t: torch.Tensor, t_emb: torch.Tensor, condition: torch.Tensor) -> torch.Tensor:\n", - " h = self.input_proj(torch.cat([z_t, t_emb], dim=-1))\n", - " for block in self.blocks:\n", - " h = block(h, condition)\n", - " return self.out(h)\n", - "\n", - "\n", - "class ConditionalFlowMatchingModel(nn.Module):\n", - " def __init__(self, input_dim: int, latent_dim: int, hidden_dim: int, condition_dim: int, time_dim: int, n_genes: int, n_batches: int, num_blocks: int, dropout: float):\n", - " super().__init__()\n", - " self.encoder = MLPEncoder(input_dim=input_dim, latent_dim=latent_dim, hidden_dim=hidden_dim, dropout=dropout)\n", - " self.decoder = MLPDecoder(latent_dim=latent_dim, output_dim=input_dim, hidden_dim=hidden_dim, dropout=dropout)\n", - " self.condition_encoder = ConditionEncoder(n_genes=n_genes, n_batches=n_batches, condition_dim=condition_dim, dropout=dropout)\n", - " self.time_encoder = TimeEmbedding(time_dim=time_dim)\n", - " self.vector_field = ConditionalVectorField(\n", - " latent_dim=latent_dim,\n", - " time_dim=time_dim,\n", - " condition_dim=condition_dim,\n", - " hidden_dim=hidden_dim,\n", - " num_blocks=num_blocks,\n", - " dropout=dropout,\n", - " )\n", - "\n", - " def encode_x(self, x: torch.Tensor) -> torch.Tensor:\n", - " return self.encoder(x)\n", - "\n", - " def decode_z(self, z: torch.Tensor) -> torch.Tensor:\n", - " return self.decoder(z)\n", - "\n", - " def reconstruct(self, x: torch.Tensor) -> torch.Tensor:\n", - " return self.decode_z(self.encode_x(x))\n", - "\n", - " def condition(self, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", - " return self.condition_encoder(gene_idx, batch_idx)\n", - "\n", - " def velocity(self, z_t: torch.Tensor, t: torch.Tensor, gene_idx: torch.Tensor, batch_idx: torch.Tensor) -> torch.Tensor:\n", - " condition = self.condition(gene_idx, batch_idx)\n", - " t_emb = self.time_encoder(t)\n", - " return self.vector_field(z_t, t_emb, condition)\n", - "\n", - "\n", - "model = ConditionalFlowMatchingModel(\n", - " input_dim=adata.n_vars,\n", - " latent_dim=LATENT_DIM,\n", - " hidden_dim=HIDDEN_DIM,\n", - " condition_dim=CONDITION_DIM,\n", - " time_dim=TIME_DIM,\n", - " n_genes=len(gene_vocab),\n", - " n_batches=len(batch_vocab),\n", - " num_blocks=NUM_FLOW_BLOCKS,\n", - " dropout=DROPOUT,\n", - ").to(DEVICE)\n", - "\n", - "ae_params = list(model.encoder.parameters()) + list(model.decoder.parameters())\n", - "flow_params = (\n", - " list(model.vector_field.parameters())\n", - " + list(model.time_encoder.parameters())\n", - " + list(model.condition_encoder.parameters())\n", - ")\n", - "optimizer = torch.optim.AdamW(\n", - " [\n", - " {\"params\": ae_params, \"lr\": LEARNING_RATE},\n", - " {\"params\": flow_params, \"lr\": LEARNING_RATE * FLOW_LR_MULT},\n", - " ],\n", - " weight_decay=WEIGHT_DECAY,\n", - ")\n", - "\n", - "if val_conditions:\n", - " _n = min(len(val_conditions), VALIDATION_STEPS)\n", - " VAL_EVAL_SUBSET = sorted(val_conditions)[:_n]\n", - "else:\n", - " VAL_EVAL_SUBSET = []\n", - "\n", - "lr_scheduler = None\n", - "if LR_SCHEDULER == \"plateau\":\n", - " lr_scheduler = torch.optim.lr_scheduler.ReduceLROnPlateau(\n", - " optimizer, mode=\"min\", factor=PLATEAU_FACTOR, patience=PLATEAU_PATIENCE, min_lr=PLATEAU_MIN_LR\n", - " )\n", - "elif LR_SCHEDULER == \"cosine\":\n", - " lr_scheduler = torch.optim.lr_scheduler.CosineAnnealingLR(optimizer, T_max=EPOCHS, eta_min=COSINE_MIN_LR)\n", - "elif LR_SCHEDULER == \"onecycle\":\n", - " total_steps = EPOCHS * STEPS_PER_EPOCH\n", - " lr_scheduler = torch.optim.lr_scheduler.OneCycleLR(\n", - " optimizer,\n", - " max_lr=[LEARNING_RATE, LEARNING_RATE * FLOW_LR_MULT],\n", - " total_steps=total_steps,\n", - " pct_start=0.1,\n", - " )\n", - "\n", - "print(model)\n", - "print(\n", - " \"LR scheduler:\",\n", - " type(lr_scheduler).__name__ if lr_scheduler is not None else \"none\",\n", - " \"| VAL_EVAL_SUBSET:\",\n", - " len(VAL_EVAL_SUBSET),\n", - " \"| flow_lr_mult:\",\n", - " FLOW_LR_MULT,\n", - ")\n" - ] + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.\u001b[0m\u001b[33m\n", + "\u001b[0m" + ] }, { - "cell_type": "code", - "execution_count": 26, - "id": "91749254", - "metadata": {}, - "outputs": [], - "source": [ - "def sample_controls_like_targets(target_idx: np.ndarray, rng: np.random.Generator) -> np.ndarray:\n", - " sampled = []\n", - " for idx in target_idx:\n", - " batch_name = batches[idx]\n", - " pool = control_idx_by_batch.get(batch_name, control_train_idx)\n", - " sampled.append(rng.choice(pool))\n", - " return np.asarray(sampled, dtype=np.int64)\n", - "\n", - "\n", - "def make_condition_batch(condition: str, split_name: str, rng: np.random.Generator) -> dict:\n", - " target_pool = indices_by_split_gene[split_name][condition]\n", - " size = min(BATCH_SIZE, len(target_pool))\n", - " replace = len(target_pool) < size\n", - " target_idx = rng.choice(target_pool, size=size, replace=replace)\n", - " source_idx = sample_controls_like_targets(target_idx, rng)\n", - "\n", - " if MAX_EVAL_CELLS_PER_GENE is not None and len(target_idx) > MAX_EVAL_CELLS_PER_GENE:\n", - " target_idx = target_idx[:MAX_EVAL_CELLS_PER_GENE]\n", - " source_idx = source_idx[:MAX_EVAL_CELLS_PER_GENE]\n", - "\n", - " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", - " target_x = torch.from_numpy(take_dense_rows(adata.X, target_idx)).to(DEVICE)\n", - " gene_idx = torch.full(\n", - " (len(target_idx),),\n", - " fill_value=gene_to_idx.get(condition, gene_to_idx[\"__unk__\"]),\n", - " device=DEVICE,\n", - " dtype=torch.long,\n", - " )\n", - " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", - "\n", - " return {\n", - " \"condition\": condition,\n", - " \"source_idx\": source_idx,\n", - " \"target_idx\": target_idx,\n", - " \"source_x\": source_x,\n", - " \"target_x\": target_x,\n", - " \"gene_idx\": gene_idx,\n", - " \"batch_idx\": batch_idx,\n", - " }\n", - "\n", - "\n", - "def compute_loss_terms(batch: dict) -> dict[str, torch.Tensor]:\n", - " source_x = batch[\"source_x\"]\n", - " target_x = batch[\"target_x\"]\n", - " gene_idx = batch[\"gene_idx\"]\n", - " batch_idx = batch[\"batch_idx\"]\n", - "\n", - " z0 = model.encode_x(source_x)\n", - " z1 = model.encode_x(target_x)\n", - " z1_bar = sinkhorn_barycentric_targets(\n", - " x0=z0,\n", - " x1=z1,\n", - " epsilon=SINKHORN_EPSILON,\n", - " n_iters=SINKHORN_ITERS,\n", - " )\n", - "\n", - " t = torch.rand(z0.shape[0], device=DEVICE)\n", - " sigma = NOISE_STD * torch.sqrt((t * (1.0 - t)).clamp_min(1e-6)).unsqueeze(-1)\n", - " z_t = (1.0 - t).unsqueeze(-1) * z0 + t.unsqueeze(-1) * z1_bar + sigma * torch.randn_like(z0)\n", - "\n", - " target_velocity = z1_bar - z0\n", - " pred_velocity = model.velocity(z_t, t, gene_idx, batch_idx)\n", - " pred_z1 = z_t + (1.0 - t).unsqueeze(-1) * pred_velocity\n", - " pred_x1 = model.decode_z(pred_z1)\n", - "\n", - " source_recon = model.reconstruct(source_x)\n", - " target_recon = model.reconstruct(target_x)\n", - "\n", - " recon_loss = F.mse_loss(source_recon, source_x) + F.mse_loss(target_recon, target_x)\n", - " flow_loss = F.mse_loss(pred_velocity, target_velocity)\n", - " endpoint_loss = F.mse_loss(pred_z1, z1_bar)\n", - " mmd_loss = rbf_mmd_loss(pred_x1, target_x)\n", - " mean_loss = F.mse_loss(pred_x1.mean(dim=0), target_x.mean(dim=0))\n", - "\n", - " total_loss = (\n", - " RECON_WEIGHT * recon_loss\n", - " + FLOW_WEIGHT * flow_loss\n", - " + ENDPOINT_WEIGHT * endpoint_loss\n", - " + MMD_WEIGHT * mmd_loss\n", - " + MEAN_WEIGHT * mean_loss\n", - " )\n", - "\n", - " return {\n", - " \"total\": total_loss,\n", - " \"recon\": recon_loss.detach(),\n", - " \"flow\": flow_loss.detach(),\n", - " \"endpoint\": endpoint_loss.detach(),\n", - " \"mmd\": mmd_loss.detach(),\n", - " \"mean\": mean_loss.detach(),\n", - " }\n", - "\n", - "\n", - "@torch.no_grad()\n", - "def evaluate_split_loss(\n", - " split_name: str,\n", - " conditions: list[str],\n", - " seed: int,\n", - " steps: int,\n", - " fixed_condition_subset: list[str] | None = None,\n", - " batch_rng_seed: int | None = None,\n", - ") -> dict[str, float]:\n", - " if fixed_condition_subset is not None:\n", - " if len(fixed_condition_subset) == 0:\n", - " return {}\n", - " elif not conditions:\n", - " return {}\n", - "\n", - " model.eval()\n", - " if fixed_condition_subset is not None:\n", - " subset = fixed_condition_subset[: min(len(fixed_condition_subset), steps)]\n", - " else:\n", - " rng = np.random.default_rng(seed)\n", - " subset = conditions.copy()\n", - " rng.shuffle(subset)\n", - " subset = subset[: min(len(subset), steps)]\n", - "\n", - " brng = batch_rng_seed if batch_rng_seed is not None else seed\n", - " rng = np.random.default_rng(brng)\n", - "\n", - " rows = []\n", - " for condition in subset:\n", - " batch = make_condition_batch(condition, split_name=split_name, rng=rng)\n", - " losses = compute_loss_terms(batch)\n", - " rows.append({key: float(value.item()) for key, value in losses.items()})\n", - "\n", - " return pd.DataFrame(rows).mean().to_dict()\n", - "\n", - "\n", - "@torch.no_grad()\n", - "def predict_gene_expression(\n", - " condition: str,\n", - " split_name: str,\n", - " rng: np.random.Generator,\n", - " n_stochastic: int = 1,\n", - ") -> tuple[np.ndarray, np.ndarray, pd.DataFrame, np.ndarray]:\n", - " target_idx_all = indices_by_split_gene[split_name][condition]\n", - " if MAX_EVAL_CELLS_PER_GENE is not None and len(target_idx_all) > MAX_EVAL_CELLS_PER_GENE:\n", - " target_idx_all = target_idx_all[:MAX_EVAL_CELLS_PER_GENE]\n", - "\n", - " pred_blocks = []\n", - " pred_std_blocks = []\n", - " real_blocks = []\n", - " obs_blocks = []\n", - " gene_id = gene_to_idx.get(condition, gene_to_idx[\"__unk__\"])\n", - " condition_source = \"trained_gene_embedding\" if condition in gene_to_idx else \"unk_gene_embedding\"\n", - "\n", - " for start in range(0, len(target_idx_all), PRED_BATCH_SIZE):\n", - " target_idx = target_idx_all[start : start + PRED_BATCH_SIZE]\n", - " if n_stochastic <= 1:\n", - " source_idx = sample_controls_like_targets(target_idx, rng)\n", - " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", - " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", - " gene_idx = torch.full((len(target_idx),), gene_id, device=DEVICE, dtype=torch.long)\n", - " z0 = model.encode_x(source_x)\n", - " z1_hat = rk2_integrate_latent(model, z0, gene_idx, batch_idx, steps=INFERENCE_STEPS)\n", - " pred_x = model.decode_z(z1_hat).cpu().numpy().astype(np.float32)\n", - " pred_std_blocks.append(np.zeros_like(pred_x, dtype=np.float32))\n", - " else:\n", - " runs = []\n", - " for s in range(n_stochastic):\n", - " sub_rng = np.random.default_rng(int(rng.integers(0, 2**31 - 1)) + s * 1000003)\n", - " source_idx = sample_controls_like_targets(target_idx, sub_rng)\n", - " source_x = torch.from_numpy(take_dense_rows(adata.X, source_idx)).to(DEVICE)\n", - " batch_idx = torch.as_tensor(batch_index_per_obs[source_idx], device=DEVICE, dtype=torch.long)\n", - " gene_idx = torch.full((len(target_idx),), gene_id, device=DEVICE, dtype=torch.long)\n", - " z0 = model.encode_x(source_x)\n", - " z1_hat = rk2_integrate_latent(model, z0, gene_idx, batch_idx, steps=INFERENCE_STEPS)\n", - " runs.append(model.decode_z(z1_hat).cpu().numpy().astype(np.float32))\n", - " stacked = np.stack(runs, axis=0)\n", - " pred_x = stacked.mean(axis=0)\n", - " pred_std_blocks.append(stacked.std(axis=0))\n", - "\n", - " real_x = take_dense_rows(adata.X, target_idx)\n", - " pred_blocks.append(pred_x)\n", - " real_blocks.append(real_x)\n", - "\n", - " obs_chunk = adata.obs.iloc[target_idx][[\"perturbation_gene\", \"condition\", \"split\"]].copy()\n", - " obs_chunk[\"target_cell_id\"] = adata.obs_names[target_idx].astype(str)\n", - " if n_stochastic <= 1:\n", - " obs_chunk[\"sampled_control_cell_id\"] = adata.obs_names[source_idx].astype(str)\n", - " else:\n", - " obs_chunk[\"sampled_control_cell_id\"] = \"stochastic_mean\"\n", - " obs_chunk[\"condition_source\"] = condition_source\n", - " obs_blocks.append(obs_chunk)\n", - "\n", - " pred_gene = np.vstack(pred_blocks).astype(np.float32)\n", - " pred_std_gene = np.vstack(pred_std_blocks).astype(np.float32)\n", - " real_gene = np.vstack(real_blocks).astype(np.float32)\n", - " obs_gene = pd.concat(obs_blocks, axis=0)\n", - " return pred_gene, real_gene, obs_gene, pred_std_gene\n", - "\n", - "\n", - "@torch.no_grad()\n", - "def evaluate_gene_metrics(split_name: str, conditions: list[str], seed: int) -> tuple[pd.DataFrame, ad.AnnData, ad.AnnData]:\n", - " rng = np.random.default_rng(seed)\n", - "\n", - " pred_blocks = []\n", - " real_blocks = []\n", - " obs_blocks = []\n", - " metrics_rows = []\n", - "\n", - " for condition in conditions:\n", - " pred_gene, real_gene, obs_gene, pred_std_gene = predict_gene_expression(\n", - " condition=condition, split_name=split_name, rng=rng, n_stochastic=UNCERTAINTY_SAMPLES\n", - " )\n", - "\n", - " pred_blocks.append(pred_gene)\n", - " real_blocks.append(real_gene)\n", - " obs_blocks.append(obs_gene)\n", - "\n", - " pred_mean = pred_gene.mean(axis=0)\n", - " real_mean = real_gene.mean(axis=0)\n", - " pred_delta = pred_mean - control_mean\n", - " real_delta = real_mean - control_mean\n", - "\n", - " row = {\n", - " \"perturbation_gene\": condition,\n", - " \"n_cells\": int(real_gene.shape[0]),\n", - " \"mse_mean\": float(np.mean((pred_mean - real_mean) ** 2)),\n", - " \"mae_mean\": float(np.mean(np.abs(pred_mean - real_mean))),\n", - " \"pearson_mean\": safe_pearson(pred_mean, real_mean),\n", - " \"pearson_delta\": safe_pearson(pred_delta, real_delta),\n", - " \"delta_l2\": float(np.linalg.norm(pred_delta - real_delta)),\n", - " }\n", - " if UNCERTAINTY_SAMPLES > 1:\n", - " row[\"pred_stochastic_std_mean\"] = float(np.mean(pred_std_gene))\n", - " metrics_rows.append(row)\n", - "\n", - " pred = ad.AnnData(\n", - " X=np.vstack(pred_blocks).astype(np.float32),\n", - " obs=pd.concat(obs_blocks, axis=0),\n", - " var=adata.var.copy(),\n", - " )\n", - " real = ad.AnnData(\n", - " X=np.vstack(real_blocks).astype(np.float32),\n", - " obs=pd.concat(obs_blocks, axis=0),\n", - " var=adata.var.copy(),\n", - " )\n", - " metrics = pd.DataFrame(metrics_rows).sort_values(\"perturbation_gene\").reset_index(drop=True)\n", - " return metrics, pred, real\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "markdown", - "id": "e2bf6516", - "metadata": {}, - "source": [ - "## Train The Conditional Flow Matching Model\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "code", - "execution_count": 27, - "id": "27f82aeb", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "{'epoch': 1, 'train_total': 0.7099, 'train_flow': 0.2342, 'train_recon': 0.2873, 'val_total': 0.336, 'lr': 0.001}\n", - "{'epoch': 2, 'train_total': 0.3107, 'train_flow': 0.0055, 'train_recon': 0.2766, 'val_total': 0.3071, 'lr': 0.001}\n", - "{'epoch': 3, 'train_total': 0.2989, 'train_flow': 0.001, 'train_recon': 0.273, 'val_total': 0.3069, 'lr': 0.001}\n", - "{'epoch': 4, 'train_total': 0.298, 'train_flow': 0.0006, 'train_recon': 0.273, 'val_total': 0.3047, 'lr': 0.001}\n", - "{'epoch': 5, 'train_total': 0.3001, 'train_flow': 0.0004, 'train_recon': 0.2751, 'val_total': 0.3051, 'lr': 0.001}\n", - "{'epoch': 6, 'train_total': 0.3016, 'train_flow': 0.0003, 'train_recon': 0.2763, 'val_total': 0.304, 'lr': 0.001}\n", - "{'epoch': 7, 'train_total': 0.2976, 'train_flow': 0.0002, 'train_recon': 0.2728, 'val_total': 0.3046, 'lr': 0.001}\n", - "{'epoch': 8, 'train_total': 0.2963, 'train_flow': 0.0002, 'train_recon': 0.272, 'val_total': 0.3057, 'lr': 0.001}\n", - "{'epoch': 9, 'train_total': 0.2977, 'train_flow': 0.0001, 'train_recon': 0.2733, 'val_total': 0.3036, 'lr': 0.001}\n", - "{'epoch': 10, 'train_total': 0.3011, 'train_flow': 0.0001, 'train_recon': 0.2761, 'val_total': 0.3054, 'lr': 0.001}\n", - "{'epoch': 11, 'train_total': 0.2978, 'train_flow': 0.0001, 'train_recon': 0.2736, 'val_total': 0.3036, 'lr': 0.001}\n", - "{'epoch': 12, 'train_total': 0.2999, 'train_flow': 0.0001, 'train_recon': 0.2756, 'val_total': 0.3037, 'lr': 0.001}\n", - "{'epoch': 13, 'train_total': 0.299, 'train_flow': 0.0001, 'train_recon': 0.2748, 'val_total': 0.3038, 'lr': 0.0005}\n", - "{'epoch': 14, 'train_total': 0.2968, 'train_flow': 0.0, 'train_recon': 0.2732, 'val_total': 0.3032, 'lr': 0.0005}\n", - "{'epoch': 15, 'train_total': 0.2944, 'train_flow': 0.0, 'train_recon': 0.2713, 'val_total': 0.3045, 'lr': 0.0005}\n", - "{'epoch': 16, 'train_total': 0.2981, 'train_flow': 0.0, 'train_recon': 0.2745, 'val_total': 0.3032, 'lr': 0.0005}\n", - "{'epoch': 17, 'train_total': 0.2954, 'train_flow': 0.0, 'train_recon': 0.2722, 'val_total': 0.3036, 'lr': 0.0005}\n", - "{'epoch': 18, 'train_total': 0.2971, 'train_flow': 0.0, 'train_recon': 0.2734, 'val_total': 0.3034, 'lr': 0.00025}\n", - "{'epoch': 19, 'train_total': 0.2946, 'train_flow': 0.0, 'train_recon': 0.2715, 'val_total': 0.3034, 'lr': 0.00025}\n", - "Early stopping at epoch 19 (no val improvement for 5 epochs).\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
epochtrain_totaltrain_recontrain_flowtrain_endpointtrain_mmdtrain_meanval_totalval_reconval_flowval_endpointval_mmdval_mean
010.7099310.2872590.2341950.0788070.1904180.0129340.3359940.2699050.0157060.0056500.2318840.016517
120.3106610.2765750.0054750.0022250.1821470.0065460.3070920.2680130.0007000.0006300.2214960.015565
230.2988690.2729720.0009560.0007340.1818770.0059090.3069030.2684030.0002670.0004970.2200680.015845
340.2980130.2729530.0005760.0006080.1810310.0057890.3046990.2671930.0001760.0004780.2179660.015207
450.3001270.2750720.0003910.0005450.1818530.0060110.3051270.2675890.0001000.0004540.2186150.015301
560.3016340.2762750.0002670.0005010.1827650.0064310.3039580.2669010.0000620.0004190.2173080.015023
670.2975960.2727880.0002180.0004850.1832950.0059090.3045770.2672640.0000600.0004440.2179270.015209
780.2963100.2719740.0001680.0004680.1817970.0056700.3056660.2680870.0000400.0004280.2176520.015540
890.2976870.2733050.0001400.0004560.1822610.0057180.3036460.2667290.0000500.0004270.2166250.014965
9100.3011060.2761250.0001080.0004480.1827370.0063200.3054050.2679310.0000310.0004330.2177320.015437
10110.2978330.2736290.0000950.0004430.1802840.0058120.3036330.2667980.0000220.0004270.2158060.015008
11120.2998510.2755990.0000770.0004350.1808350.0058360.3037350.2669170.0000170.0004050.2159540.014993
12130.2990170.2748170.0000660.0004350.1807200.0058110.3038450.2669250.0000160.0004260.2163720.015046
13140.2968120.2731850.0000480.0004260.1806660.0052760.3031550.2665270.0000060.0004110.2157280.014840
14150.2943950.2713050.0000410.0004260.1790310.0049120.3044630.2672800.0000050.0004150.2179480.015173
15160.2980510.2745210.0000380.0004240.1796230.0052990.3031930.2665920.0000040.0003930.2154800.014850
16170.2954460.2721800.0000350.0004220.1785580.0051460.3035950.2668210.0000040.0004230.2162110.014935
17180.2971450.2734400.0000320.0004210.1795670.0054900.3034470.2667320.0000050.0003980.2162520.014884
18190.2945520.2715450.0000300.0004200.1782210.0049300.3033980.2667580.0000030.0004080.2153360.014897
\n", - "
" - ], - "text/plain": [ - " epoch train_total train_recon train_flow train_endpoint train_mmd \\\n", - "0 1 0.709931 0.287259 0.234195 0.078807 0.190418 \n", - "1 2 0.310661 0.276575 0.005475 0.002225 0.182147 \n", - "2 3 0.298869 0.272972 0.000956 0.000734 0.181877 \n", - "3 4 0.298013 0.272953 0.000576 0.000608 0.181031 \n", - "4 5 0.300127 0.275072 0.000391 0.000545 0.181853 \n", - "5 6 0.301634 0.276275 0.000267 0.000501 0.182765 \n", - "6 7 0.297596 0.272788 0.000218 0.000485 0.183295 \n", - "7 8 0.296310 0.271974 0.000168 0.000468 0.181797 \n", - "8 9 0.297687 0.273305 0.000140 0.000456 0.182261 \n", - "9 10 0.301106 0.276125 0.000108 0.000448 0.182737 \n", - "10 11 0.297833 0.273629 0.000095 0.000443 0.180284 \n", - "11 12 0.299851 0.275599 0.000077 0.000435 0.180835 \n", - "12 13 0.299017 0.274817 0.000066 0.000435 0.180720 \n", - "13 14 0.296812 0.273185 0.000048 0.000426 0.180666 \n", - "14 15 0.294395 0.271305 0.000041 0.000426 0.179031 \n", - "15 16 0.298051 0.274521 0.000038 0.000424 0.179623 \n", - "16 17 0.295446 0.272180 0.000035 0.000422 0.178558 \n", - "17 18 0.297145 0.273440 0.000032 0.000421 0.179567 \n", - "18 19 0.294552 0.271545 0.000030 0.000420 0.178221 \n", - "\n", - " train_mean val_total val_recon val_flow val_endpoint val_mmd \\\n", - "0 0.012934 0.335994 0.269905 0.015706 0.005650 0.231884 \n", - "1 0.006546 0.307092 0.268013 0.000700 0.000630 0.221496 \n", - "2 0.005909 0.306903 0.268403 0.000267 0.000497 0.220068 \n", - "3 0.005789 0.304699 0.267193 0.000176 0.000478 0.217966 \n", - "4 0.006011 0.305127 0.267589 0.000100 0.000454 0.218615 \n", - "5 0.006431 0.303958 0.266901 0.000062 0.000419 0.217308 \n", - "6 0.005909 0.304577 0.267264 0.000060 0.000444 0.217927 \n", - "7 0.005670 0.305666 0.268087 0.000040 0.000428 0.217652 \n", - "8 0.005718 0.303646 0.266729 0.000050 0.000427 0.216625 \n", - "9 0.006320 0.305405 0.267931 0.000031 0.000433 0.217732 \n", - "10 0.005812 0.303633 0.266798 0.000022 0.000427 0.215806 \n", - "11 0.005836 0.303735 0.266917 0.000017 0.000405 0.215954 \n", - "12 0.005811 0.303845 0.266925 0.000016 0.000426 0.216372 \n", - "13 0.005276 0.303155 0.266527 0.000006 0.000411 0.215728 \n", - "14 0.004912 0.304463 0.267280 0.000005 0.000415 0.217948 \n", - "15 0.005299 0.303193 0.266592 0.000004 0.000393 0.215480 \n", - "16 0.005146 0.303595 0.266821 0.000004 0.000423 0.216211 \n", - "17 0.005490 0.303447 0.266732 0.000005 0.000398 0.216252 \n", - "18 0.004930 0.303398 0.266758 0.000003 0.000408 0.215336 \n", - "\n", - " val_mean \n", - "0 0.016517 \n", - "1 0.015565 \n", - "2 0.015845 \n", - "3 0.015207 \n", - "4 0.015301 \n", - "5 0.015023 \n", - "6 0.015209 \n", - "7 0.015540 \n", - "8 0.014965 \n", - "9 0.015437 \n", - "10 0.015008 \n", - "11 0.014993 \n", - "12 0.015046 \n", - "13 0.014840 \n", - "14 0.015173 \n", - "15 0.014850 \n", - "16 0.014935 \n", - "17 0.014884 \n", - "18 0.014897 " - ] - }, - "execution_count": 27, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "history_rows = []\n", - "best_state = None\n", - "best_val_total = float(\"inf\")\n", - "rng = np.random.default_rng(RANDOM_SEED)\n", - "epochs_no_improve = 0\n", - "\n", - "train_sampling_conditions = [g for g in train_conditions if g in indices_by_split_gene[TRAIN_SPLIT]]\n", - "if not train_sampling_conditions:\n", - " raise ValueError(\"No train perturbation conditions are available.\")\n", - "\n", - "for epoch in range(1, EPOCHS + 1):\n", - " model.train()\n", - " epoch_rows = []\n", - "\n", - " for _ in range(STEPS_PER_EPOCH):\n", - " condition = rng.choice(train_sampling_conditions)\n", - " batch = make_condition_batch(condition=condition, split_name=TRAIN_SPLIT, rng=rng)\n", - "\n", - " optimizer.zero_grad(set_to_none=True)\n", - " losses = compute_loss_terms(batch)\n", - " losses[\"total\"].backward()\n", - " nn.utils.clip_grad_norm_(model.parameters(), max_norm=GRAD_CLIP_NORM)\n", - " optimizer.step()\n", - "\n", - " epoch_rows.append({key: float(value.item()) for key, value in losses.items()})\n", - "\n", - " if lr_scheduler is not None and LR_SCHEDULER == \"onecycle\":\n", - " lr_scheduler.step()\n", - "\n", - " train_summary = pd.DataFrame(epoch_rows).mean().to_dict()\n", - " val_seed = VAL_BATCH_SEED if FIXED_VAL_BATCH_SEED else (RANDOM_SEED + epoch)\n", - " val_sub = VAL_EVAL_SUBSET if (FIXED_VAL_SUBSET and VAL_EVAL_SUBSET) else None\n", - " val_summary = evaluate_split_loss(\n", - " split_name=VAL_SPLIT,\n", - " conditions=val_conditions,\n", - " seed=RANDOM_SEED + epoch,\n", - " steps=VALIDATION_STEPS,\n", - " fixed_condition_subset=val_sub,\n", - " batch_rng_seed=val_seed,\n", - " ) if val_conditions else {}\n", - "\n", - " row = {\"epoch\": epoch, **{f\"train_{k}\": v for k, v in train_summary.items()}}\n", - " row.update({f\"val_{k}\": v for k, v in val_summary.items()})\n", - " history_rows.append(row)\n", - "\n", - " current_val_total = val_summary.get(\"total\", train_summary[\"total\"])\n", - " if current_val_total < best_val_total:\n", - " best_val_total = current_val_total\n", - " best_state = deepcopy(model.state_dict())\n", - " epochs_no_improve = 0\n", - " else:\n", - " epochs_no_improve += 1\n", - "\n", - " if lr_scheduler is not None and LR_SCHEDULER == \"plateau\":\n", - " lr_scheduler.step(current_val_total)\n", - " elif lr_scheduler is not None and LR_SCHEDULER == \"cosine\":\n", - " lr_scheduler.step()\n", - "\n", - " display_row = {\n", - " \"epoch\": epoch,\n", - " \"train_total\": round(train_summary[\"total\"], 4),\n", - " \"train_flow\": round(train_summary[\"flow\"], 4),\n", - " \"train_recon\": round(train_summary[\"recon\"], 4),\n", - " \"val_total\": round(current_val_total, 4),\n", - " \"lr\": round(optimizer.param_groups[0][\"lr\"], 8),\n", - " }\n", - " print(display_row)\n", - "\n", - " if EARLY_STOP_PATIENCE and epochs_no_improve >= EARLY_STOP_PATIENCE:\n", - " print(f\"Early stopping at epoch {epoch} (no val improvement for {EARLY_STOP_PATIENCE} epochs).\")\n", - " break\n", - "\n", - "if best_state is not None:\n", - " model.load_state_dict(best_state)\n", - "\n", - "history = pd.DataFrame(history_rows)\n", - "history\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "markdown", - "id": "fc2eba64", - "metadata": {}, - "source": [ - "## Figures — training dynamics\n", - "\n", - "Loss curves (total / validation) and training component losses. Files: `OUTPUT_DIR / \"figures\"` as PNG and PDF.\n" + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAmkAAAHaCAYAAABbx96UAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAQ6wAAEOsBUJTofAAArz1JREFUeJzs3XdUVFfXBvBn6L2piF2x995Fh2Iv0USjMdZEjbGXYE8ssetr1JhYEiPRaExssWMBsUSR2HuLYldERESk3+8PPiYicw7DMCADz2+tWUtmz75z7hQ83HvP3ipFURQQERERUa5i8r4HQERERETpcZJGRERElAtxkkZERESUC3GSRkRERJQLcZJGRERElAtxkkZERESUC3GSRkRERJQLcZJGRERElAtxkkZERESUC3GSRkRERJQLcZJGpIdp06ZBpVJpboUKFYKXlxeOHj1q0Oc4fvy4wbaXSqVSYeHChQbZ1uLFi7Fnz55095cuXRrDhg0zyHMY0pkzZ9CoUSPY2NhApVIhMjIS8fHx6N+/PwoVKgSVSoXFixcb7PmCgoIwe/Zsg20PABo0aIAffvhB87Ofnx82bNhg0OeQbVetVqNDhw7S3NDQUNja2iI0NNTg4yLKTzhJI9KTtbU1Tpw4gRMnTmD58uV4/vw5vL29cenSJYNsf/r06dkySTMk0SRt27Zt+Oqrr97DiORGjBiBpKQk7N69GydOnIC9vT3Wrl2LdevWYfHixThx4gR69OhhsOcz9CRt27ZtCA0NxWeffaa5L6cnabooXbo0unbtiqlTpxp4VET5i9n7HgCRsTIxMUGjRo00Pzdo0AClS5fGihUrsGzZMr23++bNG1hbWxtiiDmyXW1q166dI8+TWdeuXcOQIUPg6emZ5r6iRYvi008/fY8j083ixYvxySef6PU+JiUlITk5Gebm5tkwsvQ+//xz+Pj4YOHChShUqFCOPCdRnqMQUaZNnTpVsbW1TXe/q6ur0q5dO83Pa9asUapXr65YWloqRYsWVSZNmqQkJiamiQNQjh8/rvj4+Cg2NjbK0KFDFQDpbocOHVLu3LmjAFA2bdqU5nlHjhyplCpVKsPtKoqiAFDmzJmj+Pr6KgULFlTs7OyUvn37KlFRUZr86OhoZejQoUqFChUUa2trpVSpUsoXX3yhREZGah5TqlSpdGNcs2aNJpb6fKm2bNmi1KxZU7G0tFSKFCmijB49Wnnz5o0mfujQIQWAsn//fuWTTz5R7OzslJIlSyrz5s3T6T3ZtWuX0qBBA8XKykopWLCgMnjwYCU6OjrNtt++tWjRQus+3LlzR1EURbly5YrSqVMnxcHBQbGxsVHatWun3Lp1K81zJiUlKf/73/+USpUqKRYWFkrhwoWVrl27KpGRkcrUqVO1PqeiKMr9+/eVbt26Ka6uroqlpaVSunRpZdSoUdL9u337tqJSqZQjR45o7mvRokW655g6daom1r59e8XPz0+pUKGCYmpqqvzzzz/Cz66jo2Oa3Iy2u2nTJqVChQqKra2t4unpqfW1cXNzUxYvXqzL20dEWvBIGpGBREVF4fnz5yhatCgAYNGiRRg3bhxGjx6N//3vf7h69SomT56MpKQkzJ07N01uz549MWjQIEyaNAk2Njbo1asXGjdujOHDh6Nnz54AgCpVqiAiIiJTY3p3u6m+//571KlTB7/++ivu3LmDCRMmIDY2Fhs3bgQAxMTEICkpCbNmzUKhQoVw//59zJo1C507d8ahQ4cApJx6a9euHZo1a4axY8cCAMqWLat1HDt27EDXrl3Ro0cPzJ07F9euXcOkSZNw7949bN68Oc1jBw8ejN69e2Pbtm3466+/MH78eNSoUQNt2rQR7ufmzZvRvXt39O/fH9OnT8fjx48xYcIEvHjxAhs3bkSdOnVw4sQJtGzZEt27d8eAAQPg4OCAuLg4zJs3D4cPH8a2bdsAAEWKFMHt27fRpEkTVKtWDX5+fjAxMcGsWbPg7e2N69evw9LSEgAwfPhwrFy5EqNHj0bLli3x6tUr7N69G9HR0RgwYAAePHiADRs2IDAwEADg4OAAAOjTpw8ePXqEpUuXonDhwrh37x5OnTolfS8DAgJgZmaGBg0aaO778ccf0atXL9jY2GiuMyxevLgmfurUKYSGhmLGjBlwdnZGiRIlpM+h63bPnTuHBQsWYO7cuUhKSsKYMWPQq1cvnDhxQvOY1CPNBw4cwMiRI3V6XiJ6x/ueJRIZo9SjEQkJCUpCQoJy584d5cMPP1QAKP7+/kpUVJRiZ2enTJw4MU3e8uXLFWtrayU8PFxRlP+OeM2dOzfdcwBQFixYkOa+zB5JE223TJkyaY7orV69WlGpVMrVq1e17m9CQoJy7NgxBYBy/fp1zf3ajphpu7927dpK48aN0zxm5cqVCgDlwoULiqL8d7TL19dX85jk5GSldOnSyueff651XKmPKVWqlPLJJ5+kuX/v3r2KSqVSLl26pLnv7aNFqd597RRFUfr06aO4u7unOdIXFham2NnZKT/88IOiKIpy/fp1RaVSKbNnzxaOTXTUytbWVlm6dKkwT5tBgwYpVatWTXd/6pEtbfebm5sr9+7d02lM7742su3a2toqYWFhmvtSP2/3799P91wFCxbMcN+ISDsuHCDS0+vXr2Fubg5zc3OUKVMGhw4dwrJly9C6dWscP34c0dHR6NatGxITEzU3Hx8fvHnzJt3igvbt22fLGEXb7dixI0xNTTU/d+3aFYqiICQkRHPfunXrULt2bdjZ2cHc3BzNmjUDANy4cSNTY4iOjsa5c+fQtWvXNPd3794dAHDs2LE097dq1Urzb5VKhcqVK+PBgwfC7d+4cQN3797Fxx9/nOa1btGiBUxMTDI8QqXN/v370alTJ5iZmWm25+zsjNq1a+Off/4BAAQGBkJRFHz++eeZ3n6dOnWwcOFCLF++HLdu3dIp5/Hjx5m+tqtGjRo6Hz3LjFq1aqUZS5UqVQAg3ftUsGBBhIeHIyEhweBjIMoPOEkj0pO1tTX++ecfzSml8PBwDB06FAAQHh4OIOU/49SJnLm5OcqXLw8AuH//fpptFS5cOFvGKNquq6trmp8dHBxgZWWFx48fA0g5ldmnTx80aNAAf/75J4KDgzWnA2NjYzM1hsjISCiKkm4sjo6OsLS0THcK18nJKc3PFhYW0udMfa27dOmS5rW2sbFBUlJSutdaF+Hh4Vi8eHGa7Zmbm+Po0aOa7T1//hxmZmbpXktd/PHHH/D29sbkyZNRvnx5VKpUCVu3bpXmxMbGak6z6iq7Plfa3iMg/WcjdbyZ/cwQUQpek0akJxMTE9SrV09rzMXFBQCwdetWrUcyypQpk+ZnlUql03NaWVkBAOLj49Pc/+LFC62PF203LCwszc9RUVGIjY1FkSJFAACbNm1CrVq1sHLlSs1jDh8+rNMY3+Xk5ASVSpXuOV++fIm4uDjNa6Wv1Pxly5ahYcOG6eKp1whmdpvt27fHkCFD0sXs7e0BAAUKFEBiYiLCwsIyPVErUqQIfvnlF/z88884ffo0Zs6cie7du+P69etwd3cXjimzdce0vf9WVlbpjmwlJCQgOjo6U9vWRWRkJCwsLDSvGRFlDidpRNmgcePGsLGxwYMHD9ClSxe9tmFubp7uCISrqyvMzc1x9epVzX3x8fGZnkDt3LkTixYt0pzy3Lx5M1QqFerXrw8gpVxH6tGRVOvXr0+3nYyOcgGAnZ0datWqhc2bN2P06NGa+//8808A0JxG1VelSpVQvHhx3L59W3MkM6t8fHxw6dIl1K5dO81p4bd5eXlBpVJhzZo1GD9+vNbHWFhYIC4uTvg8JiYmqF+/PmbOnIkdO3bg1q1bwklaxYoVNYs23n2OzBypKl68OOLj4/Hvv/9qFnoEBgYiKSkpS9vVJjQ0FBUqVMjSNojyM07SiLKBk5MTZsyYgXHjxuHBgwdQq9UwNTXF7du3sX37dmzZsiXNakttKleujO3bt8PDwwO2traoWLEi7O3t8eGHH2LZsmUoV64cChYsiGXLlkFRFJ2PxgFAXFwcOnfujCFDhuDOnTsYP348unbtisqVKwMAWrZsiaFDh+Lbb79F48aNsWfPHgQEBGgdY2BgIA4cOABnZ2eUKVMGBQoUSPe4adOmoXPnzujVqxd69eqF69evY9KkSfjoo49QvXp1ncetjUqlwqJFi9CzZ0+8fv0a7du3h62tLe7evYvdu3dj9uzZmZ4oTJ8+HfXr10fr1q0xaNAgFC5cGE+ePMHhw4fh4eGBTz75BBUqVMDgwYMxZcoUREREwNvbGzExMdi9ezemTZuGYsWKoXLlykhMTMSSJUvQpEkTODg4wM3NDa1bt0bv3r1RsWJFxMfH4/vvv4eTkxPq1KkjHFPTpk0xY8YMPHjwIM1Ky8qVK+PXX3/Fzp07UaRIERQtWlR69LBt27awtbXFwIEDMX78eDx48ABLlizRHKXVd7vanDp1Ch4eHpnKIaK3vN91C0TGSbRC7l2///67Ur9+fcXa2lpxcHBQateurXz99ddKQkKCoij/rYp79uxZutyjR48qderUUaytrTV10hQlZZVh586dFQcHB6VYsWLK4sWLhas7tW0X/18nbcyYMYqLi4tiZ2en9O7dW3n58qXmMYmJicrYsWOVQoUKKfb29krXrl2V4ODgdCtLL126pHh4eCj29vYZ1knbvHmzUqNGDcXCwkJxc3NTRo0apbVO2j///JMm74MPPtDUF5PZv3+/ZuWhra2tUrVqVWXs2LFparvpurpTURTlxo0byscff6wUKFBAU8usT58+aVaLJiUlKfPnz1fKly+vmJubK25ubkr37t01r2VCQoIyZMgQpXDhwopKpVJatGihxMbGKgMGDFAqVqyoWFtbKy4uLkqrVq2UkJAQ6f7FxcUpBQoUUFatWpXm/gcPHijt2rVTnJyctNYz08bf31+pWrWqYmVlpTRq1Eg5e/ZsutcmM9s9e/Zsms+ooijK06dPFVNTUyUgIEC6X0QkplIURXlfE0QiItLd2LFjcfbsWU3dtdzshx9+wHfffYebN29m6igvEf2HkzQiIiPx+PFjlCtXDsePH0fNmjXf93CEkpOTUalSJUyZMgV9+vR538MhMloswUFEZCSKFCkCPz8/PHv27H0PRerRo0fo168fevXq9b6HQmTUeCSNiIiIKBfikTQiIiKiXIiTNCIiIqJciJM0IiIiolyIkzQiIiKiXIgdBwyoQ4cO2Lxpk9bYsznDpblrW4wTxrpXd5PmFrARv42OTy5IcxUT7S1vAODjQ8nSXO8q4ubNHSsWlOa6xT8RxsIs5fvr+uaRMDbtfJIwBgAJndoLY7Pv7JbmvnAoKYy5hF+R5l62Ele8P/XwpTTXrVtHYWzHXXmu0649wpjvnTXS3EVlPxPGZlSU93mMdq0qjM08dFua276K+P3/50GkNPfzOuKK+KYZlOqyfhMuDppaiGMAjr8Qfwc9TOUN3k9/MVYYS1ih/fdJKuvRnwhj1X9YKs1VzK2EsYEHnktzv2xaRhirp5Lvb+LNM8LY+Fd1pblda4rfX3MT+TEHdbeJwtirXdrbeqV67vedMJYUK277BQAFWzQXxtpfKi3N3WZ/QBiz7PCFNHf3Y/EHvl2YeLsA8J2qsTD2lXuMNNesVO4tDWOMeCTNgLLa546IiIgoFSdpRERERLlQnpukJSQkwMnJCRcu/Heab968efjoo48yfNyXX36JFi1aoHHjxggODs6xMRMRERG9K89N0oKCgvDBBx9gx44dmvtOnjwJExMTxMXFCR938eJFmJqa4vDhw/j7779Rtar4mhoiIiKi7JbnJmm7du3CzJkzcfbsWQDAs2fP4OrqitatWyMgIED4OCsrK1y6dAn379+HiYkJ7O3tpc+zYMECuLq6prm9efMm+3aMiIiI8pU8N0l79OgRSpQogWLFiuHJkyfYvXs3OnbsiE6dOmH37t3Cx5UvXx59+/bFhx9+iGbNmuHhw4fS5/H19UVYWFiam7W1dXbvHhEREeUTeaoEx4ULF3D+/Hm0adMGERER2LlzJwICAhAeHg4zMzM8evRI+LiBAweif//+6N+/P3bu3ImlS5di3rx573mPiIiIKL/KU5O0HTt2YP369ahfvz7i4+PRvXt3WFhY4ODBgwCAb7/9FqdPn8bevXvTPK5379746KOPYGpqCkdHRxQqVAjJyfIaYURERETZKU+d7gwMDETt2rUBABYWFtixYwdq1KihiTdv3hw7duxI97jo6Gg8ePAAHTp0QIsWLTB58mQMGzbsvewDEREREQCoFEVR3vcg8goftQd2zxiiNeZfQC3NbfNUXAHav3BL+fNeWC2M3dl1QppbtkdbYczEo7s099Ir8YHYWvE3pLl9j4s7HcxpX0maa2UmrqTtmCCvwC8TYeoojTtf2CGMqap5SnNNYl4IY3F//yXNvdZksDB28oF8f/s92iyMrS/RTZrbp7y4Iv31N+IYABT8ZYIw9mqg/DKCEEkHhqbr5JXh//1ysTBW2M5SmlvBUfw3a6JKftLB+u4/wtid5T9Kc4/1mS+MlXWxkeY+fR0vjLVyd5bmhjTwEMb+miH+nQIAS2qJC3evrtdXmlvh1DFhzHlCL2lu1W98hbGgHvLPRsWPagljAZ2+keZGx4s7mew5/1iaO+8DcaWAwDvyzg53w18LY/PryI+x/P7MQRhbeeCmNHeP6yFhrOBS+fXa8Wd/kcYpc/LUkTQiIiKivCJPTdJ0KWQ7bdo0NG3aVPPzxx9/jMGDU45UVKlSBWq1Gk2bNsXVq1dzbuBERERE78hTkzRdC9nGxcUhPDwc8fHxCAsL09xfsmRJBAUFYeXKlVi4cGGOjp2IiIjobXlqkqZrIdv27dtjz549OHz4MFq0aJFuO1WrVsWTJ09ybNxERERE78pTkzRdC9l6eXnh0KFD2L17N9q3b59uOydOnIC7u7v0ubR2HIgVX8RLRERElBl5pk6aroVsAcDSMmWV1+PHj+Hq6qq5/969e/D09ISDgwN+/FG+IsvX1xe+vmlXGvmoxauliIiIiDIjz0zSdC1km6pHjx6IiIhIs42SJUvC398/R8dNREREpE2eOd2payHbVK1bt8Ynn3yS4+MkIiIi0gWL2RqQj0cTbBugvfDsWKvO0tw+3w8Xxr7uOluau7JnLWGstFmMNFcxNRfGrg34VJrr3qGhMGbVUp4b41BcGHsYnSDNtbcQF8JdePi2NLeim7jA40dVCklzTX6ZIoxZfyF/j84+ERelbGzxTJp726KYMFb86Cpprpm6pzD2CPbSXAfJ62x3fpc017SAmzB2c+Eiaa7N/N+EsUev5Nd9Vne1FsbMksQFWAFg3MH7wth85wvCGADE3xUXBz0wTlxQGAA67f9OGDthX0eaW9vNVhiTFdgFgNhzR4SxSnvEnzkAWBEkHnPRE+LtAkCy5L+c8gfln42bPmPEz7tO/P0EgMC2E4Wxru7izw0ALDwlLjo7oJ78tXKC+HN3N078+xcAXvbrIozV+UFcBBkAPLe8EsY8qhaW5n4/f60w5uJeU5oburqHNE6Zk2eOpBERERHlJZykEREREeVCeW6SpmvXgTp16qB58+aYPHmyJq9bt25Qq9Vo0qQJQkNDc3roRERERBp5bpKma9eBH3/8EUeOHMGlS5fw6tUr+Pv7o2HDhggKCkJQUBAKF5afsyciIiLKTnlukqZr1wEASE5ORnR0NJKSkmBtbY1//vkHz549g4WFBayt5ReSEhEREWWnPDdJ07XrwJAhQ1C2bFnUqFEDTk5O8Pb2Rr169eDl5YUOHTrg1SvxyhhAe8eBmDj5qkQiIiIiXeWpSdrbXQeCg4Oxc+dO+Pv747vvvkOfPn1w9OhRzWN//PFHnD59GtevX4eiKFCpVPD19cXFixfRunVr/PabuAwAkNJxICwsLM3NxlK+nJqIiIhIV3mm4wCQ+a4DLi4uqF+/Pg4ePIhq1arB2dkZVlZWKFSoEJ4/F9fFISIiIspueepIWma7DgBAv379sHr1aoSGhsLT0xNqtRrr1q1Dr169cnTsRERERG9jxwEDaqb2Qs85v2iNdTm0QJr7vL+4Yr2zlbjyOwAUfiSpLG5fUJqb/OC6MFbyuyfS3NOr+gljBc0yuD7vn53CkNK4mzT1x1OPhLHhNtekuaaOBcTPm5QkzV3+vKgwZiepzg8ATUo6CWOXqjeW5rbfJK6knvg4VJqLjqOEoSevE6Wppx9FCWMflLaS5iYf+V0Yi27RX5p7TtKd4WWcfMwTfgwWxs7/r60011QRb1v1z3Zp7ouTJ4WxnZ5fSXO7VBJ/R8Ni5Pu78fxjYax2MUdpbnVXcbcCy2Xiyv4A8HiAuNp9+b3y33UWTuKuH+ae8lZ99f53WRhbM6yJNLdWzBVhTLFxkuYiWfLZkMQA4NacWcLYdz+dkeYuuyv+Pal6/UKaO/CcnTC2648gae7xkiHCWL9ig6S5x8Z7SeOUOXnqSBoRERFRXsFJGhEREVEuZPSTNEN2GFi9ejUaNWqEJk2a4PDhwzm+L0RERESpjH6SZsgOAz4+PggODoa/vz/mzp37PnaHiIiICEAemKQZssNAqVKlAKSsDM2ItmK28bFvDL+DRERElC8Z/SQtOzoMzJkzB3379pU+r7ZithZWbCVFREREhmHUk7Ts6DAQGBiIe/fuoUePHu9rt4iIiIiMu+OAoTsM3Lt3D7NmzcLOneLaNERERKQ/jyq1kGCu/zEi96qVsGHDBgOOKPcy6klaYGAgJkyYAOC/DgMzZszQxFM7DKhUKs19/fr1w8SJEzFy5Eh8+OGHsLS0hK2tLTZs2IBx48bh4cOHaNeuHdzc3LBx48Yc3yciIqK8LMHcBHUvRuidf8P1mQFHk7ux44AB1W+mRqFPtXcO2NqnujQ3uHlLYSx5o/zInkf0KWEsqby8CvfBe+LFDm1N/5XmJrmUEMZqzTknzf20fSVh7Ks68krppq/ChDHVG3GVfAC4bF9NGDv/RJ77SQHxL5Wr5qWluZ+tklSkH9VUmusaflEYu2xTWZpbwT5ZGKv37d/S3Bl96gpjzQ6IK84DwMtPpwtjbnbm0tzQl/HCWDF7eW6Vz/2Esdu9xK8FALxo1k8YuxERK81tcGGdMPbKS16h3TnplTC26Lw4BgCjGoq7YEAlP1Kx6Wq4MPZxeXG1egDoukHc2WNaO/F3GwC2XhJ3MvnaQ7I/AMwi7gljSQ5u0lzTu2eFscTy8u/gfV9xl4zi7TyluS8viTsdFPxE/tlQTMWf9zdOpaS5rxaOFMaiH4rfewCIixJ/3n/8Q7w/ALBCCZXGAaBRzTpocEneMUHmqlc5HDhwQO98Y2LUR9KIiIjIuKgAmKoyfBghF0zSEhISUKhQIRw5ckTTDH3evHkICQnBli1bAACnTp3CqFGjYGJigvLly2P16tXw8vLSLACoWbMm5s6dC0tLS7Rp0wb+/v4AAD8/P1hZWWkK1FaoUAH29vb4448/MH/+fOzYsQOWlpbo0aMHkpKSsGPHDly7dg3u7u6wsLDA9u3b4egoP6pDRERElB3e++pOXYrRzp07F9u2bcORI0ewaNEiACnXoB06dAiBgYEoXbo0lixZIn2eTp06ISgoCK1bt9ZM/n788UccP34cW7ZswbBhwxAUFIQ2bdpg69atCAoK4gSNiIgoG5iqVHrf8pP3PknTpRitlZUVDh48iISEBK0Tp9QJli6qV6+Ohw8fan5WqVQoW7YsIiL0v4iRiIiIdKRKOd2p7y0/ee+TNF2K0c6dOxd79+5FxYoVsXjx4nTbMDU1Rer6h3v37kGtVkOtVmtt7fT333+jfPnymp/j4+Nx69YtFCpUKFPj1tZxIClefnExERER8Uiart7rNWlvF6ONiIjAzp07ERAQgPDwcJiZmeHRo0cAgOLFi2Pt2rWIj4+Ht7c3Pv/88zTbSUpKgolJynyzZMmSaa5JS7Vjxw5cv34dFStWxLhx43Dx4kUMGTIEDg4OGDlyJExNTTM1dl9fX/j6+qa5r34zdSZfASIiovyFCwd0914naboWo3V2dtZczG9vb59uO8uXL4darZY+V6dOnbBixYo09/34449o1KiRwfaHiIiIyFDe6yRN12K0iqLgwIEDMDU1RceOHWFvb4/4+Hh4enqmWd1JREREuV9+O22pLxazNSBvHx+s37xda+yz389Lc+cfmyWMVVj6kzQ3ycZZHMvg3U3+Q3vxXQAw7zFJmrv81CNhrNuBOdLc6PvigrQVvpkqzT2aLC7iWL+IrTQ37OuBwphDmSLS3M5v2gljy88tlOZWnjVTGEs2t5bmnk5wFcYcreR/ZyWN/UQYqzhmiDTX1LGAMKYkJUlzo4J2CWPWpcpIczcWbCOMdapYUJors+O6vICnuan4Et2PKshXeiepxO+DqZIozY1OEj9v0ir5d/DyB1OEsWbPj0lzIw4HCmObPcZIc5uXdhHGnK3kl444Worj1q8eCmMA8PdrJ2EssZP4cwMAzYO2CmN/3JdPGKq5pj+Dk2rPdfHvMgAYmxAkDnqIv58AUOfrQ8LYxq9aSHO9B68Sxqwc5d+jg9biAur1rpeT5r46LC90DQBNatVBm+vy4uEyR5uVYTFbIiIiouzAI2m64SSNiIiIcgwXDuguz07SMupk0KdPH9y7dw8XLlxAjRo1UKZMGZQqVQo7duyAg4MDWrRogenTxf0HiYiIiLJTnp2kvd3JIHWSdvLkSZiamiIuLg5r164FgDRtpKZNm8YVn0RERNmMpzt1896L2WYXXToZEBERUc5jxwHd5NlJmi6dDLQZMmQI1Go1tm3bJt2+to4Db2JiDL0bREREeQ47DugmT57u1LWTgTa6nu7U1nHA28cny2MnIiLKy7hwQHd5cpKmayeDunXrvueREhER5T/57YiYvvLk6c7AwEDUrl0bwH+dDFIXDwD/dTIgIiIiyq3YccCAWnqp4b9ce+eARFd5lWbTB5eEMZW9uLo3AFh3+V4Yi/pbHAOAZ7Hit9/tX3FFcgB4ff6keEzd5BXL1994LYyZZ3AcvKW7+PVwMpNXdze5JN6npxVaSnMdLMR/09hG/CvNfe7oLozZ/CWv0N0zVlxJ/a/mCdLc5JhX4mCxitJck1hxrmJmJc19bFNCGNtzU175f8WOq8JY7zYVpLn9aom7RjjcllfgP+kgPrJexslSmnsoNFIY6xb3jzT3sEtTYay5q/y70HvbHWFsfTt5Vfnkq8eFMbPSVaW5sf/sF8b2DP1Vmtt57wJhzHbAFmlu+LEfhLGMuhUot84IYwl1Okpzzz0VX2887OcQaW5A8SBhbIJLD2nuDx7iDirK7XPSXFRpLgy9MBV3UACAqU7i93/M4HrS3ArLN8vHBaBZ7TroE6r/Ndyb6pVgxwEiIiKi7MBr0nTDSRoRERHlGBXy3ypNfRnVNWkJCQlwcnLChQsXAAAuLi5Qq9Vo1KgRjh9POXR/8uRJeHl5oXnz5ujWrRuio6Ph5eWledwPP6QcLl+2bBmKFSuGuXPnara/Z88eNGnSBB07dkR0dHTO7yAREVE+wDppujGqSdrbXQQAoEGDBggKCsK2bduwZMkSxMbGYsKECdiyZQuOHDmCadOmIT4+HhYWFggKCsKJEyfw559/AgC6d++OdevWpdn+4sWLERgYiN69e8PPzy+nd4+IiIhIw6gmae92EUj16tUr2NnZITg4GG3atIGzszMAoGrVqnBx+e8i84SEBJiZpZzhLVSoEExM/tv9169fw9nZGVZWVvD29kZIiPxiUG3FbGPexBpqV4mIiPIsFrPVjVFN0t7tIhASEgK1Wo3mzZujf//+ePLkCYoUSb+6Kz4+Hmq1GpUrV0arVq20bjsyMhIODg4AAEdHR0RGRkrH4uvri7CwsDQ3G2v5ijciIqL8LrWYLU93ZsxoJmlvdxEIDg7Gzp07Nac7L1++jEWLFsHNzU1rN4HU0503btxAYGAgEhLSly1wdHREVFQUACAqKgpOTk7ZvUtERET5Uk4dSYuLi8OAAQNQpkwZ2Nvbo2rVqtiwYUM27ZXhGc0kLbWLgL+/P44dO6bpHgAAtra2iI6ORqNGjbBv3z68ePECAHD16lXNvwHA1NQUKpVK6yTNzs4OL168QFxcHAICAtCgQYPs3ykiIqJ8KKeOpCUmJqJo0aIICAhAVFQUVq5ciS+//BInTpzInh0zMKMpwREYGIgJEyYASDkyFh0drTndGRsbiylTpsDKygpz587FRx99hMTERLi6usLPz09zujMxMREtW7aEjY0N1q9fj0WLFiEqKgrR0dGYOXMmRo4cCU9PT7i4uOD3339/z3tMREREWWFra4sZM2Zofm7WrBmaNm2K48ePo3Hjxu9xZLphxwEDaunthT1rf9QaC7MpLs0tqHojjG0LjZfmflBRXIE/8c+5whgAhLf/Sjym3QuluRalygtjtyt1kub+efGxMDahgbxSetP/nRLGxn1UTZrbrKSjMOb6+r4095ppUWGsgtlLaW6sTSFhbPye69LcdiN7iWOHVkpzE+5cFsbimom3CwCvE5KFsYx+aZiumiiMdTP5UJrbpVFJYaxFaXn3jVpJocLYg5VLpLmxo5YKY2VMIqW5u5+aC2MdCoq7awDAdUX82Rix6YI0d0+VW8KYiZW4Wj0AJD4TV+h/cUHc9QEAXn4h7hpw4Ym8fFHH62uFsbsthkhz+60Sdzk52uyJNFeJFy/sirkj7xjituqZMJZRVxez8NvCWP2VD6S5/uPEXQMWHhF3mwCAr5qXEcb2lJH3rf70xkFhbEShZtLcFUqoNA4AzWvXxVdP4jJ8nMg8d0fcvHkzzX2+vr7w9fXNMPf169coU6YM1q1bh9atW+s9hpxiNEfSiIiIKG8wycIqTRsbG4SFhWU6Lzk5Gf369UP9+vWFiwhzG6O5Jk3EkAVuJ06cCLVajWbNmqFWrVrva5eIiIjyLJUKUJmq9L7pQ1EUDB48GI8ePcIff/wBlZGU8jD6I2lvF7itUaMGGjRoAH9/fzx+/BijRo1CnTp1MGHCBGzduhXOzs64fPmypsCtv78/FEWBWq3G0KFDMWfOHAAp17/t3y9uIExERET6M8nBWhqKomDo0KE4d+4cDh48CDs7uxx77qwy+iNphixwm2rHjh3o1El+TRURERHpR2Vqovcts4YNG4bg4GDs27dPUw/VWBj9JC07CtyePHkSjRo1kj6v9o4D4ov/iYiIKGfdvXsXP/74I65cuYISJUrAzs4OdnZ2mD179vsemk6MepKWHQVuL1++jMqVK6dpGaWN9o4D1tmyn0RERHlJTl2TVqpUKSiKgtjYWERHR2tukyZNyqY9MyyjnqRlR4HbnTt38lQnERFRdlGlXJOm7y0/MeqFA4YucAsABw4cwIgRI97nbhEREeVhKqhM8tdkS19GP0l72+7du7U+rmHDhuke++7PqQICAgwzOCIiItIqvx0R0xc7DhiQT/Om2DH2Y62xf+v3k+aWPfGTMHa7yUBpbrLkHaySeE+a+8LJXRhzCb8izX264RdhrFDrttLciArewphz/HNp7sHnlsJY85LylTshD8XV0OsWkVdoL+QxTBiLnFJFmmtWrKwwFnXymDT35xqDhbHaReX7W6eIeKn5nUh5xe+KLlbCmFWivIo+zu0ThpJfv5Kmtr0ufi0PthZ3QQCA24XEldQdLU2luYmSL5LLMT9p7psH4m4VFvY20tykzuIq6UfvRUlzzz8Wx2sXE3fXAICAauLFUTG/b5PmDv5zgjBW9Rt51fedKvH728lcXJ0fAJJcSghjM4p7SnO7dRJ3SKnot0ma+4GfuPND1/rybjL9LcUdRY7Z1ZHmJiSJP+8V1oyT5i6oN1oYszCTX+n0cc1iwthvp+WdWZZ9VFMaB4AWdethaoL+V1vNKeyIAwcO6J1vTIz6SBoREREZH32L0uY3Rr1wAMh6xwG1Wo2xY8cCALy9vWFnZ4cnT+Q94IiIiEhPqpytk2bMjP5IWlY7Drxtw4YNGD9+/HvaEyIiorxPBV6Tpiujn5JmtePA2woXLpzt4yUiIsrvVCYqvW/5idFP0rLacUCtVmPu3LmZfl6tHQfiEgyxS0RERETGfbrz7Y4DERERmo4D/v7+eP78OQYOHIgRI0YgODg4Xa62052Z4evrC1/ftKuYfJo31Xt7RERE+YVJPru2TF9G/SoZouMAERER5aycagtl7Iz6SJohOg4AQMWKFbFy5Ur07dsXBw4cwO3btzFr1ix4eHi8x70jIiLKg1QswaErFrM1IG8fH6zfvF1r7MqzGGluGWdx4dCilonS3OuvxAdEi9jJ5+GOV/YLY7vtxMUuAaB50CJhLC5SXDQWAFz7iVtvJVvZS3OTTvwljL28Ii4cCQB2w+cLY5Yv7kpzn9uKC2kWjJIX4YQiLkoZV6iCNPX4A3HxV/MMThk0vrtHGEtu3E2a+zRG/Lm7/1JeCLegjYUwVtZMXqD10HNzYay0k7U091q4uMhug+0zpbkXPvpGGNt3LUya+22M9u89AETdeSzNbfG4pTB26RP5dz+0XCthrIwSLs198sNsYaxtzAfS3OAZPsKYacBqae6jRn2FsYi+naW5NX9bL4yNKCD/fTV+TDNhrGXMh9LcJ5f+FsYqeYrfAwA42ddJGLtsXUma+/mqk8LYrtHyy2ts/hL/rpvi8JE098M5XwhjjQLEv1MAwNLeWRoHAHX9+pjnJC60nZEpMGMxWyIiIqLskKUjaUmGG0duZ9TXpBERERHlVXlikmaIrgPbtm3D8ePHUaVKFbRp0+Z97g4REVGepQJgYqLS+5af5InTnYbqOhAVFYXTp0+jS5cu73FviIiI8rD/bwult3xUkjRPHEkzVNcBBwcHWFvLL0xOpa2Y7ZsY+eIAIiIiSmkLpe8tP8kTkzRDdB24c+dOpp7T19cXYWFhaW7WNjaG2iUiIqI8i3XSdGP0pzvfZ9cBIiIiouxi9EfS2HWAiIjIuKhMTfS+5SdGv7eBgYGoXbs2gPRdB9RqNUaMGJGm60Dz5s3x9ddfw9w8fcHMS5cuwcfHB6dOnULHjh1zeleIiIjyPhWvSdMVOw4YkJe3D1b/sU1rrOglcUVyALheQTwpzKhrgEv4FWHslm1Fae7GC+Jq6F5lC0hz69/eKYyZ1PSW5t6ZPEoYM7cVd18AgJC+C4SxTv/+Ls1dV7yrMNYv7oQ016RcXWFsZai4Sj4AvHwjXo7Uu1ZRaa7sD0eHHeLXAgASPxovjFmr5BUhTW8cE8ZUBYpLcxVLW2Hs72gHae4cf3HXiEd35UfAT3cSL95JePivNPdlq+HC2LNB4s8NAFSdJn6dr7nUkea2myD+3fDp79OkuVPWDRTGbnsMkeZWTggVB6MjpLkvSjQQxhyS5QuoTGIk76GJ/LjBhQTx7yQ7C1NpbvU2o4SxL78RxwBgomdZYWyac1Vp7rIL4g4M936Rd2co8O3PwljDSfKK+3u/8RLG1p59JM31jRL/bv+3sfgzBwDVizpK4wDg2bA+vi/nmuHjREaHxbPjABEREZHhqWCSz05b6suoXyV9i9g+ffoUXbt21awA3b17NwCgb9++cHZ21rrIgIiIiCgnGfWRNH2L2I4aNQrffPMNatSogYSEBJw7dw4AMG/ePJQpU+b97hQREVEel99KaejLqI+k6VPE1sHBAa9fv0aNGjUAAObm5qhfvz4AwM3NLWd3gIiIKJ9Rqbi6U1dGvbf6FLENDw9HwYIFAQCnT5+GWq1G3759M/3cWjsOvGHHASIiooyoTEz0vuUnRru3bxexDQ4O1hSxDQoKwuXLl7Fo0SK4ubnh0aO0q1gKFiyIZ8+eAQDq1q2LoKAgPH36NNPPr7XjgDU7DhAREWXExNRE71t+YrR7q28R21evXsHW1hbnz58HACQmJr6X8RMREeVXPN2pG6NdOBAYGIgJEyYASF/ENjY2FlOmTElTxDYxMRGurq7w8/PDkiVLMGTIEERERMDExARDhw4FAIwbNw6bN2/Gnj174Ovri27dur3PXSQiIqJ8zKgnaW9LLaPxroYNG6Z7rJ2dHbZs2ZLusfPnz8f8+fMNN0giIiJKS6XKd0fE9MWOAwbk4+ON3Zs2aI39dFW+qODu89fCWP/6JaS5FWNuCGN1f3omzd1pvlUYsy4grxx9cqG44nOi/z5p7qUnr4Sxf8OipbmtKhcWxj4sIV/WrYqNEsYeWhWT5hZVXgpjipmlNDfxwC/C2JU1+6W5U9rPEMbKu9lJc8e2cBfG4ifJF8yYzfpVGLvXykeaW3iveJ/KPz4uzU0IvSaMmbQaIM293k989Lvy2C+kuZeLewpjVmby/1Bk/9+USQ6T5iY5ij93t1/GS3PHbxd3G9m34idp7r8dxYOOH/6dNLdU2Clh7KJjbWmus7W4M4CbvNkITKPFv88SgsVV8gHg9e1QYWx+uc+luXNqi8esihP/TgGAMReshbFPass7dzhbi4+jlLaMk+a2XSv+Hv18Z7k016WquBRV4utYaW7Bkf+TxgHAq3FD/NxI3MUhI19cesaOA0RERETZQWUqb+NFKYzmeKMu3QXatGmjebyfnx82btyI0NBQFC1aFN7e3vD09MSOHTsAAP/++y9q1qyJatWqaXK2bt0Kd3d3DB48OAf3jIiIKH/hwgHdGM3evt1dAICm3Ma2bduwZMkSaW6nTp0QEBCAXbt2YenSpbh//z6KFCmCY8eOoXjx/w43q9Vq7NsnP01HRERElBOMZpKWUXcBXdja2mLgwIE4ePAgbGxsYG9vnybu4uICc3Nzg42ZiIiI0jMxMdH7lp8YzTVpou4C165dw+bNmwEA9+7dg1qtBgA8efIE06ZNS7cdNzc3nDhxIsvjWbBgARYsWJDmvvLlymV5u0RERHkaV3fqzCgmaW93F4iIiNB0F/D398fz588xcOBANGvWDCVLloS/vz+AlGvStHn8+LFBenT6+vrC19c3zX0+Pt5Z3i4REVFepgI4SdORUUzSUrsL1K9fH/Hx8ejdu7cmltpdQBcxMTH45ZdfsHr16uwaKhEREWUgv/Xg1JdRvEqBgYGoXTul9s673QXUajVGjBghzd+xYwe8vLzQvn17DB06FCVKlEB4eDh8fHxw6tQp+Pj4ICoqCgEBAejVqxd27dqFgQMH5sSuEREREWnFYrYG1NJLDf/ls7TGXh3dI821a/mxMHbXRlyQFACSJG9hQUkxRADYcFHcXN7NXl6gtbPTC2FMsRAXcASAv6MdhLFKBeS555+Kj5w+fiUv8NjbRby/irmNNDfZWlzc99gL+Wv10/FQYWzNx9WEMQBYHPxAGIuMSZDmDm9SUhhzspTXKTK/4C+MmVjbSnOTi1YSxtr9cU+a26R8QWHsgyriQsYAUMP8uTD2apu8uOvDDyYJYyp5jWQcCRV/F8oXkH+uvJLERUfvLP9Rmvu9p3jM01rKr5E9XrmRMKbskxdYDrkfKYwNqCcv0FrYTPwdfaWSV7O1k/w6e7NupjTX5uNRwtjw/Y+kuSOai38Hv4yV939e515XGLuzbL0095eeNYUxt7Dz0tykZw+FsREe46S5Q7pXEcZ+/ENcQBkAViih0jgAeDdphHXt5EWPZfoevsVitkREREQGp+I1abriJI2IiIhylAknaTrJda9STnQWWLJkieZ6tgIFCmju/+OPP1C3rvjQNBEREWWVCioTE71v+Umu29uc6CwwcuRIBAUF4ZdffkGrVq009+/duxeVKlXCgwfia4CIiIgoa9gWSje5bm9zorNAqh07dqBjx44AgMTERMTExKB///6aCSIRERHR+5LrJmmizgLNmzdH//79AfzXWUCtVmPu3Llat+Pm5oanT8Ur+QDA398f7dq1AwAcO3YMLVq0gFqtxrFjxzIc54IFC+Dq6prmFvMmNpN7S0RElM+oeCRNV7lq4UBOdhaIjIwEADg5OQFIOYJ36tQp7Ny5Ezdu3EB0dLT0yJ22jgMtvdQ67ysREVF+ld+uLdNXrpqk5WRnAX9//zQLEG7duoWgoCAAwLp167Bv3z589NFH+u0IERERaaUCYGIqr9NIKXLVVDanOgsAwM6dO9GpUycAwNWrV1G48H9FMps3b87r0oiIiLIJT3fqhh0HDMjbxwfrNv2lNdZ03F5p7vbprYSxSuYvpblmL8XVsi/bVJbmlr/whzBmWkDeiD65QlNh7PoXvYUxAKg4uKcwFnX2tDT3Mxtxd4b1vWpJc21uHxfGVGbm0tyRV8QdB746+T9p7t5VJ4WxQWd+k+YGq8TVzhshVJp7d6l4XCW++kaam3T1hDBmUru1NPfZsmnCWPxQ+Wt1Oey1MFYkgy4YsYnJwpiNufwvdycrcbzIqd+luQ/q9BDGSprFSHOTJJ0sXi6VV4Z3Giqusm9294w0N+qIuKOEbY8x0tzbceKuIGUc5d+j16vFn7uEXtOkud/suyGMTT4nX/1f8OvlwlhSBv8LHrsfJYz9I+m+AABmJuJ2FYnJ8ieeNW6eMPbqkDgGAH/eFl8jffRmuDR36OaJwlj9W0WlufFnf5HGAcCnWWNs7uOZ4eNEum36hx0HiIiIiAxPle+OiOmLrxIRERHlHBVytJjtsmXLUK9ePVhaWqJHD/FR79woT07SMupasGjRIqjVari5uaFJkyZQq9U4f/58us4EREREZHg5eU1a0aJFMWXKFAwcODAb9iR75clJWkZdC8aMGYOgoCC0adMGW7duRVBQEMqXL5+uMwEREREZXk5O0j788EN07twZBQsWzIY9yV55cpKmT9cCWWcCbbQVs30TI79AmIiIiLImJiYm3f+/CxYseN/DyhZ5cuGAqGvBtWvXsHnzZoM8h7Zitt4+PgbZNhERUV6WlWK2NjY2CAsLM+Bocq88N0nTtWsBERERvQcqFVQmLGarizw3STNU1wIiIiLKJpyk6STPXZOmb9cCUWcCIiIiMjATE/1vmZSYmIjY2FgkJiYiOTkZsbGxSEhIyIadMjx2HDCgek1aAD6TtMaO1LwozW1/v4EwdqCLuCI5AJxRlRDG6rw4Jc3daVpdGPv1xF1p7pyOVYSxUn+vkuaau1cVxqY9KS3N/aBqYWGslKO8In2BByHCmMrKRpp72ry8MBbeXF49u/jxI8LYqYfyjhJ97UKFMZW5hTQ3ydpJGHuzf70016pyHWHsYTn59ZdFzeOFsYlB4g4ZADD5/lphzLac+D0AgF3FOghjXeweS3PjLhwTxkwdC0hzv1MaCmPdq8s7dzzo2EYY+2e++LUAgAF1xNXfLU5slOberS7uTVwu+qo0d29cKWGsVWFx1wcAUM4fFMYSGnWT5n6y7pww1jqD13lwCfHCrvgz4jEBwN2G/YWxL36Td3bYP7i2MObzg7y7yv5K4v83nKadl+aeLPNQGIt+LD+jVKyR+P0tOUBexsKsasadBHyaN8XOMV0zfJxIpx/2ZKrjwLRp0zB9+vQ09/Xt2xd+fn56jyGn5LkjaURERESppk2bBkVR0tyMYYIG5JNJ2rvFbXft2oXmzZvDw8MD/fun/IVUpUoVeHl5Qa1WY+bMmeABRiIiouygSrkmTd9bPpIvJmlvF7d9+vQpli1bhv379+Po0aP47LPPAAAlS5ZEYGAgDh06hIiICGzatOk9j5qIiCiP4iRNJ/likvZ2cdu9e/eif//+sLKyAgB4eHikeaxKpcL48eOxe/fu9zFUIiKivC2He3cas3yxt+8Wty1SpIj08a6urnj27Jn0Mdo6DiQlxBly2ERERHkTj6TpJM9P0t4ubhscHIwCBQrg0SP56rKnT5/C1dVV+hhfX1+EhYWluZmay1cWEhEREekqzxWzfde7xW1bt24NS0tLdO7cGVZWVjh27Fi6DgT/+9//0L59+/c0YiIiojwunx0R01een6QFBgZiwoQJAFKK29rY2OCLL75Aq1atoCgKypUrh2bNmuHevXvw8vJCcnIyfHx80LWr/jVciIiISESV764t0xeL2RpQSy819q2aqzWW6CguOgkAf9xXCWOfmMoLS8r+IolxbyJNPf/0tTAWHiOvyNwxSVxoccaTktLcSaYnhLGkJj2kuStPi09X77vwRJq7q0shYezZL99Jcy92myaM1do5W5r7vNcMYcxUJX7vAcDGXPzLLD5J/vV1+kP8vDe2iAv7AkDlT8U9bm07fi7NvY2CwlhpK/nn6vWG+cLYuhpfSHMH1BZfb2p+Tr4YaPBdd2Fscew2aa5/ncHC2OYz4qKiAPBbe/GlFREW4s8rAPxTRVxEd/8TecHS/4UHC2Ob7iZKc2sXsRfGSh1dIc01a/mZMBa2aKI0t+DE74Wx1j+clOau7V9XGHuTIP8e/RshLoQb9lpcuBkAfNxdhLFyPto74KSK3TRIGHvgXE2aW9hMfI30rGD5NddhbcQFlnvckhdIV5cVf/dT+bTwwO6p8qK4Mh3mrMtUMVtjluePpBEREVEuw9OdOuHxRiIiIqJcKM9M0nTpKtDm/w/h9uvXD4MGpRxGDg0NxeDBKacq+vbtC2dnZwQHi08DEBERURaoAJWpqd63/CTPnO58u6tA4cKFNV0FrKyscPTo0XSPP336NF68eJHmvnnz5qFMmTI5NWQiIqJ8SAVw4YBO8syrlJmuAgDQv39//PLLL2nuc3Nzy5GxEhER5WssZquTPDNJy2xXgS5dumDXrl1ITk7W6/m0dRyIeROr17aIiIjyCxUAlYmp3rf8JE9M0vTpKmBqaopWrVphz549ej2nto4DNtZWem2LiIgoXzEx0f+Wj+SJa9L06SoAAAMHDkTLli3RsKG41hARERHR+5AnpqSBgYGoXbs2gP+6CgwZMgStWrWCh4cHVq9erTWvYMGCqFWrlubncePGYe3atRgxYgQ2bdqUE0MnIiLKX1Qqnu7UETsOGJBPCw/s+lp7Ne3xr8TVrgFgUaVIYexFMXluwrJxwtj9Xt9Kc4utmSCMmY2SV+B3uREgjJkUkF8T+GTdT8KYc7UK0tyoW3eFMY97XtLca2PEnR8W3BNXBgeA4Q2LC2OmyfKq46pEcfxRovw0+cqT94WxSoXtpLlWZuJfaN0c5d0ZEm6eFcZUDbtIc6+/Fh+kPxIaIc0d7BgqjJ20qSHNLWhjLg5+3U+au7HbHGGsdXl55f/qJ34UxtaU+lSa62wtHnPnigWkuYW8xd/9hYvGSnMbFncUxioVkH8mzzwRdypp/OqMNDeptPj32YpLL6W5/W+sEcZmu34izZ1mekwY+9W5lTS3XyVxh4WfLkdJc3vXKCyMBYbK97dJCQdhrEDkLWnujgY9hbH9YeL3DwCcJV1OZl77Q5pr6l5fGgeAlp4tsDeD7hIybcf9L1d2HLhy5QoePHgADw8PWFtbIzo6GpaWljA3l/xeykCWTneGhobizJkziIiIwLtzvYED9W/5QERERHlYHrq2LDw8HGPGjMH169ehKAq2bt2K4sWLY9myZTAxMcG4ceI/pjKi9yRt06ZNWLBgAQoUKICCBQtC9U7/QU7SiIiISJu8VJR2wYIFKFKkCFasWKEpmg8A3t7emDNHfIReF3pP0vz8/DBixAj06tUrSwPIjISEBBQqVAhHjhxBjRo1sGvXLsyfPx+KoqBcuXJYs2YNrl27huHDhyM+Ph5Vq1bFkiVLMHnyZISEhODChQuoUaMGrK2t8fXXX2PAgAEoWbIk/P39AQBffvklLl68CEtLS/z2228ZlvEgIiKi/O306dP46aefYGNjk+b+YsWK4enTp1natt6TtJiYGKjV6iw9eWbp0lVg+PDhWLNmDYoXL47ly5djxYoVmD9/PoCUtlCpE7KoqCicPn0aXbr8d23NhAkTUKpUKQQEBGD58uWYMWNGju4fERFRvpCHFgAkJiZqvT8sLCzdxC2z9D4p3LFjRwQEiC8czw4ZdRUIDQ1F2bJlUbx4ygXeAwcO1EzK3uXg4ABra+s095UqVQoAYG5unu707bu0FbN9ExeX1V0kIiLK21SqPNVxoFGjRvjjj/8WVKhUKrx58wY//fQTmjRpkqVtZ2nhgJ+fH06ePImyZcvCzCztpkaOHJmlgWnzbleBd3f+yZMnKFr0v9V7ZmZmme4ooCgK5syZgx9/FK/WAlKK2fr6+qa5z6dF+vZTRERElJYqDy0cGDVqFEaMGIEePXogLi4OU6dOxb1792BnZ4fp06dnadt6T9KuX7+OChUqICkpCTdu3EgTy+golD7e7ioQERGBgQMHpusqULhw4TT3JSYmwiSTH4QZM2agW7dubLRORESUXXLhETF9ubm5YcOGDThw4ABu3LiBN2/eoH379mjbtq3mbJ++9J6krVy5MktPnFm6dhW4desWHj58iGLFiuGnn35Cq1by+jdv27NnDx48eICpU6dm454QERFRXpCYmIhevXph7ty5aNu2Ldq2bWvQ7Wf5eGNsbCxu3bqFW7duITY2+xqM69pVYOnSpejTpw9atGiB8+fP48svv9S6vUuXLsHHxwenTp1Cx44dAQCjR4/GxYsXoVarMXfu3GzbFyIionxNZaL/LRcxMzNDZGRktm1f744DCQkJ+P7777FlyxbEx6dUU7ewsEDXrl0xfPjwdNeo5Qc+Hk3w1xettcZue42S5q47La4qP7OZvNr59S/7C2Mr/rgizZ3x8rIwZq/IJ92zgp8JYxVd5ZXwm5dyEsai4+XXEf7yzz1hbIqnuzT3sz8uCmOLu1SV5hZODBfGboweKs2tOP9/wphiLj8cfivZWRir8OyUNFdlK65YnuRYTJpr+kq8dDz5+SNhDADib10Qxu75n5DmvpzqJ4wlfNJRmuu8dY8wVvHBYWnuBTfxNaUui4dLc+1LiqvK29esLc19fVX8mbTpM0WaaxL7Shj764H8V/viPdeEsZld5Z0dijtaCmNJ4+VlmU4OXSaM9Sos3h8AaPar+HO3uJe8M8uMveL93dnWWhgDgIEnxK/lilK3pbmfXC0hjK11CJLmWjZoI4z9O0t+rdPQGmOEMd+W5aW5X63+Rxgb9lE1ae4XDUtL4wDQ0ssT+35ekOHjRFp/MTFXdRz4+eef8fjxY0yaNAmmBq7/pvdMavHixQgKCsL06dNRs2ZNAMC5c+ewePFiJCcnY+xYeUsSIiIiyp+UXHZELCtu3ryJ4OBgBAcHo1y5cukqR2TlzJzek7T9+/dj5syZaNiwoea+li1bwt7eHt988817naRlVPS2evXq2LFjB65duwZ3d3dYWFhg9OjRGD16NEqWLAmVSoVNmzahYMGC720fiIiI8qw8NEmzsbGBl5e8b7S+slTM1tXVNd39rq6ueP1a3rw1u2VU9NbDwwNjxoxBv379MHfuXLi5uSEoKAiDBg3ChAkT8Pvvv2PTpk3C69mIiIgoC7KhCsT7kp2LDfWeylarVg0///wzEhISNPfFx8fj559/RvXq1Q0yOH1lVPQ2I69evYKdnfyaKiIiIqJUDx48wNGjR3H06FE8ePDAINvU+0ja2LFjMXz4cLRr1w4VK1YEkFI7zdTUFMuWiS8MzQkZFb0VWbVqFXbt2oX79+/jwgXxhc9ASseBBQvSXvhYrkwpvcdMRESUb+ShYravXr3CjBkzEBQUBHNzcwAppTlatGiBqVOnZumgj96vUoUKFbB9+3Z8+eWXcHd3h7u7O4YMGYK//voL5cqV03tAWfV20dvg4GAUKFAgXdFbkUGDBuHYsWNYsWIFli9fLn2sr68vwsLC0txsLMwNsQtERER5lypl4YC+t9xm/vz5uH//Pvz8/HD8+HEcP34cv/zyC+7fv5/uYE5mZalOhpWVFT788MMsDcDQdC16K+Po6JitdU+IiIjytVw42dLXsWPHsHTpUlSt+l8Zp2rVqmHixIkYNWpUlradqUnasWPH0KhRI5iZmeHYsWPSx2Y0EcougYGBmDBhAoD/it5+8cUXaNWqlWZ1p2hsq1atgr+/PxRFwZo1a3Jy2ERERPmEKk9N0hITE7W2f7KyskJSUlKWtp2pSdro0aOxb98+uLi4YPTo0cLHqVQqhISEZGlg+goMDEzz8+7duwEAnTp1SvdYPz8/zb/VajVu35YXJSQiIiJ6W7169fC///0Ps2bNQoECBQAA4eHhWLx4MerVq5elbevdcYDS8/Hxxu7NG7XGlp5/Kc0t6iCuOu9R0kma62gp/ouk2pd/SHPvzJJ8gF6GSXPXvBFfe2huIl9eXa+YozBWzslCmnv1087C2JC6vtLcw19UEsYeQjwmAHidIO6EEB6TIIwBwI9HxH8A/Op6WpqrklSwNnFwkebecm8pjLk7yl/nZnOOCGNLD8+U5rpu2i2MHa/cUBgDgPqXTwpjZx5HSXM/uLxaGDvbRN4VouyGr4Ux1w+6SnOjju0Xxg42GSHN/bCAeJ+SreWfyeR/dglj9+r0kOaeeSSu7t8lSvzeA4CqUmPxmGzEHTIAwDxc/F24Zye/njnqy27CWMFVm6W5jUdtFcZ+vfCDNLfFpiXC2LR/naS5XuXE9TafvIqT5qpLi7c959C/0lyfoT2FsbYHvpfmxpZpJIwdqSKOAUC70PPSOAC09PbC3t/07//dts+QXNVx4PHjxxgzZgxCQ0NRpEgRzX1lypTBokWL4Obmpve29b4mbdeuXWjVqhUsLNL+ok9ISMC+ffvQoUMHvQdFREREeVduXACgryJFimDDhg04efIkQkNDAQClS5dGw4YNocpiPTi9X6UZM2YgOjo63f2vX7/GjBkzsjQomYSEBDg5OWlKZOzatQvNmzeHh4cH+vfvj0WLFkGtVsPNzQ1NmjSBWq3G9u3b07VlOHnyJHx8fKBWq3Hu3DnN/VWqVMGOHTuybfxERET5Xh5psJ5KpVKhUaNG6NGjB3r06IFGjRpleYIGZGGSpiiK1gE8e/YsWwvBvt1N4OnTp5puAkePHsVnn32GMWPGICgoCG3atMHWrVsRFBQER8e0pwwURcHixYuxZ88eBAUFoVatWgCAW7duoW7duprr2IiIiCgbqFT633KZadOm4bfffkt3/4YNG7J80CrTpzv79OkDIGXWOGzYsDQd35OTk/HgwQM0aiQ/Z50Vqd0ERo0apVc3AQD4999/8fr1a3Ts2BFubm5YsWIFrK2tsX37dgwePBjfffedcBJKRERElOr48ePo2TP9NYD169fHr7/+mqVtZ3qSllq+4urVq2jQoEGabu/m5uYoUqQIvL29szQoGX27CbwtLCwMd+7cwalTp7B27Vr8+uuvGDx4MEJCQjB69GicOHECISEhaZrHv0tbx4Hy77GILxERkdHIpact9fH69WtYWlqmu9/CwkLrZWGZkelJ2qBBgwAARYsWRcuWLbUOLLu83U0gIiICAwcO1LmbwNscHR3RqFEjWFpaonnz5lixYgVevHiBkJAQtGvXDm/evEFkZKR0kubr6wtf37QrCX18sm9ySkRElBcoUOWphQPu7u4IDAxE//7909wfEBCAUqWy1i5S79Wd72P1piG6CQBA+fLl8fDhQyiKgosXL6J06dLYvXs35s+fj27dUpZ4d+7cOZv3hoiIKJ/KQ707P//8c0yYMAEPHz5E3bp1AQCnTp3Cnj17MHv27CxtW+9JWkJCAn766SccPHgQT548QWJiYpp4dhSzzUo3gdWrV+PgwYMAgJUrV+Ljjz9G8+bNYW9vj40bN2LgwIFYvHix5vEuLi64c+cOypQpY/D9ICIiyrdUyFOnO9VqNZYtW4aff/4Zhw4dApByMGjZsmWaSZu+9C5m+9133+Hw4cMYMGAA5syZg5EjR+Lp06fYtWsXhgwZgg8++CBLAzNGPj7e2L1pg9ZYpIl8xavsXVj5zwNp7timJYUxyxd3pblXVUWEsUq390lzk2PERThNq8qPaP7yQFy8t0+YuEBnRpb3XCqNdx4gLt5r+c0qaW5UnLiYbbn7QdLc+KriorKT9t2U5v6vrngBi2JqLs1VST5Yj21KSHPrfiZ+PRyLyf94OdfssjC2sKC4ICkATKwpvoRizFF5UegldRKFsXNm7tLc6vcDhLHo6u2kuXd6pu9okqrW/G+kuUp8rDB2u4h8EVbZGHFB0wtjJkhzKw8Wvw99HtaQ5q5v7SSMmcSJi+QCwHFFfPqn4aNAYQwAIv4+KoxdWntCmvvnNz8JY+6u8t/P304RF7sdPmmQNHeaqbiF4svGn0pzLUzF3/2bEeLPDQBYjhUXs/2tz0Jp7vgW4u/3p+vOSXP3DW0qjQMp/1fu2fx7ho8Tad/t01xVzDY76T2VDQgIwMSJE9GhQweYmpqicePGGD58OIYMGYKgoCADDpGIiIgod4qMjERkZKTm5zt37uDnn39GQID4Dz9d6T1Ji4yM1FwQZ2tri6iolKMq9erVw6lTp7I8MCIiIsqj8lAx23HjxuHw4cMAgBcvXuCzzz7D/v37MWPGDKxbty5L29Z7b4sVK4bHjx8DSGl/kDpj/Pvvv2Fvb5+lQYno223A3d0dnp6e6NSpE+Lj4wEAffv2hbOzM4KDgzXbnzFjBjw8PDBmzJhsGT8RERGltIXS95bb3Lp1CzVqpFwicODAAZQoUQJ//vknZs6cia1bxT1jdaH33nbo0AHXrl0DAPTr1w9//PEHmjZtioULF6J3795ZGpSIvt0GBg0ahEOHDkGtVmP//pRGyPPmzcPIkSM1237y5AkuX76Mo0ePIjY2FpcuXcqWfSAiIsrfVHnqSFp8fLymqH5wcDDUajWAlMUDYWFhWdq23qs7356INWzYEJs2bcK1a9dQokQJlC9fPkuDEslqt4HIyEikrpN4tyv96dOnNS+st7c3QkJCUK1aNeG2WMyWiIhIT3moo4+7uzu2bNmCpk2b4uTJk5p6ss+ePYODg0OWtm2wKWnRokXh5eWVbRM0IH23gSJFxCsT37Zq1SrUqlUL+/fvR+vWrbU+JjIyUvNiOjo6prkIUBtfX1+EhYWluVnbWEtziIiIKG8ZMWIEtmzZgi+++ALe3t6oVKkSAODIkSOoWrVqlradpUlaQEAAPv/8c3h7e8Pb2xuff/65phaZob3dbSA4OBgFChTQudvAoEGDcObMGRQtWhTPnz/X+hhHR0fN4oeoqCg4OTkZauhERET0thw83RkZGYmPP/4Y9vb2KFq0aJqaqIZQr149HDx4EAcPHkzTUL1Lly6YNGlSlrat9yTNz88P33zzDSpWrKhpkVSxYkVMmzYNfn5+WRqUNqndBvz9/XHs2DFs2LABfn5+iI1NqRVz7Ji4Fg0AmJiYYPjw4Vi2bJnWeN26dTWrMwICAtCgQQPD7gAREREByNmFA8OGDUNcXBwePnyIffv2Yfbs2di7d69B98fU1BQODg7w8/PDq1cptQKLFi0KFxeXLG1X70nahg0bMHHiRIwbNw5t2rRBmzZtMG7cOEyYMAEbNmgv6JoVgYGBqF27NoD/ug0MGTIErVq1goeHB1avXp3hNtRqNYKCgpCUlIRx48Zh7dq1GDFiBDZt2oQiRYqgUqVK8PDwgLm5ufR6NCIiIsqCHDqS9vr1a2zatAmzZs2Cg4MDqlevjoEDB+KXX37Jlt1as2YNXr6UF93ODL07DqjVaqxduxYlS6atdn/v3j307t1bc1QqP2nu6Y2xy37TGvv5WKg097su4vPWj6Pjpbm1j4urYVu2zGClrSKuom8aHS5NvWlfRRgrsHG6NNdJrf3aQACIv3VBmmvq3VcY81h6Vpq7YNdUYazhzM+luYqHuDr4zf4fSXPNF28Uxo7cjZDmvo5PEsbW7rkuzT3eQBwPqiyuSA4APpHHhbH7ZX2kucUvbxfGTAvIryVNfHxHGDMvX1uaezRZXM3+zCP5L84hbw4JY2aVGkpzEwuUFsZMYl5Ic99YFxTGHkYnSHNlHkbFSeMBN8Xf7+FNxF1MACAhWfzfxqZLT6W5k0eJexne7Sa/rtd82HxhbN+/8te5WynxOrm7X4+Q5pacvVwYS9ip/exMqivrxP8XRi/7Q5p77+UbYax6YXmpq37f/y2Mjfn5K2lu1XPi735I9cbS3GEv5b+TAMDHxwc7d+nfWcbH2xs3b6bt1JJ6Ru9dZ8+eRYMGDZCQ8N93adOmTfjmm29w9epVvccg0rx5c2zYsAHFixc3yPb0Xt3Ztm1bbN++HcOHD09z/86dO4UX5xMRERHpd3gohY2Njc6lLaKjo+Ho6JjmPicnJ80pydwuU5O0JUuW/JdoZobNmzfjxIkTmlODly9fxoMHD9Cpk7iPXVYkJCSgUKFCOHLkCGrUqIFdu3Zh/vz5mubq1atXx44dO3Dt2jW4u7vDwsICo0ePxujRo1GqVCnY29tj8+bNsLCwQN++fbFjxw7s3bsXjRr91x+vSpUqmDt3brbtAxEREeUMOzs7zaLAVC9fvsy2ovt//vknChUqZLDtZerk7pUrVzS3GzduoFKlSrC3t8fdu3dx9+5d2NnZoVKlSukOQxpKdhazBVKqBtetWxe7d+/OlvETERERkKwoet8yo0KFClCpVLh8+bLmvnPnzhn0uvPp06dj1/+fvnVzc4OpqSmAlKN406fLL/3JSKaOpK1cuTJLT5ZV2VnMFgC2b9+OwYMH47vvvoOiKFDloWJ7REREuUUWznZmiq2tLbp27YrJkydj3bp1uHv3Ln7++WesWbPGYM+xa9cuHDhwAFevXsXYsWNhYpJy/CsuLg67d+/G1Knia6EzotfqzsTERLRo0QL//vuv3k+sj+wsZgsAISEhaNy4MRo1aoSQkBDpNhcsWABXV9c0t7hY8UWeRERElCJZ0f+WWT/88APMzc1RpEgRtGzZEhMmTEDbtm0Nuj9Lly5FcHAwhg0blu70alboNUkzMzODo6Mj9FwYqpfsLmb74sULhISEoF27dti5cyd27twp3aa2jgOWVuw4QERElBFFUfS+ZZaTkxM2bdqE6OhoPH78GKNGjTL4/pQuXRq//vorzMzM0LdvX9y5I16lnhl610nr168ffvjhB0RHRxtkIBnJ7mK2u3fvxvz58+Hv74/Dhw+zwToRERFlKPXSKDs7OyxZsgSenp7o378/jhw5kuVt612C46+//kJoaChat26NYsWKaa4NS7V27dosD+5tgYGBmDBhAoD/itl+8cUXaNWqlWZ1Z7NmzaTbUKvVmDJlCpKSkjBx4kRs3rwZe/bsga+vL3bu3JmmVYSLiwvu3LmDMmXKGHQ/iIiI8jMF+p22zK3ePrqnUqkwYsQIVKhQATNnzszytvWepDVr1izDSZEhBQYGpvk5dQWmtlIZb7elUqvVUKvVAFJevL//TinwN3/+fMyf/19hxG7duqXZRnZVIyYiIsrv8tAcDStWrICDg0Oa+9q0aYPy5ctnuWCu3h0HKD1Pbx/8sH6L1ljwfXm1837O4uvrEgqWlebeGtFfGLP/TlzpHgCKPvlHGLs2e6E0t/zAT4SxzY5qaW6p4T2EsSqfNpXm2nUfLYwN3PtQmvtLndfCWFD3sdJc93f+UHjbm0Rx5wYAKP9YXMH7auEm0lwnS1NhzN5SfsWCXbT4c3V39hRp7m6/M8JY17Fqae7rL8Wfnd/PP5bmTqxtK4w1+/GKNLdaeXH1/p8aijs3AECwyl0Yq/m3vKr8fR/xZ7LZZ99Lcx/+OUwYi/xB/h4NdhF3wdjiJV+dHn9F/N3PyOTkFsJYj9rFpLl134j/00pyKSHN/S1UHOtdQt6Z5eLgIcJY1TXrpLmPkmyEsQcZdHYIjxF3jbCzEH+3ASDstXifMsqt7SauA9ZgsJ80N3SglTD2dVwjYQwAFnTMuLSFt48PNm7dkeHjRHp+9AEOHDigd74x0ftIGhEREZE+eHxIN3pP0urXry+tI5ZRCQt9ZGfHgdDQUDRp0gQVKlRAQkIC1qxZgwoVKhh8H4iIiIh0ofck7bvvvkvzc2JiIq5fv46dO3di4MCBWR6YNm93HChcuLCm44CVlRWOHj0KDw8PjBkzBv369cPcuXPh5uaGoKAgDBo0CBMmTMCiRYuwf/9+dOjQAfPmzUu3KKBTp05YsWIFTpw4gdWrV2PevHnZsh9ERET5mfwCEUqVpYUD71Kr1Shbtix27NiBDz74IEsD0ya7Ow6kevXqFezs7AwzaCIiIkqDZzt1Y/Br0ipWrIgzZ8QXHGfFux0HmjSRX3CdatWqVdi4cSOsrKwwZYr4QtwdO3bg6tWruHHjBo4ePSrd5oIFC7BgwYI095UtV06n8RAREeVneakER8uWLVGnTh3UqVMHdevWRTkDzgX0LmYbGxub5vbmzRs8ePAAP/30E4oVk6/u0Ud2dxwAUk53Hj58GIGBgVi4UL6yUVvHAStr8QogIiIiAqDkbMeB7DZ58mQULlwYO3bswKeffgpvb2989dVX2LBhA65du5albet9JM3DwyPdwgFFUeDq6opZs2ZlaVDapHYcqF+/PuLj49G6dWtYWlqic+fOsLKywrFjx6R1297uOJDR+BwdHREZGWngPSAiIiIFeeuatLfrsUZHR+Ps2bM4ePAglixZAkVRsrSQUu9J2ooVK9L8bGJiAmdnZxQvXhxmZoav7JHdHQfq16+vWRkaHx+fpvsAERERkcjz589x5swZnDlzBqdPn8ajR49Qu3Zt1KlTJ0vb1buY7ZkzZ1CjRo10E7LExERcuHAhywMzRrXqNsSV5KpaY9E/tJPmvjz5tzBmNlB+5M8+4l9h7LZVKWlu2ejrwljsaXHxVgC4WF+8irf27V3SXNRuLQwp5uJCigCgUsR/g5ncCpbm3ioiLpR7J/KNNLeNmbhhbnyovKq0eZkqwlji/ZvSXLNSlYWxJEf5pQXJIeKCkU/qi4sRA0Dw/Shh7MOS8kKayoWD4mDNltLc5xAXs3WxkP+6Mr0h7uH7Y0xFae4XdYoIYybRz6S5L3/7ThizKS5epAQAFk27iJ/35RNpbqBpJWGs2fU/pbn3GvQRxorbm0tzzaOfCmNbH8lzu5QS/xFvGiXf3yiX8sKY/Y1D0tzkCuJrmI88FhecBYCqhcSXsrg+OCnN3Zosfo/qrx0vzXVtoP3/EwAw9xIXMgYAxdxaGGv582Vpbtg9cfH120e3S3Pjz2bcrcfL2wdr/vwrw8eJDOjRJVcVs/3www8RHh6OatWqoW7duqhTpw6qVasGc3P5d0EXel+TNnjwYERFpf8lHh0djcGDB2dpUERERJR3JSuK3rfcJvVaOZVKleZmCHqfl0wd0LuioqJgbS2ewRMREVH+lvumWvrbtm0bwsPDcfr0aZw5cwazZ8/G48ePUb16ddSpUwcDBgzQe9uZnqSlXhemUqnw7bffwsLCQhNLSkrCzZs3UaNGDb0HlFmG6kKQkJCATp06ISEhASVLlsTatWthYqL3gUYiIiISyEslOACgYMGCaN26NZo1a4azZ8/iwIED8Pf3x6lTp3J2kpZ6lExRFFhaWmqKyQKAmZkZPvjgA3TpIr7GwtAM1YWgZcuWWL9+Pdzc3PD1118jKCgIXl5eObYfREREZHyOHDmiWTBw48YN2NnZoVatWhgxYgTq1q2bpW1nepI2depUAICVlRWGDRsGW1vxRb45wVBdCCwtLTVdCMzNzQ12PpmIiIjSyoWXlunt22+/Re3atdG+fXt88803KFeu3Pu9Ji0pKQlbt25F9+7d3/skzdBdCJ49e4aAgABMnjxZmq+t40CRYiUzvwNERET5THIeuiotO1ea6nXRlampKUqUKIHXr18bejyZYuguBElJSfj888+xbNkymJrKSwxo6zigMpHnEBERUcqRNH1vuVFcXBz++usvLF68GIsXL8b27dsRFxeX5e3qfWX8sGHDsGTJEoSGhmZ5EPpK7ULg7++PY8eOYcOGDfDz80NsbCwA4Ngxcc0kIG0XAgCYNGkSunXrhurVq2f72ImIiPIjBSkLB/S95Ta3bt1Cly5dsGzZMty8eRM3b97E999/jw8//BC3bt3K0rb1LsExc+ZMxMTE4OOPP063gADI3sN/qQzZheDhw4dYunQpGjZsiNWrV2PChAlo06ZNtu8DERERGa+FCxeiZs2amDp1qmYuFBsbi+nTp2PhwoXpOjRlht6TtFGjRun9pIYSGJi2Iv7u3bsBpDRKf5efn5/m32/32VKpVPj775Rq/2/eyCvOZ8TFyR6Xfl+sNZZ4Yo00V1aV3OK1vAo3XoYJQ4VcykpT90WIr6P72y796/i2r2+JK0+bVBFX9geAGHN7YczSVH7B5Z9Xw4Wxaq6NpbnhUbHCWKuES9LcRMlrGVG3mjTX1lx80Pos5O9RpT+nC2P2pYtKcy82/EIYqxdzT5pbX3KNpWIaL839yUL8/ptck3/PetVwEMZMr8j/+DN1dhXGviwmP3GQFOgnjKkvVJDm7nYRd8m4739CmmvvM1QY23XfQhgDgH5xAeLnbSjuKAAAhbaIO5lYtfpQmht34bgwVrF+f2lu8rHfhLHI6/LuG5c/mCKMlfx1nTS3RFdxB42AVzWluZ424t9Jnv7yz1V05AVh7HKg/P193VNczurRom+kuYc+miGMHRgg7mQAAPae4k4IK++K3/vMyK2nLfVx8eJFrFu3Ls3BKisrKwwYMAB9+si/hxnRe5LWoUOHLD0xERER5U95aeGAjY0NwsPD4e7unub+Z8+eZXlxZZaqtYaHh2PdunWYM2cOIiMjAQBnz57FgwcPsjQoIiIiyrvy0sIBHx8fzJgxAwEBAQgPD0d4eDgOHjyImTNnomVLea/ijOh9JO3SpUsYOnQo3N3dcfXqVfTu3RtOTk44ffo0/v33X8yZMydLA9OVoToOWFhYoHv37njy5AnMzMywadMmuLi45Mg+EBER5Se5sQenvkaNGgWVSoUpU6YgKSkJQEoVjA8//BAjRozI0rb1PpK2ePFi9O3bF2vWrEnTGqpRo0a4cEF8Dt7Q3u448PTpU03HgaNHj+Kzzz7DmDFjEBQUhDZt2mDr1q0ICgqCo6MjBg0ahEOHDkGtVmP//v0AgN9++w2HDx9Gv379sHHjxhzbByIionxDAZKS9b/lNpaWlhg3bhwCAwOxfv16rF+/HoGBgfD19YWSxcmo3pO0GzduoHXr1unud3FxwYsXL7I0qMxI7Thw9uzZLHUcAFI6DQApqzIqVaokzVuwYAFcXV3T3N68icni3hAREZExsra2Rvny5VG+fHmYmprit99+wwcffJClbeo9SbO1tUVERES6+2/cuIFChQplaVCZ8W7HgSJFiuiUt2rVKtSqVQv79+/XTDajo6PRtGlTLF26FBUqyFdzaStma21tk+X9ISIiystS6qQpet9yi/j4eCxbtgy9e/dG//79cfDgQQApNVw7deqEjRs3olevXll6Dr0naa1atcL333+PFy9eaHpUXbp0CUuWLEHbtm2zNChdGbrjgJ2dHf7++2/MnTsX3333XXYOnYiIKJ9SkKTof8stVqxYgb1796Ju3bp4/fo1Jk+ejG+//RYbNmzAmDFjsH37dvTu3TtLz6H3JG3o0KEoVaoU2rVrpylq+9lnn6Fy5coYMGBAlgalK0N2HFAUBQkJCQAAR0fHdMV5iYiIyDDywpG0gIAAfPPNNxg1ahQWLVqE5ORkJCcn4/fff0erVq0ybC+pi0yv7kxOTsbatWtx5MgRJCYmom3btvDy8sKbN29QoUIFlCpVKsuD0pUhOw7Ex8ejdevWUKlUsLCwSFP8loiIiAwnNy4AyKynT59qaqMVL14cFhYW+PTTTzVnFw0h05O0X375BWvWrEG7du1gaWmJvXv3QlEUTJ061WCD0pWhOw4EBQVlaTwmKgWlLBO0xhTvftLcsDjxjNvOQn7A0+r2TmHsiEkVaW4Ha3FNu1ZV5UcTVzxQC2NNk52lufYxicLY42h5NfsWW8WfNY+H3tLc00s6C2MnwuVVxxslRwpjdhbyv5jiJb+RGrskSXNDuon398qzaGluOwdxRXP30aeluVeXuwtjPX+/Is1dY7JLGLPx6SHNvflK+3cIAMoVky/oufs/cZV1t5k/SXPNaos/O7uaF5fmWh25IYyVV8sLfydYiI8MfFZa/j/ZM4uOwliZyOvS3MSe44Qx5ZW4iwkAxPqIO1k4x8s/z8cqfiyMqYv+I8193rSFMFbonLyzg8rytTA2M/qZNDfx2hlh7FCrYtLcKQ/EBy7WD5dfVqNKFF++0/p1F2muun5zYaz7jq+kuTInX4i7tgCAvN9E3pKcnAwzs/+mUaamprC2tjboc2R6krZ79258/fXXmr6Wbdq0wcCBA/H111/DxCRLtXGJiIgoH8hNpy31pSgKvv32W00Zsri4OMyfPz/dRG3u3Ll6P0emJ2lPnjxBnTp1ND9Xq1YNJiYmePbsGQoXLqz3QPRlyGK2t27dwvDhwxEbG4vx48ezwToREZGBKUCuWgCgr3fbY7Zr187gz5HpSVpSUpKmnlgqU1NTJCaKT19lp7eL2RYuXFhTzNbKygpHjx6Fh4cHxowZg379+mHu3Llwc3NDUFAQBg0ahAkTJmDRokXYv38/OnTogJkzZ2Ljxo1wdHR8L/tCRESUHyQb/xwtRy7zyvQk7d3De4D2Q3xZObyXGanFbEeNGpWlYrbx8fG4f/8+evXqBTMzM/z0008oWLBgdg+fiIgo30nKC7O0HJDpi8g6dOgAR0dHWFtba27t2rWDi4tLmvtyiqGK2YaHh+PixYtYt24dhgwZgoULF0rztXYciHljiF0iIiIiyvyRtPexilPk7WK2ERERGDhwYKaK2Y4bNw5du3bF8+fP4ejoiJo1a8LJyQnNmzfHjz/+KM339fWFr69vmvt8fOQrC4mIiChvLBzICUa9HNOQxWxtbW1hZmaG+Ph4XLx4EaVLl86BPSAiIspfFAVIysItP8n0kbTcxJDFbJOSkjBq1Ch4eXnB3Nwca9euzYldICIiynd4JE03KkXhK2Uo6nr1MFPQTurw9F+kuROKi4spjrliL81dVEtc/PNFAXnxT/skcTFUxdJOmhs6/FNhzNRKXEQVAGxnil+PQrcPS3NfVfAUxixM5ZWeE9aJi53alK8szX1e+0O9n9fcRBx/NukzaW78pJXCmAry57U1Fx8sL/bqpjQ3+dl9Yaz4nDvS3Lu/iVvDRcprFePlOHGvu+ivV0tzXW3NhbHCieHS3Iufi8dccv12aa7Tm6fC2PhgcRFVAOhSXXwdbWPTh9LccMdywtiJB1HS3AZbxJeu/NhgtDR32NllwpjLkGnS3CgTG2HMMfGVNHfNTXEh1c6VCklzh2y+JIx9uWioNLfmCfHvJAeV/ANt8kr8u93kzQtpbpJTCWEs2Ur+/4LqtLig9HZHcaFbAOjV72thrHSz9MXi33bj+87SOAA09/TG6O9/y/BxIj+O7oMDBw7onW9MjPpIGhERERkfHknTjVFfk0ZERESUVxn9JC0hIQFOTk64cOECgJS6ac2bN4eHhwf69++PRYsWQa1Ww83NDU2aNIFarcb27dvh7u4OT09PdOrUCfHxKYerv/vuOzRq1Aienp54+lR8+oKIiIj0x4UDujH6SdrbHQeePn2q6Thw9OhRfPbZZxgzZgyCgoLQpk0bbN26FUFBQXB0dMSgQYNw6NAhqNVq7N+/HwCwdetWBAcHY/jw4fj999/f854RERHlPQpSTnfqe8tPjH6Sltpx4OzZs1nqOAAAJUqUQHx8PF6+fAkXF5dsHTcREVF+lZys6H3LT4x+4cC7HQeaNGmiU96qVauwceNGWFlZYcqUKQCABg0aoFKlSrC0tMSpU6ek+QsWLMCCBQvS3OdevLh+O0FERJSP5LfTlvoy6iNpb3ccCA4ORoECBTLVceDMmTMoWrQonj9/jqioKOzcuRM3b97E3Llz8f3330vzfX19ERYWluZmZWLULycRERHlIkY9qzBkxwGVSgVbW1uYmprCxcUFL1++zIldICIiymf0vx6N16QZkcDAQNSuXRvAfx0HhgwZglatWsHDwwOrV8sLXwIpHQeCgoJgY2ODWrVqoVmzZpg4cSIGDhyY3cMnIiLKl5IURe9bfsKOAwbkofZC/wV+WmM25qbS3NYnxRW8LXp/I8099zRGGCvmIK/8/youWRgrsklcnR8ArneaJIy9jEuU5j55FSeMJSSLxwQAQdfEFbx/6lZNmhs6sKswNrH5FGnuws5VhbGzj+WV0oPrituTTVkk7mQAACGtxwljD6PEFdgBoF9x8esc7SC/hvJ5jPg9PPdU3KkCAEo6au+8AQDVC1lLc5N3LBbGzH36SnPDl38rjB3/QFxFHQBsP2gnjDkEBMifN0Zcdb5NSfFrAQAxKkthzP7OCWnumzNHhLFhZvLK8L9ILt99slbc5QIAdrURf/et27WV5nZd9bkwZlGprjT3RoE6wtjN5+LfgwDQtISDMKasln/3F1caJIxNryb/ffVqj7iyvm3X4dLcgPriz6Tnj0OkuU/qfyKM9VwVIs098qmrMFZsVKA09+nWkdI4ADRTe6Hf/DUZPk7kj8kD2HGAiIiIKDtw4YBujPp0JxEREVFWHTp0CJ6ennB0dISbm9v7Ho6G0U/S3u044OLiAi8vL7Ro0QI///yz1scFBgZCrVajUqVKqFKlCtRqNQ4fPgwPDw94eHjgq6++el+7Q0RElOfltoUDtra2+Oyzz7Bo0aJs2b6+jP5059sdB2rUqIEGDRrA398fiYmJ6NmzJ6pXr46GDRumedyUKVPg5eUFPz8/WFlZoUePHoiMjMSuXbvg6OiI3r174/bt23B3d3/fu0dERJTn5LYFAA0aNECDBg0QFBT0voeShtEfSXu748DbzMzMMHbsWOzZs0f6uFROTk5wdHQEAJibm0OlUkmfd8GCBXB1dU1zi4t9Y4A9IiIiyrsUAEnJit63mJiYdP//vltcPq8w+iNp73YceJubm5umUfq7jxOdc7527RqeP3+OMmXKSJ/X19cXvr6+ae7zUHtlYU+IiIjyh6QstHeysbFBWFiYAUeTexn1kbR3Ow7s3LkzTfzx48dwc3PL8HGpXr9+jWHDhmH58uU5MXwiIqJ8KStH0jKra9euUKlUwltuZtSTtHc7Dhw8eFATS0xMxJIlS9CuXTvp4972xRdfYPLkyShatGhO7QIRERFlo82bN0NRFOEtNzPqSdq7HQeio6MREhICLy8v+Pj4wMvLCw0aNND6uJiYtIUPT5w4gV27dmH69OlQq9U4d+5cTu8OERFR3qfk7JE0XSQnJyM2Nhbx8SmFqWNjYxEXJy4GnlOM+pq0wMC0lY93796dqcf169dPc1/jxo0RGRmZpfFYJsehe+jvWmP/1BFX2QYA+/Z9hLFERV69v7CtuTD29z15D9KPi4orpSdVrizNrW/3Whh7snqaNNf7q3nioCKv4L31lHjMZpDnOlcoJoxtayeuSA4Az63EXSNiE5OkuQvu7xPG3uxbJ80t4SCuWO8TfliaGxStFsZaWEZKc2N/niWMVeo/W5q71V1cOd7639PS3GLtJBXL/b+X5hZqL66y/1vwPWnuT+f+Fsaej/5UmjuqhLj6e5ux9aW5YWN6CWMWLRtKczdVEf9eWVnFWZp7sFYLYazE0UPS3FZm4r/vB8/4WRgDAPuK4lXz7V8FS3NbTFwljD3+QVydHwCmHH0hjM1o3FyaO/HWn8LYuDPyjiGyPhc7HklTUeW4+Pv9wZZL0tzDX4tLSf20Ut5hIe6EuEvC4/me0lxdpC4cyE2OHDkCT8//9s3a2hqlSpVCaGjo+xsUjHySRkRERMYnt03S1Gp1rjz1adSnOwHDFrNt3rw5WrRogc6dO2sOeRIREZFh5bbTnbmV0U/S3i5SC0BzDVpAQAD279+PkydPpnucl5cXgoKCMGHCBHzzzTcICgpCixYtEBAQgMOHD6Nu3brCxQVEREREOcHoJ2mGKmYLpBSxBVIuIGS3ASIiouyg/1E0HkkzMvoWs9Xm1q1baNCgAQ4cOJBhGQ5tHQdi4niKlIiISCarHQfyE6OepBm6mG25cuUQEhKCTz75BL/++qv0uX19fREWFpbmZmNpYbB9IyIiypNyYQmO3MqoJ2mGLGabmJioWdnh6OgIKytx2QMiIiLSX2KyovctPzHqSZohi9k+ffoUarUanp6e2LJlC3r27Jnj+0NERESUyqjrpBmymG2xYsVw+LC8MCgRERFlXX47bakvo56k5TYqUzNYVmmgNVZ65RhprvLJJ+Lt2rtKc62XTxTGunfuKs09ElVbGHtZtL00d95KccXro8MmSXMTDm8QxkwsraW5333UXRhzaDpcmvvaX1xpO2bPGmmuVTfx69wjMkCaO/puHWHsW8l2AaByxL/CWEAhcdV4ACjyTT9h7NEPf0hzi302Whhz/Psnae6kU5J4Qqg0N6TDMGHMobi8K8TZseIj4PfuXZbmPohKED/vXHlXiGOXtwtjv19/Jc1VO9kJY2Ge4k4GANDLRNz1IyJBfqJk+mcLhDH/vQuluS/vPBbGtkyQd4Xw/1dc+V99SP7+HvvpC2HsZIz4/QOAX34SX2c8ZcOX0lzcuiAMzXcPk6ZGPRJ3QbkeFi3N7Vhc3OVka3/x7xQA6Jgkft66ReWvs2mEmzAWX6ymNFeXC4VyY8eB3IqTNCIiIspRSbmwun9uZJTXpOnSZSA0NBRFixaFp6cnfHx8EB4eDgDw9PTE5MmTAaTUQ3Nzc8PGjRvx6tUr1KtXD05OTu9ln4iIiPILru7UjVFO0nTtMtCpUyccOnQIQ4YMwcaNGwEAlpaWuHXrFgAgJCREUw/N2toa+/btQ6NGjd7DHhEREeUTLMGhM6OcpOnaZSBVZGRkmsapZcqUwb1797Bz50507NhRk1ugQIHsHzwRERGRDozymjRduwzs2LEDZ86cwYsXLxASEqJ5TMeOHbFz507cvn0brVu31msMCxYswIIFaS+8LV+WraSIiIhkuHBAd0Z3JE3XLgNAyunOkJAQeHt7IzQ0VPOYxo0bY8uWLShVqpTe49DaccDKUu/tERER5RdJycl63/ITo5uk6dpl4G3jxo3D//73P83PJiYm6NixIz6RlL0gIiKi7MAG67oyukmarl0G3ubu7o7w8HC8ePFffZ7Ro0ejZs209V5at26NU6dOwcfHBzdu3Mj+nSEiIspn2GBddypFYbESQ2nppYb/z9oLRCa/kBc8jK7oKYxlNJO2iXogjH26L1Kau7hzFWHs+ZskaW68pFhieEy8NHffNfHr0aN2MWluDRdxgUeTS4HCGACYFSwijN1b9aM0t1h3cRHdq0WaSXN9t4kL/6558Zs09+WAecLY+ery1cg1LwYLY7MOyP8Q6VRD/Fp1fhMijAGAUlUtjG2+JS7ACgD1JIU2l7jVkOZOfC5+ne0s5N8kO8n3aPSJN9Lc8v27CWNfPD0vzU3ev0oYS/AeKM012ysuHBsXHiHNtSlfSRxsLC+CHZMs/g4+ipYXlQ2oWF8Y+2LzOGlu/8fiQqq/eUhTUXqq+LPRp6f8e9S0jIswtuGf+9LcWVeWCWO/+ciLfrcsX1AYs7MQvwcA8Pe9SGGs5x1xMXEAiIsQ526pKy+wPLBBxpcRNfBQo8zn8zN8nEjE+sk4cOCA3vnGxCgXDhAREZHxym+N0vXFSRoRERHlHIWrO3WVJydpCQkJKFSoEI4cOYIaNWrAxcUFtWrVQlJSEnr37o0BAwYgISEBPXv2xLNnzxAfH48NGzYgKCgIVlZW6NGjx/veBSIiojyLkzTd5MlJ2tsdCWrUqIEGDRrA398fiYmJ6NmzJ6pXr46wsDA0bNgQX331FeLj45GUJL/+ioiIiLKOddJ0Z3SrO3WhS0cCa2tr/PPPP3j27BksLCxgbW2dqedYsGABXF1d09xi3sQacjeIiIgoH8uTkzRdOhJ4e3ujXr168PLyQocOHfDq1atMPYfWYrbWVobcDSIiojyJJTh0k+cmabp2JFCpVPD19cXFixfRunVr/PabvAwCERERGQYnabrJc5M0XTsSPH78GLGxKacnCxUqhOR81mqCiIjofVGSFb1v+Umem6Tp2pEgNDQUnp6eUKvVWLduHXr16gUAmDVrFnx8fODj44PXr+VFN4mIiCiTFCA5WdH7lp+w44ABefv4YMOW7VpjTibyCvzRv84WxvybjpTmdi1nK4wdeSyv/n364UthTF2mgDS39u1dwtj2At7SXNn37OnrOGluW0kV7kI28gXLtrHiKuyLLsqryo+M3ieMLXNoI82tX9xJGLMxl/+tVDfxX2EsoUg1ae6Ppx4JY3WKOkpzbz4X/5HSvrz8s3GudlNhrNllcRcEALCMjxLGHs4YLc0NGymu7m43ubc0N3zGr8JY8L0XwhgANJC8v8UWfCHN/XvgYmEsLFr+XehU2VUYO/tYfp1t67LOwphdBuv+743tK4yZ28ivzT3ea64w1r2g+PcRANxbNEsYS3gtX7jlVKGEMBZ+QfwdA4DSHcXtDB4cOCnNDbv4VBhzLCX/Dp4ft1IY+2HPNWnuvlHi7+Cyk+LuGgAw6MJyYazehXrS3LtrPpXGAaB+UzVsu07P8HEi5ntm5puOA3nuSBoRERFRXmB0k7SEhAQ4OTnhwoULAAAXFxeo1Wqo1WrcuXMH0dHR6N+/P9RqNZo1a4Y1a9YAAEaOHIkWLVqgWbNmuHXrFgCgTZu0Rz++/PJLNGvWDN7e3nj8+HHO7hgREVE+oED/69Hy2zVpRlfMVlSoNtXYsWPRvXt3tGnTBoqiIDg45dTKwoULYW5ujqNHj2LVqlWYPz99c9cJEyagVKlSCAgIwPLlyzFjxowc2y8iIqL8Ir9dW6YvozuSJipUm+rKlSuaI2QqlQqNGzcGAJibmwMAoqOjUa2a9ut4SpUqpXmsSqUy9NCJiIgIgJKs/y0/MbpJ2ruFakNCQqBWq/HBBx+kedzDhw+hVqvRunVrzX3t2rXDsGHDNKs/tVEUBXPmzEG/fv2k49DWceBNTEyW9o2IiCg/UBRF71t+YlSTNG2Fahs0aICgoCBs3552VWWxYsUQFBSU5ojYnj17sGPHDkybNk34HDNmzEC3bt1QpkwZ6Vi0dRywtrHJ0v4RERERpTKqSZqsUG2qKlWqYO/evQBSZuqpjdPj41NKYDg4OAj7dO7ZswcPHjzAZ599lk17QERERKyTphujWjgQGBiICRMmAPivUO27nQKmT5+OYcOGYd68eTA1NUXXrl0BAJ988gkiIlJqZP3www8AgPv378PHxwcA8NFHH2Hx4sVwdnaGWq1GmzZtNM9FREREBqIg363S1JfRTdLetnv37nSPsbOzg5+fX7r7t2zZku6+y5cvp/n5yy+/zNoAiYiIKEOcpOnGqCZpuZ0q/g3sA1dpjQ2KaS7NXdPmI2Gsh6m4YjUA3I+vIIzN3nNRmju2lTi3xK/yI4mPh/1PGPso8oo0N2LHRmHsQbep0lzL70YIY/ZDx0tzk63shbGvCt6T5ioVWwljI2zklcP/fib+hVTTXt4VYqC/qTB266G82vmavnWFsYy6MzS1jRQH34g7GQBA60O/CGNJ59P/cfW2O34bhLHii9dLc8X19wHrb6dJcwMTxcvGqhUWf24AoFmBJGHsRgYdNI7dChfG5jwUd0EAgN12Y4WxrlUKSXNtIm4LY789le/vw1/PCWPDwi5Icz80Fa+c/zfKQZpbetr3wlipAb9Lc8c1bCuMlW0rv564bClx/GEt+eUxZZzFHRhMM6gi8OFB8f52V5eV5o6wryKMjXwsf4/KzhSP+e/1HaS5ukrOZwsA9GVU16QRERER5RdGNUl7t9vA2x0Dpk2bpilcu3r1ajRv3hzNmjXDpEmTMG7cOKjVak13grZt28LPzw9VqlSBWq3GF1+k9Nbz9vaGnZ0dnjx5kvM7R0RElE+w44BujOp057vdBrQ5f/48Dh48iKCgIJiYmODo0aPw8EhpjtumTRtNdwI/Pz9888036NGjhyZ3w4YNGD9efrqMiIiI9KcAQD6bbOnLqI6kZdRtAAC2bduGUaNGwcQkZddSJ2i6KFy4cJbHSERERHIswaEbo5qkiboNqNVqzYrOJ0+eoEiRIjptb8aMGVCr1Vi+fHmmx6Kt40BMnPwCcCIionxPYccBXRnN6c63uw1ERERoug2knr5M7SLg5uaGR48eoWTJkhlu893TnZnh6+sLX1/fNPf5NG+q17aIiIjyk/zWg1NfRnMkTZduAwDQpUsXLF26VFPk9tixYzk5TCIiIiKDMJpJWmBgoKYxuqjbAADUrFkT3t7eaNGiBZo1a4Y9e/bo/Bx9+/bF/v378fHHH+Po0aMGGzsRERH9h9ek6Ual5LcTvNmohZc3pq7UXqS12IIvpLnlJn0tjCU9CZXmmhQTF6Td/7qgNLdBUTthLCwmUZrbYpD2wr0A8PfP8v0tZf5GGIvymyvNdfmgpzA2/aa8CGerbwcIY7VHdpTm7izfSxhrvmumNNexYmlhrOFJcdFJADjZUFwY+Il6iDT35UBxkeRizSpLczFwtjB0Re0lTXXY4S+MXXz6Sprbo0oBYezT3y9Jc9eXEBdv7nG3mjR3UzsnYSz55ilpbliNzsJYTIL8vE5pC/F3QbGQF1l9teobYczZS1y8FQCSS2hfIQ8AZ6IspLmTdlwWxvZ1lS/ASrYWF342i3wgzb02VVzo+u6k1dLcsw9fCmMfVnWT5pa5lL5jTaqrlbpIc4/cjRDGelaXv1b2puLPjknMC2nukyXThLEaISWkueGrPxbGig7fJ80N+2uMNA4AdRs3R1zTcRk+TqTI+SU4cOCA3vnGxGiuSSMiIqK8Ib/VO9MXJ2lERESUo9gWSjdGc02arvTtSjB8+HDNIoOAgIB0KzeJiIiIclKeO5Kmb1eCihUrom/fvti7dy8WLlyIDRvETZ6JiIhIPwp4ulNXee5Imr5dCVxdXdGgQQMMHjwYLVq0gLOzs/R5tBWzjXsjvgCYiIiIkFLMlr07dZLnjqSJuhIAQGhoKNq0aSPsSvDVV1+hZs2auHxZvHIplbZiti28vA2yD0RERHlZfiuloa88dSTt7a4EwcHBmq4EQUFBCAoKQr9+/QD815XgXfb29qhQoQKsra1zeORERET5R25rC7VgwQJUr14d9vb2KFmyJCZPnoykpKRsea7MyFOTNHYlICIiosxKTk7GmjVrEBERgaNHj2LXrl1YuHDh+x5W3pqk5URXAiIiIsqa3HZN2vjx41GvXj2Ym5ujVKlS+PTTT3PFARx2HDCglp4tsHep9grgq6LLSnMHqMQLHVQlq0pzH/xvhjDmpm4kzTVp/okwFvA4g44DJR2EMcvwW9Lc38LEVccL2mRQ7fzX08JYiOdDaa7DxGBhLGbtp9LcR0UbCmM25ipprl3w78JY7L070tyR1h8KY78v/EGaG7NTXEpGVvkdAG6PE3cz2PfZImmuuYn4778qruIuFwBwpJr4M1vCWn4ZbbcH4u9RRsyOa+8WAgDm5WpJcxMLlBbGDtyPk+aam4o/Oy2KWUlzu60Xd6MY7VVOmrv1QvpLPlItaiHu+gAALy3EC6ui4+UdFv55GCWMfVBavr+vVOK4Q6K8k0WS5PP+cqm8+n3Y2X+FsUI1ykhzCwwQfwdf/Sn//j7tOkUYc5d/jfA8wVQYi4qTv0eOluLvb/SkftLcCss3S+MAUKdRc7yoMSLDx4kUvfwdbt68meY+bdeJZ0X79u1Ro0YNzJkzx2Db1EeeWzhAREREuZuSrP/1XjY2NggLCzPgaNL6/vvvcfHiRaxduzbbnkNXeeZ0p6iIrZeXF9RqNdRqNbZt2wY/Pz8UL15ckzdu3DjNYydNmoRChQph40bxX9RERESUBYoCJTlJ71tmde3aFSqVSnh727p16zB79mzs378fBQrIjyjnhDxzJE1UxNbCwgL+/v81e/bz80OhQoVw5swZ1KlTB/fv39fERo4ciQoVxM3KiYiIKOuyciQtszZvzvgULACsX78evr6+OHjwICpVqpTNo9JNnjmSpksR21SdOnXCrl27cPPmzTSTssKFC2fnEImIiCgX+v333zFq1Cjs3bsX1apVe9/D0cgzk7R3i9imio+P15zuvHMn5QLtypUr49q1a9i5cyc6dOig1/Np6zgQEyu/QJiIiIgAJSlJ71t2mDRpEiIjI+Hh4QE7OzvY2dmhbdu22fJcmZEnTne+XcQ2IiICO3fu1MTePd15+PBhAEDRokWxd+9ejB49Wq/n1LaSpKVnC722RURElF8oUHL0dKcuUg/i5DZ5YpKWWsS2fv36iI+PR+/evTPM6dmzJ0qWLJnuokEiIiLKXrltkpZb5YlJWmBgICZMmABAXsT2bXXq1EGdOnXS3Ldo0SKsXr0aJiYmCA8Px7Bhw7JtzERERPmSwkmarvLMJO1tu3fv1vz77VOdADT9O9+W+pgxY8ZgzJgxhh8gERERUSax44ABtfTyhP8v2iuxb4goKM1tUsJJGLsfpf+CBBdrc2k8NlH810wt21hp7qN5E4SxwG4zpbkNiourfztbiStlA8DiY6HC2OhmpaW5AXdeCGM9nZ4IYwAQX0T/FT/7b0cKYx1s5c97xcJdGLMxl6/9iYoTv7/Xw19Lc80klfA746o0V4kXf3bu/7Zemps4cYUwduFptDS3iJ2lMNYo/po0d58irtBfwkFeCd/VVvz3rqOF/D369cJTYaxiQXlZ+WRJi5ztlx5Lcyf9u1oYK9RZ3IkEAM59pb2zii4+ryq+DrizWvxZB4AeNYoIYzY/yP/ALtSkgTA2W9Vcmjvwb3EPR8fx30tzbR+cEcZaH5D/97vixjJhzLlCCWmuU8vOwtjUGn2kuRNeXBLGvrKtLM1doYRK4wBQu0EzPCnVI8PHiVSL3I4DBw7onW9M8sSRNCIiIjIWuW/hQG5lVCU4RF0F/v77b7Rr1w4JCQlpOg14enqicePG+OuvvwCkFLLduHEjQkNDYWZmpilk269fPzx58gSrV69Go0aN0KRJE80qUCIiIjKs5OQkvW/5iVFN0t7uKpDq9u3bGD9+PNavXw9z8/9O7VlYWODQoUM4fPgwli9fnm5b5cqVS3e/j48PgoOD4e/vj7lz52bfjhAREeVjOdkWypgZ1STt3a4CL1++RN++feHn5wdnZ2etOTExMYiPj093v1qtRnBwMOLi/rveq1SpUgBSJnhERERE75NRTdLe7Spw5coVFClSBOXKpb/gN7XTgLu7O0aOHKl1ez169NDaTH3OnDno27evdCxaOw68kV9oT0RERDySpiujmaS93VUgODgYO3fuROPGjVGgQAGsWrUq3eMtLCwQFBSEnTt34siRI1q32atXL6xfn3alWWBgIO7du4cePeQrT3x9fREWFpbmZmMtXwFGRESU7ylKrmsLlVsZzSQttauAv78/jh07hoMHDwIAlixZgt9//x3BwcFa85o2bYqrV6/i5cuX6WI2NjaoUaMGTp48CQC4d+8eZs2ahR9++CH7doSIiCif45E03RjNJC0wMBC1a9cGkLargIWFBdavX49hw4alaaz+tu7du6c7YpZq6NChuHHjBgBg1qxZePjwIdq1a5fhkTQiIiLSDydpumExWwPy9vHB2k1/aY1Jak4CAIo8/kcY6x5sI821MBPPtUd7lpXmFrcXL5JwsZT3Nf3Hu5UwZvPnLmlu4O3nwthIy4vS3GMFmglj/Wfvl+aO/KyRMGZlJi+i+/kb7afNASCoWGtprrOkqHC1079Kc1/9GyqMufQaLs09EGkvjP11/pE0d3n9RGEssYC86Kgq7pU4aCpfmPPR5rvC2NY6z6S5i2OqCGP1x/SS5q4ZIz6Cvrq5/FKGM4q4yKrl2J7S3BK/bBXGvtolL8C7vEMZYWxi4ANp7vxa4tZ5kc7iwr4AYCX5nWP1/LY0VxUt/u7LiiADQL8LLsJYz68/l+Z6rRgqjF2o0lWaW+fFKXHQpZg0N/Hy38JY8usoaa6JYwFh7HLFztLcGmHHhbHnZdXS3GnOVYWxoQ/PS3OrFxUXKk9Vq34ThDp6Zfg4kfqqk/mmmK3RHEkjIiIiyk/YcYCIiIhykAIlWXwkN0Pykx55ilEfSRN1IEiV2nXAy8sLo0eP1tREu3btGlq2bIkWLVpgyJAhSEhI0ORUqVIlTbFcIiIiMixek6Ybo56kaetA8LbUrgOBgYEoXbo0lixZAgAYPnw41qxZg8OHD6N69epYsSKlofOtW7dQt25d7N69O8f2gYiIKF9ROEnTlVFP0t7tQCAzbNgwBAUFITQ0FGXLlkXx4sUBAAMHDoS/vz8AYPv27Rg8eDCeP38OrqcgIiLKHuzdqRujnqS924FAxtTUFIqi4MmTJyhatKjmfjMzMyT//7nxkJAQNG7cGI0aNUJISIh0e9o6DryJicn6ThERERHBiBcOvN2BICIiAjt37pQ+PikpCSYmJihcuDAePfqv9EBiYiJMTEzw4sULhISEoF27dnjz5g0iIyPRsGFD4fZ8fX3h6+ub5j5vH5+s7RQREVGep+S7zgH6MtpJWmoHgvr16yM+Ph69e/eWPn758uVQq9UoU6YMbt26hYcPH6JYsWL46aef0KpVK+zevRvz589Ht27dAACdO3fOgb0gIiLKf/LbtWX6MtpJWmBgICZMmADgvw4EISEh8Pn/o1lDhgxBfHw8PD09oVKpULNmTcydOxcAsHTpUvTp0weJiYmoXLkyli5dit69e2Px4sWa7bu4uODOnTsoU0ZcLJKIiIgyj5M03bDjgAH5+Phg63btp11tXoRKc//pPVgYa7BitjQ36dlDYSy5TntpbuT3E4SxVhFthDEACG5ySxizqCM/9fv7MwdhLCFJ/pH8qHJBYcz2xR1pbld/cSX8P6s/lubutRd3K6jmaifNLYEXwli7P+5Jc//AdmHsp6qDpLlu9pbCWKeK4tcRAFSShhPmJvJuFGe9xe+/05Y90tziO+cKY+d9xkpzLQd8JIw1+Hm+NDfx/k1h7DtFfOkDAHzlLrke1UT+t/APVbsIY1/+e1Ca22DRJWHs9BfiLggAkFBI3FXgzrBPpbnJSeIaV69mrJXmVne1FsYsXsm/g6qEOGFsynn5740Jj7W3BgQAVY/J0tykddOFsa+cuktzn0SKuyhs/9BNmjvgoPj3xsrCF6S5Xv+UFMYiHkdLc7dP9hTGHkaJ3wMA8CpfSBoHgFp1G+Eqqmf4OJHmLqHsOEBERERE74/Rnu4kIiIiY6TwdKeO8uQkLSEhAYUKFcKRI0dQo0YNtGnTRlMLDQCePn2KoUOHIjw8HMnJyRg/fjzat2+Pvn37YseOHdi7dy8aNRKf2iIiIiL9VKlYFoWe3dc7v1ChjE+p5hV5cpL2dieCGjVqpIuPGjUK33zzDWrUqIGEhAScO3cOADBv3jwuFCAiIspGGzZseN9DMBp58po0WSeCxMREvH79WjN5Mzc3R/369QEAbm7yizjfpq2YbUzMG8PsABEREeV7eXKSJutEEB4ejoIFU1a1nT59Gmq1Gn379s30c/j6+iIsLCzNzcZGvGqJiIiIKDPy3CTt7U4EwcHB6ToRFCxYEM+ePQMA1K1bF0FBQXj69On7GCoRERGRUJ67Ji2jTgRmZmawtbXF+fPnUbNmTSQmJr6nkRIRERGJ5blJmi6dCJYsWYIhQ4YgIiICJiYmGDp0KABg3Lhx2Lx5M/bs2QNfX19NiygiIiKinMaOAwbk4+2FPRv9tMZMw0OluRs9vhDGzAPkVcfrFxNX77fxmyLNdapTVxiLrCuu3g4Adubis+W1fP2FMQC4OF9ckf5lovws/LMY8dHPisnyiuUmUWHC2M9R4grdAPBZKfHzRtrIq7vbB/8ujJmXqCDNfRm0Wxh70VX+/nZdfEwYm/unvMp6zDbx83qUFH/mAMAhUdzZweTxVWnub2/KCmPdq8qX3puH3xbGTiXJ36OaF8Urzszdq0pzq/8UJYwFTPWW5u68ES6M9atoK829HmMhjGXUFWLr5SfCWM0i8ve37l/fCmOFOn4ozV0aWVoY+zJqvzQ37vEDYczeo500N6mgeOX+S3NHae7qIrWEsS+enpfm2sZGCGN7q7WW5ta8HCyMdZgTJM092fCKMOY8+7o0996h74Ux2ecVAAY2KCWNU+bkuWvSiIiIiPICo5+kJSQkwMnJCRcupPQxa9Mmbb/Jp0+fomvXrlCr1WjevDl27045OtC3b184OzsjODjlL5VXr16hXr16cHJyytHxExEREWlj9NekGapwrbW1Nfbt24dPP5U3FiYiIiLKCUZ/JM1QhWvNzMxQoECB7B8wERERkQ6MfpKWE4VrtdHWceDNG3YcICIiIsMw6kna+yxcq63jgLU1Ow4QERGRYRj1NWksXEtERER5lVEfSQsMDETt2rUBpC9c6+Pjg61bt2LJkiWYMWMGPD090bp1awwaNAhASuHatWvXYsSIEdi0aRMAoHXr1jh16hR8fHxw48aN97ZfREREREZ9JC0wMDDNz6nlNd61ZcuWdPfNnz8f8+fPT3Pfvn37DDc4IiIioiww6klabhOTCMw8E6M19k3xJGnux3vnCWMHLOVv08//iKtwD+w/U5rrnPxCGDP1+0aa++8JceX4gHJuwhgARCe1EsZuPJcvwKjmaiOMPU8sIc11fRMpjHWsUFCaq4r6VxhLsJQ37nh54aIwdnvOOmnuxI4zhLFvImOluf7jmgtj3ZwXSXMDXMRdA74/r/1znqphcSdhrNaVEGluA486wphZlLyjhGIhvi50zzlxtwkAcGsiXlQUNbS7NPfc8jXC2NNkaSpMVOLOAKbR8uruPRbdFMZ2TvSU5v4bFi2MdalaWJpr7eokjPU7J6/eP6CJOG5RopE0F16VhKGghuLPOgA47RZ3Myh/cK4097cvxd+Vfj+Ml+baebUUxorXl3fBcLMR/+7fO0n+/n5fYowwdu/hOWlug5HpD2ykOtfosjQXDcT/l1HmGfXpTiIiIqK8yugnaYbqOBAeHo6mTZuiefPmGDBgQM7uBBEREdE7jP50p6E6Dri4uODo0aMwMTHB559/jitXrqBKlSo5tRtEREREaRj9kTRDdRwwMTGBiUnKy2FlZZUuTkRERJSTjH6SZsiOA8eOHUPNmjXx+PFj2NvbS59XW8eBhDh2HCAiIiLDMOpJmqE7DjRr1gznz59H2bJl4e/vL31ubR0HzC3ZcYCIiIgMw6gnaakdB/z9/XHs2DEcPHgwTfztjgMApB0H4uPjNf92dHSElZVV9gyaiIiISAdGvXAgMDAQEyZMAJC+4wAADBkyBEuWLMGQIUMQEREBExMTDB06FEBKx4HNmzdjz5498PX1hbu7O0aPHg0TExOUKVMGkyZNem/7RURERKRSFEVehZN01kzthd7zftEa+6yUvG/oK9uiwpgt4qS5JjHigrRv7OTFEl/GiYvsnn8iLnYJAE1KOAhjFqbiAp0AMOdwqDA21fmaNHe3nbjgZUlH+RHQ6vcDhLHkWm2luU/eiKuSFjGRF3e98UY8rlKOFtLcyx3aCGN263dIc4tsEhfC/bak+PpMACjuIi4a3DWDYqeRseLPla2F/AC+1dJRwliRYfI/ni4MHCyMOZR0kea6DxsmjCm2ztLcZmufCGNBoxpIc3+9LP7+Bl17Js3tP0e8v80XfyHNdZgYLIwd2bZQmlurkLkwZhotH7MqVlwkOXzjz9LcpEFzhDHX24eluaG//CqMlVgojgGA+VPx76T4i39Lc+837COMbboo/twAwKtY8f8b09Tywt2mVw4JY6PvlZLmJnb9QBh7smqjNPevgRkUJKZMMerTnURERER5FSdpRERERLmQ0U/SDNVxINXSpUvTbYOIiIgopxn1wgHAcB0HgJTVn9qK4hIRERHlNKM/kmaojgMA8Oeff6Jr1646Pa+2YrbxsSxmS0RERIZh9JM0Q3Yc2LVrF9q1a6fT82orZmthxWK2REREZBhGPUkzZMeBAwcOwNPTEyqVvHQEERERUU4w6kmaITsOXL16FRs3bkSbNm1w+vRprF27NlvHTkRERCRj1AsHDNlxYMSIERgxYgSAlBWiffqICxASERERZTd2HDCghg0bwsHhvyr8MTExsLERV22n//C10h1fK93xtdIdXyvd8bUSs7KySnfpEemPk7Rs5OrqirCwsPc9DKPA10p3fK10x9dKd3ytdMfXinKKUV+TRkRERJRXcZJGRERElAtxkkZERESUC3GSlo18fX3f9xCMBl8r3fG10h1fK93xtdIdXyvKKVw4QERERJQL8UgaERERUS7ESRoRERFRLsRJGhEREVEuxEkaERERUS7ESRoRERFRLsRJGhEREVEuxEmagUVGRuLjjz+Gvb09ihYtisWLF7/vIeUay5YtQ7169WBpaYkePXqkiV26dAmNGjWCjY0NqlSpgsDAwPc0ytwhLi4OAwYMQJkyZWBvb4+qVatiw4YNmjhfr7QGDRqEYsWKwcHBAaVLl8bs2bM1Mb5W6YWHh6NgwYJo1KiR5j6+Tmn169cPFhYWsLOz09zu3bunid+/fx+tWrWCra0typQpg40bN77H0VJexUmagQ0bNgxxcXF4+PAh9u3bh9mzZ2Pv3r3ve1i5QtGiRTFlyhQMHDgwzf0JCQno2LEjOnXqhBcvXmDq1Kno0qVLvm5gnJiYiKJFiyIgIABRUVFYuXIlvvzyS5w4cYKvlxajRo3CrVu3EBUVhaNHj+K3337Dn3/+yddKwNfXF1WqVNH8zNdJuzFjxiA6OlpzK1mypCb2ySefoFy5cggPD8eaNWswcOBAXLp06T2OlvIiTtIM6PXr19i0aRNmzZoFBwcHVK9eHQMHDsQvv/zyvoeWK3z44Yfo3LkzChYsmOb+oKAgxMTEYMKECbC0tET37t1RrVo1bNq06T2N9P2ztbXFjBkz4O7uDpVKhWbNmqFp06Y4fvw4Xy8tqlSpAmtra83PJiYmuHXrFl8rLQ4fPoybN2+if//+mvv4OmXOzZs3cfLkScyaNQvW1tZQq9Xo1KkTfv311/c9NMpjOEkzoBs3biA5ORnVqlXT3FerVi3+dZWBS5cuoXr16jAx+e/jyNctrdevX+PUqVOoVq0aXy+BiRMnwtbWFiVLlsTr16/Rq1cvvlbviI+Px7Bhw/DDDz9ApVJp7ufrpN2qVavg4uKCmjVrpvlj+9KlSyhVqhScnZ019/H1ouzASZoBRUdHw9HRMc19Tk5OePXq1XsakXGIjo6Gk5NTmvv4uv0nOTkZ/fr1Q/369dGqVSu+XgJz5sxBdHQ0QkJC0LNnTzg7O/O1esfcuXPh4+ODmjVrprmfr1N6I0aMwI0bNxAWFobFixdj3Lhx2LJlCwC+XpRzOEkzIDs7O0RFRaW57+XLl7C3t39PIzIOdnZ2ePnyZZr7+LqlUBQFgwcPxqNHj/DHH39ApVLx9ZJQqVSoX78+rKysMHXqVL5Wb7l16xb8/Pwwffr0dDG+TunVqVMHBQsWhJmZGTw9PTF06FDN6V++XpRTOEkzoAoVKkClUuHy5cua+86dO5fm9CelV61aNVy8eBHJycma+/i6pUzQhg4dinPnzmHv3r3/196dRzV57H0A/yKgbIpaQBBRxAJat2oQsS6lCiLnihvqFRWRqiBSKItwceGAC4soEpa2KFWUqpXq9RZUNqsHAc/F6waC1bJIggmgIgaQsAXm/cOX5xITIFYtyJ3POTmHzDOZ/J55Jk8mM/M8QE1NDQCtL1mIRCKUlpbSuuokJycHVVVVMDIygra2Nr799lvcvXsX2traMDAwoPXUgwEDBoAQAuD1Z5DL5UIgEDDbaX1RHwSh3qu1a9eSpUuXkrq6OlJQUEBGjBhBUlJSejusPqG1tZU0NjaSXbt2kVWrVpHGxkbS0tJCWlpaiL6+PgkJCSFNTU3kl19+IUOGDCFPnz7t7ZB71bZt28i0adNITU2NWDqtL3E1NTUkISGB1NbWkra2NpKTk0M0NTVJVFQUratOhEIhqaysZB5sNptMnz6dVFZW0nqSIjExkdTV1ZG2tjaSnZ1NNDQ0yM8//8xs/+KLL4irqysRCoXk+vXrZPDgwaSgoKAXI6b6I9pJe89evnxJVq5cSVRVVYm2tjaJiIjo7ZD6jICAAAJA7OHg4EAIIeT+/fvE1NSUKCkpkfHjx5Pffvutd4PtZRwOhwAggwYNIqqqqswjKCiIEELrq7OXL1+Sr776igwdOpSoqakRY2NjEhoaStrb2wkhtK66Eh8fT2bOnMk8p/Ukbu7cuURdXZ2oqamRzz77jMTGxoptLy8vJxYWFkRZWZmMGTOGnDlzppcipfozOUL+f/yWoiiKoiiK6jPomjSKoiiKoqg+iHbSKIqiKIqi+iDaSaMoiqIoiuqDaCeNoiiKoiiqD6KdNIqiKIqiqD6IdtIoiqIoiqL6INpJoyiKoiiK6oNoJ42iKIqiKKoPop00qk+5cOECPD09medOTk5gs9ld5q+oqICJiQlKSkpkfo8jR47A3t7+XcJ8r0xMTJCdnd3bYXxQNjY2SExM7O0w/meUlJTAxMQEFRUVvR3KR6enc867eN+f9draWlhZWeHZs2fvrUyqb6GdtH6muroaYWFhWLp0KWbNmgUrKys4OTkhKSkJLS0tvR1et5qbm3HkyBFs3rxZ5teMGDECaWlp0NfX/3CBfWBpaWmYOXNmb4fxUetrHW/qrxMYGAhfX9/eDkNMV+3xfX/W1dXVsXjxYhw5cuS9lUn1LQq9HQD1/vB4PGzatAk6Ojrw8vKCvr4+BgwYgD/++APnz5/HqFGjwGKxejvMLl29ehVDhw7FxIkTZX6NvLw8NDQ0PmBUH05raysUFRU/2vipD08kEkFB4eM5TXe06b+CSCSCvLz8eyvvr4j9Q3zWbWxssG7dOnh4eGDw4MHvvXyqd308n36qRwcOHICWlhaOHz+OAQP+O0iqp6cHCwsLdP43rVVVVYiIiEBubi4UFBTAYrHg4+MDTU1NAK9/nQqFQowfPx4///wzAGD58uXYtm0bU0ZdXR3YbDauX78OkUiEyZMnY/v27cyoVlFREQ4dOoRHjx5BTk4OY8aMwZ49ezB27Fip8WdkZGDu3LkS6W1tbQgPD8fFixehoqKCjRs3YvXq1QBeT3cuWbIEZ8+exaeffgoAyMzMBJvNxvPnzzFt2jTMnz8fwcHBuH37tli5SUlJOHr0KIRCISwsLODr6yv1JF1XVwcrKyuw2WyxX8HJycmIjo5GamoqACAoKAi3b99GTU0NdHR0sG7dOixfvpzJ7+TkBCMjIxBCkJaWhkmTJiEyMhImJiaIiIhg9r2jTp89ewYNDQ0sWbIEjo6OzDGV5djU1tYiKioKWVlZEAqFGD16NLy8vDBjxgymjo4ePQoOhwMtLS0sX74c9vb2Yu2ms9u3byMqKgqlpaUYOHAgPv30U4SFhWHYsGFMPGFhYUx+X19fqKioIDAwkEl79eoVduzYgezsbKirq8PZ2RlLliwB8PoL8vDhw7h27Rrq6+uhqakJOzs7rFmzhjkGXbW1ixcvIi4uDsDr6SQACAgIgI2NjdR96czJyQl3796VSE9OTsbIkSMl0nuKs6d6T0xMxJkzZ/D06VPo6elh69atWLBgAYD/tuWQkBAkJibiwYMH2LdvHywsLPCvf/0LP/30E6qqqqCrqwt7e3um7gCgoKAAISEh4HA4MDIywrp163rcdxMTE+zYsQPXrl1DXl4etLS04O3tjTlz5jB5SkpKEBkZiXv37kFVVRVz5syBp6cn1NTUmPqT1qbf1NFGDAwMcO7cObS1tcHGxgYeHh5MR6u5uRnff/890tPT0dDQAENDQ3h6emLy5MkAgIsXL4LNZsPf3x/R0dF48uQJbG1tcenSJWZ/ACA2NhYAsHXrVmRlZUFFRQUAkJ2dDU9PT+Y8cOTIEeTk5GDZsmU4ceIEBAIBMxUpEokQGhqK1NRUDBw4EGvXroWjoyOzPwkJCbh06RL4fD6GDh2KBQsWwNXVFYMGDeq2Pb75We84RxYWFkJFRQULFy6Eh4cHBg4cyNSvsbExs/9vnv8AQF9fH1paWrh+/ToWL17c43GnPi60k9ZPCAQC5ObmYv/+/V1+0crJyQF4fQJyc3PD559/jmPHjkFOTg6xsbHw8vLCyZMnmdffvHkTI0aMQFxcHB49egR/f39MmzYNs2bNAgD4+flBRUUFMTExUFFRwdmzZ+Hq6orz589DWVkZu3fvhrGxMXbu3Ak5OTk8fPiQiUGavLw8qV+sycnJcHR0REJCArKysnDw4EHMmDFDamePz+fDz88P69evh42NDQoKChAdHS2Rj8vl4ubNm4iOjkZVVRX+8Y9/YPz48bC1tZXIO2TIEHzxxRdIT08X66SlpaXBwsICCgoKaG5uhra2Ng4cOAB1dXXcuXMHoaGh0NHRgZmZmdi+/P3vf8fx48e7rAc1NTXs2bMHGhoaePToEYKCgjBs2DCsWLGCydPTsfH19UV9fT2Cg4Oho6ODkpISpu7v3buHwMBA+Pj4YOrUqeByuQgKCsLAgQNhZ2cnEY9IJML27duxfPlyBAcHo7m5GQUFBV3G35WEhARs2bKF+fLcv38/DAwMMGnSJJw9exZZWVk4cOAARowYAT6fj9raWua13bU1S0tLlJaW4j//+Q+ioqKYOpTFwYMH0draKva8pKQEw4cPl5q/pzi7q/erV6+CzWbDx8cHLBYLV69exY4dO5CQkIDx48czZcTExMDT0xOGhoZQUlJCamoq4uLi4OvrC0NDQ/z+++/Yv38/hgwZAnNzczQ0NMDDwwOzZ8/G/v37UV5ejkOHDsm0/7GxsXB3d4ePjw9+/fVX+Pr64sKFC9DW1kZ9fT1cXFxga2sLb29vNDY24vDhwwgMDBQrX5Y2DQC5ublQUlJCXFwcysvLsXfvXmhqamLDhg1M3XM4HISGhkJDQwNpaWnMMdbS0gIACIVCnDp1Cnv27IGqqio0NTUhEAjQ1NSEXbt2AXg9BZifny/T/nM4HGRnZyM8PFzsvJmcnAxbW1ucPHkS9+/fR2hoKPNjF3g9gu/r64uRI0eivLwcISEhGDRoEFxdXWVuj42NjXBzc8P06dORkJCA6upq7Nu3D/Ly8vD29haLpafz34QJE3Dv3j3aSeuPCNUvFBQUEBaLRR4+fMikCYVCMmfOHOZx7NgxQgghly9fJqtWrRJ7fUNDAzE1NSUFBQWEEEICAgLI0qVLSXt7O5PH0dGRREVFEUIIuXfvHvnqq69IS0sLs72trY1YWVmRjIwMQggh8+bNIxcvXpQp/rq6OsJisUh+fr5Y+pYtW4izs7NYmpWVFTl37hwhhBA+n09YLBYpLi4mhBASFRVF7OzsxPIfOnSIsFgs5nlsbCyZO3cuEQqFTNrOnTuJn59fl/FlZGQQc3NzZn+fP39OTE1NJeLtLCAggOzdu1dsX+zt7SXysVgskpWV1WU5sbGxxMXFRazc7o7N7du3iampKeHxeFLLc3FxISdPnhRLO3fuHFm5cqXU/AKBgLBYLHL79m2p2wMCAoiPj49Ymo+PDwkICGCeL168mHh6eorl+fbbb8nOnTsJIYSEhYWRrVu3iu1TB1naWmxsLFm/fr3U+GR17tw5Ym5uTjgcTpd5uouzp3p3dHQkISEhYmlubm5k165dhJD/tuXExESxPEuXLiVXrlwRS4uKiiJubm6EEELOnz9PLCwsSHNzM7M9ISGBsFgswufzu9wXFotFDh48KJa2evVqEhMTQwghJC4ujri7u4ttLysrIywWi7x48YIQ0nWbflNAQIDUGK2srAghhFRWVhJTU1NSXV0t9joHBwdy4sQJQgghycnJhMVikZKSEomy32x/t27dIiwWizQ0NDBpWVlZEueBWbNmEYFAIPbaLVu2kDVr1oilhYaGEkdHxy73Lzk5mSxdulSsbGntsfNn/Z///CeZP38+aWxsZLZnZGQQMzMz5tzU0/mvQ3h4uNg5guo/6EhaP6akpIQzZ84AALZv386MGBQXF4PL5UpMLba1tYHH42HSpEkAAAMDA7GRr08++QQ1NTUAXg/Tv3r1CvPnzxcro7m5GTweDwCwZs0a7Nu3DykpKTAzM4OlpSV0dHSkxtrc3AwAzDB/Zx3TmB00NDTw8uVLqeVwuVyJNW3S1rjp6upCWVlZrMw//vgDAHD8+HHEx8cz286dO4e5c+eivb0dN27cgLm5Oa5cuQJtbW1MmTKFyffLL78gOTkZlZWVaGlpQWtrq8QawAkTJkiNu7OMjAycPXsWPB4PjY2NEIlEEvXW3bEpLi6GtrY2dHV1pZZfVFSE/Px8ZkoGANrb29He3i41v7q6OqytreHu7g5TU1PmWHY12tSVjnbVYfLkybh69SoA4G9/+xtcXV1ha2uL2bNnY968ecwUoSxt7V3dvXsXERERCA8Px5gxYwAAwcHBzFQ28Hq6rLs4e6r3srIyrFq1Sixt6tSpuHbtmljaZ599xvzd2NgIHo+HgIAA7Nmzh0nv3CY6pjg7f3Y6pgh7Iu2YlJWVMftz8+ZNqUsQeDwec/xladMApMZYXV2NV69eoaSkBG1tbVi2bJnYa1paWmBoaMg8HzRoEMaNGyfT+8li5MiRUFdXl0h/s16mTJmCtLQ05vnNmzcRHx8PDoeDhoYGtLe3o62t7a3eu6ysDMbGxlBSUmLSpk6ditbWVvB4PGa/ZTn/DRo0CE1NTW/1/tTHgXbS+olRo0ZBTk4OHA6HmTqRk5ODnp4eAPHOj1AoxMSJE8VO+h06f/G+uWBZTk6OWdcmFAqhpaWFH374QaKMIUOGAABcXFxgbW2NnJwcZGdn48iRI4iIiICpqanEa9TV1SEnJ4f6+nqJbdLi6KpDIavuyrS1tYWlpSWzTUNDAwoKCjA3N0d6ejrMzc2RlpYGKysrJk96ejoiIyPh5eWFiRMnQlVVFUePHsXTp0/F3qdzx1Ca+/fvw9/fHy4uLpg5cyZUVVWRlJSEjIyMHuMnndYcdqexsREuLi748ssvZcoPAPv27cPatWtx48YNXL58GT/88AOOHTuGcePGSX1vkUgkc9nA645JcnIybty4gZs3b8Lb2xsLFy7E7t27ZWpr76Jjunvbtm1iU9Nbt26VuEKvuzjfl85f2kKhEMDrNU1vdoY+9AUFjY2NMDc3h6urq8S2jrWrQM9tWhZCoRAKCgo4ffq0xJIIVVVV5u/OddOdjqnLzu1SWpv8M7FXVFTA09MTq1evxjfffIPBgwcjNzcXBw8efOuyZCHL+a+urg7Dhg37IO9P9S7aSesnhg4dipkzZyIhIQGWlpbdXvVkbGyMq1evYvjw4WInwLcxfvx4PH/+HIqKitDW1u4yn76+PvT19bF+/Xq4u7sjNTVVaidNUVERY8eOxePHj5mRiT9jzJgx+Pe//y2W9uDBg7cqQ11dXeqva2tra3h7e6OoqAgPHjxAQEAAsy0/Px/Tpk0TW9P25MkTqSOD3bl//z50dXWxceNGJu1t73VlaGiIqqoq8Pl8qaM6xsbG4HK5TAdeVhMmTMCECROwadMmrF69GteuXcO4ceMwbNgwcLlcJl97eztKS0sxbdo0sdcXFhZKPO+8rmbw4MFYtGgRFi1ahFmzZsHf3x87d+6Uqa0pKiq+9UgGADQ1NTGL5d9cbD98+HCpo4VdxdlTvY8dOxb5+fmwtrZm0vLz82FgYNBlfJ988gk0NDTA5/OxcOFCqXn09fWRlpaGlpYWpr29WdddKSwsxKJFi8Sed4ycGRsb4/r16xg5cuR7uYqyqKhIIkYNDQ2oqanByMgIIpEIAoFAbHRaFoqKihKdlo4Oy4sXL5hzXFFRkcxlvnnOKCgoYNrqw4cPAQAeHh7M9uTkZImYemqPY8eORUpKCpqampjOZ35+PhQVFTFq1CiZYwVej8p1XKRA9S/0Pmn9iJ+fH6qrq+Ho6IjMzExwuVyUlZUhKSkJFRUVzInW2toaampq8PHxQV5eHvh8Pm7duoWQkBCpI1nSmJqaYuLEifD29kZubi74fD7y8vIQGRkJDoeDpqYmhIWF4e7du6isrMSdO3dQXFzc7f3MzMzMkJeX9051sGLFCjx+/BjfffcduFwuUlJSJEah/qwZM2ZATU0NAQEBMDIyEvtyHT16NAoLC5Gbmwsul4vIyEhm2uht6OnpoaKiAleuXAGPx8NPP/0k0ensCYvFwueffw4fHx/cunULfD4fWVlZzFVtmzZtQnJyMn788Uc8fvwYjx8/RmpqKo4dOya1PD6fj5iYGNy/fx9VVVXIycnB06dPmWPJYrFQWFiI9PR0cLlchIeHQyAQSJRz9+5dnD59GlwuF6dOncKNGzeYqyJPnz6NjIwMcDgccDgcZGZmYvTo0RgwYECPbQ14PW3F5/NRVFQEgUAg8z0Bg4KCIBKJ4OzsjOrqaubR1Rdsd3H2VO/r1q1DUlISLly4gPLycsTHxyM3N7fHKzE3b96M48ePIzExEVwuF8XFxbhw4QLOnz8PAFi0aBHa29sRHByMsrIyZGZmynzj4IyMDFy6dIlps1wul/mhsWrVKrx48QK7d+/G77//Dh6Ph5ycHAQFBclU9puam5vFYoyPj2eOv76+PiwtLeHv74/MzEzw+XwUFhYiLi4Od+7c6bZcHR0dZgmHQCCASCSCnp4ec2FNeXk5fvvtN4mOVHd4PB6io6PB5XJx6dIlJCUlMbHq6emhpaUFiYmJ4PF4SEpKYq4w7SBLe7S2toaCggL27NmD0tJS5Obmgs1mY+XKlW81wtfc3IyHDx+KjQJT/QcdSetHRo0ahdOnT+P48eM4fPgwnj17xqzh2LJlC3N1oLKyMuLi4hAVFcVctTVixAiYmZnJPPIzYMAAREVF4bvvvkNgYCBqa2uhoaEBFouFIUOGQF5eHgKBAP7+/qipqcHw4cOxaNGibr+Qli1bhg0bNqChoeFPj/Dp6uoiNDQUbDYbp0+fxvTp07Fx40bExMT8qfI6k5eXh6WlJc6ePQt3d3exbStWrMCjR4/g5+cHeXl5WFtbY8mSJcw6N1l9+eWXsLOzQ2hoKEQiEebNmwcHBwdcuHDhrco5cOAA2Gw2/Pz80NTUhDFjxsDLywsAMHv2bBw+fBg//vgj4uPjoaioCAMDA4n1Uh2UlJRQVlaGixcvoq6uDlpaWvj666+ZKeHZs2fDwcEBhw4dQnt7O+zs7KTesNPe3h75+fn4/vvvoa6ujp07dzJrp5SVlXHixAk8efIE8vLymDx5MnNLj57aGgDMnz8f165dg7OzM+rr65lbHgQGBqKiogJHjx6Vum95eXmorKyUuCquq1twdBdnT/W+YMECVFdX48SJEwgLC4Oenh5CQkLEruyUpuNL+9SpU2Cz2VBRUYGRkRFzVaSqqioOHz6M0NBQrF27FoaGhvjmm2+Yqx274+TkhJSUFAQHB0NLSwsHDhxgRiu1tLRw7NgxREdHY9u2bWhtbYWuru5bTZN3ZmZmBi0tLWzevBkikQg2NjZYv349s33v3r2Ii4tDeHg4nj9/juHDh2PKlCliywqkWb58Oe7cuQN7e3sIhULExsbCxMQE+/btQ2hoKOzs7DB9+nRs2bIF+/fvlynWJUuWoK6uDhs2bICioiK+/vprZiTTyMgInp6eiI+PR1RUFGbMmAEXFxexsrtqj50pKysjOjoahw4dgr29PXMLDjc3N1mrFMDrtZJvro+l+g85IutCFor6C3TcFqLzyftdxcTEIDs7m/5bov9BTk5OYLFYcHZ27u1Q+pw379n1IUm7lx71fmzatAmrVq0Sm7am+g863Un1KR4eHjIvDu5KYmIiMz3z66+/IjExkd4/6H9QQ0MD+Hw+/XdRVL9VW1uLefPm9TjaSH286HQn1afo6upi5cqV71RGx3qfuro66OjowMnJSaY7sFP9i6qqKi5fvtzbYVDUB6Ourg4HB4feDoP6gOh0J0VRFEVRVB9EpzspiqIoiqL6INpJoyiKoiiK6oNoJ42iKIqiKKoPop00iqIoiqKoPoh20iiKoiiKovog2kmjKIqiKIrqg2gnjaIoiqIoqg+inTSKoiiKoqg+iHbSKIqiKIqi+qD/A9kLWXDx8gfNAAAAAElFTkSuQmCC", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "code", - "execution_count": 28, - "id": "a01ce791", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "_fig_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", - "plot_training_history(history, tag=_fig_tag)\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "markdown", - "id": "01fea9eb", - "metadata": {}, - "source": [ - "## Predict the evaluation split and save outputs\n", - "\n", - "Run inference on the held-out split, compute per-gene metrics, and persist `pred` / `real` AnnData plus CSV summaries.\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "code", - "execution_count": 29, - "id": "de3913d9", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Aggregate summary:\n", - "method_name conditional_flow_matching_method\n", - "input_path /root/final/data/processed/scPerturb/predictio...\n", - "eval_split test\n", - "train_split train\n", - "val_split val\n", - "n_control_cells_train 8653\n", - "n_eval_cells 29965\n", - "n_train_genes 2056\n", - "n_eval_genes 2056\n", - "n_eval_genes_missing_from_train 0\n", - "mse_mean_avg 0.019163\n", - "mae_mean_avg 0.097516\n", - "pearson_mean_avg 0.975644\n", - "pearson_delta_avg 0.089579\n", - "delta_l2_avg 8.494132\n", - "dtype: object\n", - "\n", - "Saved:\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/conditional_flow_matching_model.pt\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/pred_conditional_flow_matching_method_seen_cell_split_test.h5ad\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/real_conditional_flow_matching_method_seen_cell_split_test.h5ad\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/metrics_by_gene_conditional_flow_matching_method_seen_cell_split_test.csv\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/aggregate_conditional_flow_matching_method_seen_cell_split_test.csv\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/summary_conditional_flow_matching_method_seen_cell_split_test.csv\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/training_history_conditional_flow_matching_method_seen_cell_split_test.csv\n", - "- /root/final/baseline/outputs/conditional_flow_matching_method/config_conditional_flow_matching_method_seen_cell_split_test.json\n" - ] - }, - { - "data": { - "text/html": [ - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
perturbation_genen_cellsmse_meanmae_meanpearson_meanpearson_deltadelta_l2
0AAAS210.0066970.0621510.991311-0.0466045.387447
1AAMP100.0194500.1045790.9742160.2490969.181193
2AARS40.0448530.1606990.9435050.29963813.942513
3AARS2180.0072880.0660640.9902740.0369565.620068
4AASDHPPT50.0301150.1334670.9641450.01852711.424403
\n", - "
" - ], - "text/plain": [ - " perturbation_gene n_cells mse_mean mae_mean pearson_mean pearson_delta \\\n", - "0 AAAS 21 0.006697 0.062151 0.991311 -0.046604 \n", - "1 AAMP 10 0.019450 0.104579 0.974216 0.249096 \n", - "2 AARS 4 0.044853 0.160699 0.943505 0.299638 \n", - "3 AARS2 18 0.007288 0.066064 0.990274 0.036956 \n", - "4 AASDHPPT 5 0.030115 0.133467 0.964145 0.018527 \n", - "\n", - " delta_l2 \n", - "0 5.387447 \n", - "1 9.181193 \n", - "2 13.942513 \n", - "3 5.620068 \n", - "4 11.424403 " - ] - }, - "execution_count": 29, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "eval_conditions_to_run = [condition for condition in eval_conditions if condition in indices_by_split_gene[EVAL_SPLIT]]\n", - "if not eval_conditions_to_run:\n", - " raise ValueError(f\"No conditions found for EVAL_SPLIT={EVAL_SPLIT}.\")\n", - "\n", - "metrics, pred, real = evaluate_gene_metrics(\n", - " split_name=EVAL_SPLIT,\n", - " conditions=eval_conditions_to_run,\n", - " seed=RANDOM_SEED + 999,\n", - ")\n", - "aggregate = build_aggregate_frame(metrics)\n", - "aggregate_summary = pd.Series(\n", - " {\n", - " \"method_name\": METHOD_NAME,\n", - " \"input_path\": str(INPUT_PATH),\n", - " \"eval_split\": EVAL_SPLIT,\n", - " \"train_split\": TRAIN_SPLIT,\n", - " \"val_split\": VAL_SPLIT,\n", - " \"n_control_cells_train\": int(len(control_train_idx)),\n", - " \"n_eval_cells\": int(pred.n_obs),\n", - " \"n_train_genes\": int(len(train_conditions)),\n", - " \"n_eval_genes\": int(len(eval_conditions_to_run)),\n", - " \"n_eval_genes_missing_from_train\": int(len(set(eval_conditions_to_run) - set(train_conditions))),\n", - " \"mse_mean_avg\": float(metrics[\"mse_mean\"].mean()),\n", - " \"mae_mean_avg\": float(metrics[\"mae_mean\"].mean()),\n", - " \"pearson_mean_avg\": float(metrics[\"pearson_mean\"].mean(skipna=True)),\n", - " \"pearson_delta_avg\": float(metrics[\"pearson_delta\"].mean(skipna=True)),\n", - " \"delta_l2_avg\": float(metrics[\"delta_l2\"].mean()),\n", - " }\n", - ")\n", - "\n", - "config = {\n", - " \"name\": METHOD_NAME,\n", - " \"input_path\": str(INPUT_PATH),\n", - " \"split_strategy\": SPLIT_STRATEGY,\n", - " \"train_frac\": float(TRAIN_FRAC),\n", - " \"val_frac\": float(VAL_FRAC),\n", - " \"eval_split\": EVAL_SPLIT,\n", - " \"random_seed\": int(RANDOM_SEED),\n", - " \"latent_dim\": int(LATENT_DIM),\n", - " \"hidden_dim\": int(HIDDEN_DIM),\n", - " \"condition_dim\": int(CONDITION_DIM),\n", - " \"time_dim\": int(TIME_DIM),\n", - " \"dropout\": float(DROPOUT),\n", - " \"num_flow_blocks\": int(NUM_FLOW_BLOCKS),\n", - " \"batch_size\": int(BATCH_SIZE),\n", - " \"epochs\": int(EPOCHS),\n", - " \"steps_per_epoch\": int(STEPS_PER_EPOCH),\n", - " \"validation_steps\": int(VALIDATION_STEPS),\n", - " \"learning_rate\": float(LEARNING_RATE),\n", - " \"lr_scheduler\": LR_SCHEDULER,\n", - " \"plateau_factor\": float(PLATEAU_FACTOR),\n", - " \"plateau_patience\": int(PLATEAU_PATIENCE),\n", - " \"early_stop_patience\": int(EARLY_STOP_PATIENCE),\n", - " \"fixed_val_subset\": FIXED_VAL_SUBSET,\n", - " \"flow_lr_mult\": float(FLOW_LR_MULT),\n", - " \"uncertainty_samples\": int(UNCERTAINTY_SAMPLES),\n", - " \"weight_decay\": float(WEIGHT_DECAY),\n", - " \"recon_weight\": float(RECON_WEIGHT),\n", - " \"flow_weight\": float(FLOW_WEIGHT),\n", - " \"endpoint_weight\": float(ENDPOINT_WEIGHT),\n", - " \"mmd_weight\": float(MMD_WEIGHT),\n", - " \"mean_weight\": float(MEAN_WEIGHT),\n", - " \"delta_mse_weight\": float(DELTA_MSE_WEIGHT),\n", - " \"delta_cos_weight\": float(DELTA_COS_WEIGHT),\n", - " \"delta_mmd_weight\": float(DELTA_MMD_WEIGHT),\n", - " \"delta_bulk_weight\": float(DELTA_BULK_WEIGHT),\n", - " \"noise_std\": float(NOISE_STD),\n", - " \"sinkhorn_epsilon\": float(SINKHORN_EPSILON),\n", - " \"sinkhorn_iters\": int(SINKHORN_ITERS),\n", - " \"inference_steps\": int(INFERENCE_STEPS),\n", - " \"n_train_conditions\": int(len(train_conditions)),\n", - " \"n_eval_conditions\": int(len(eval_conditions_to_run)),\n", - " \"n_unseen_eval_conditions\": int(len(set(eval_conditions_to_run) - set(train_conditions))),\n", - " \"feature_gene_columns\": FEATURE_GENE_COLUMNS,\n", - " \"model_path\": str(MODEL_PATH),\n", - "}\n", - "\n", - "pred.uns[\"baseline\"] = config\n", - "real.uns[\"baseline\"] = config\n", - "\n", - "tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", - "pred_path = OUTPUT_DIR / f\"pred_{METHOD_NAME}_{tag}.h5ad\"\n", - "real_path = OUTPUT_DIR / f\"real_{METHOD_NAME}_{tag}.h5ad\"\n", - "metrics_path = OUTPUT_DIR / f\"metrics_by_gene_{METHOD_NAME}_{tag}.csv\"\n", - "aggregate_path = OUTPUT_DIR / f\"aggregate_{METHOD_NAME}_{tag}.csv\"\n", - "summary_path = OUTPUT_DIR / f\"summary_{METHOD_NAME}_{tag}.csv\"\n", - "history_path = OUTPUT_DIR / f\"training_history_{METHOD_NAME}_{tag}.csv\"\n", - "config_path = OUTPUT_DIR / f\"config_{METHOD_NAME}_{tag}.json\"\n", - "\n", - "torch.save(model.state_dict(), MODEL_PATH)\n", - "pred.write_h5ad(pred_path, compression=\"gzip\")\n", - "real.write_h5ad(real_path, compression=\"gzip\")\n", - "metrics.to_csv(metrics_path, index=False)\n", - "aggregate.to_csv(aggregate_path)\n", - "aggregate_summary.to_frame(name=\"value\").to_csv(summary_path)\n", - "history.to_csv(history_path, index=False)\n", - "with config_path.open(\"w\") as f:\n", - " json.dump(config, f, indent=2)\n", - "\n", - "print(\"Aggregate summary:\")\n", - "print(aggregate_summary)\n", - "print(\"\\nSaved:\")\n", - "for path in [MODEL_PATH, pred_path, real_path, metrics_path, aggregate_path, summary_path, history_path, config_path]:\n", - " print(\"-\", path)\n", - "\n", - "metrics.head()\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "markdown", - "id": "viz-eval-metrics-md", - "metadata": {}, - "source": [ - "## Figures — held-out evaluation\n", - "\n", - "Per-perturbation metrics (MSE vs Pearson, ranked Pearson, delta L2, mean vs delta correlation). Seen vs unseen training genes are colored when both cohorts exist.\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "code", - "execution_count": 30, - "id": "97a06a5d", - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "_fig_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", - "plot_evaluation_metrics(metrics, tag=_fig_tag, train_genes=train_conditions)\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "markdown", - "id": "bio-extended-md", - "metadata": {}, - "source": [ - "## Extended biological analysis figures\n", - "\n", - "Additional plots for manuscript-style interpretation: gene-level calibration, MA-style view, perturbation–gene heatmaps, perturbation similarity, cell-space PCA, and error distributions. Requires upstream cells that define `pred`, `real`, `adata`, `metrics`, `control_mean`, and `eval_conditions_to_run`. Figures append to `FIGURE_DIR` with the `bio_` prefix.\n" + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqAAAAIrCAYAAAAwSyveAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAQ6wAAEOsBUJTofAAA1LxJREFUeJzs3Xd4VFX6wPHvnZLJZCaVJCShJKGXgFRBmopUQVyVolhAFBv409VFWXUXdBULuuourmVd0bWsEhXrsgKKCiqI9NBbQgkhvUzKtHt+fwQGhiSQhDTg/TzPPA9z7rn3vndmmHlz7imaUkohhBBCCCFEAzE0dgBCCCGEEOLCIgmoEEIIIYRoUJKACiGEEEKIBiUJqBBCCCGEaFCSgAohhBBCiAYlCagQQgghhGhQkoAKIYQQQogGJQmoEEIIIYRoUJKACiGEEEKIBiUJqBBCCCGEaFCSgApxgfr+++/RNI3PPvus3s+VmpqKpmm8/fbb9X6uUx2/zu+//77Bz32qyy67jMsuu6yxwzgrb7/9NpqmkZqa2tihNIgL7XqFaCiSgApRQ8d/kI4/rFYrnTt3Zvbs2RQUFNTZecrKypg7d26TSJyq63//+x9z585t7DBENX344Ye89NJLjR2GEOICJAmoELX01FNP8e677/LSSy+RlJTEs88+y8iRI1FK1cnxy8rKePzxx8+5BPTxxx+vUB4fH09paSk333xzI0QlqlKbBPTmm2+mtLSU+Pj4+glKCHFBMDV2AEKcq6688kp69OgBwJ133smECRP4+OOP+eWXXxgwYECtj+v1evF4PHUUZUUulwuDoWH/9tQ0jcDAwAY9p6hbxcXF2Gw2jEYjRqOxscMRQpzjpAVUiDpyvG/f8b5iuq7z/PPP07lzZywWC7Gxsdx77704HA6//TRN4/777+ff//43nTp1wmKx8J///Ifw8HAAHn/8cd/t/uO3t6vqSzh16lQSEhJ8z4/3vXzxxRd54YUXSExMxGq1cujQIV8dj8fD7Nmzad68OXa7neuuu46MjAy/465cuZIJEybQunVrLBYLrVq14ve//z2lpaV+53755Zd913T8cXIcp/YB/fbbbxk4cCBBQUGEh4czYcIE9u/fX+GawsLCOHjwIOPGjcNutxMVFcUf/vAHvF5v1W/IGfz8888MHz6ckJAQbDYbw4YNY926db7t8+fPx2Aw+L1Wx919990EBwf7Xf9XX33FgAEDsNlshIaGcs0117Bnz55axXb8M/HBBx/QqVMngoKCGDJkCLt37/bF1rp1a6xWK+PGjSM3N7fCMc4Uz2WXXcbnn39OWlqa7706/tk53m82OTmZP/7xj8TFxREcHAxU3Sfy66+/ZsiQIdjtdkJDQxkwYMAZ+xcff2/37NnD8OHDsdlstGrViueee65C3eeff54BAwbQrFkzrFYrvXv35uOPP67ytfvkk0/o2rUrFouFrl278r///a9C3YMHD3LLLbcQHR2NxWKhe/fu/Oc//zltzAC//fYbI0eOJDIyEqvVSmJiItOmTTvjfkKIE6QFVIg6sm/fPgCaNWsGwG233cYHH3zAtGnTuP/++9m9ezcLFixg27ZtLF++3JecASxdupSPPvqIGTNmEBYWRs+ePXn99de58847ueaaa7j22msB6N69e61ie/PNN3G73dxzzz0YDAbsdrtv2xNPPIHJZOKRRx7h8OHDvPzyy+zdu5e1a9diNpsBSE5OpqSkhLvvvptmzZrx66+/8ve//51Dhw6RnJwMlLcCZ2Rk8M033/Duu++eMably5czevRoOnTowF/+8hcKCwt5+eWXGThwIJs2bSIqKspX1+12M2LECAYOHMjzzz/PsmXLeOGFF2jbti133313jV+P5cuXc+WVV9K/f3+eeOIJdF3njTfeYMiQIaxdu5YuXbowceJEHnroIZKTk/n973/v29fr9fLpp58ybtw4rFYrUJ6UTZs2jbFjx/Lcc89RVFTE3//+dwYNGsSmTZto3rx5jWNcsWIFn332Gffccw9ut5t58+ZxzTXXMHnyZBYvXswf/vAH9u/fz8svv8ysWbP417/+5du3OvE8+uijOBwO0tLSePHFFwH8PhcAc+fOJSgoiIcffpiioqIqY/3Xv/7F7bffzkUXXcSjjz5KcHAw69atY9myZfzud7877XW63W5GjRrF4MGDee655/jqq694+OGHAXjooYd89V5++WXGjRvHjTfeiMvl4sMPP2TChAl89dVXjBkzxu+YP/zwA8nJydxzzz3Y7Xb+9re/cd1113HgwAHf/88jR47Qv39/AgICuO+++4iIiODzzz9n8uTJOJ1Opk6dWmm8mZmZjBgxgsTERB577DFsNhv79+9n8eLFp71OIcQplBCiRhYuXKgAtWLFCpWVlaXS0tLUm2++qQIDA1V0dLQqLi5WP/74owLUxx9/7Lfvhx9+qAC1ZMkSXxmgjEaj2rlzp1/dvLw8Bag5c+ZUiOHSSy9Vl156aYXyKVOmqPj4eN/z/fv3K0CFhYWp7Oxsv7orVqxQgGrdurUqKirylb/zzjsKUG+//bavrKSkpMK5nn76aaVpmkpLS/OV3Xfffaqyr5XjcSxcuNBX1qNHDxUTE6Py8vJ8Zb/88osC1AMPPOB3TYCaN2+e3zF79uypevfuXeFcpzp+nStWrFBKKeX1elW7du3U2LFj/erl5+ermJgYNWnSJF9Zv379VP/+/f3qLV++XAHq888/V0opVVRUpEJDQ9XMmTP96u3bt09ZrVb18MMP+8qqet9OBajAwEB14MABX9mzzz6rANW+fXtVVlbmK7/hhhuUxWJRLperxvFcffXVfp+X446/Zu3bt1elpaV+245//vfv36+UKn/d7Ha7GjBggF9cSiml6/ppr/P4e3vy++31etUVV1yh7Ha73+fy1M+gy+VSSUlJaujQoX7lgLJYLGrfvn2+sk2bNilA/f3vf/eVTZs2TbVs2dLv86eUUqNGjVKxsbHK6/VWer2LFy9WgMrMzDzttQkhTk9uwQtRS5dffjlRUVHEx8dz++23k5CQwJdffklQUBAff/wxERERXHrppWRnZ/seQ4YMwWg0VhhYNHToUDp06FBvsU6YMMHX8nOqKVOm+LV8TZ48mfDwcJYsWeIrO97SB+V9AbOzsxkwYABKKTZs2FDjeI4cOcLGjRuZNm0aYWFhvvL+/fvTv39/vv766wr73HnnnX7PBw8e7Gt1rolNmzaxZ88ebrjhBr/3xu12M3jwYL/3ZuLEiaxZs4YDBw74yhYtWkRoaCgjR44EYNmyZRQUFDBx4kS/4wUHB3PRRRfVehDZiBEjaNWqle95v379gPL3x2Kx+JU7nU6OHDlS5/FMnTr1jH13ly5disPh4I9//KNfXIBfK//pzJgxw/dvg8HA3XffjcPhYNWqVb7ykz+DeXl5FBQUMHjwYNavX1/heCNHjiQxMdH3vHv37oSEhPg+L0opXyu2x+Pxe51GjRrFkSNH2LVrV6WxHv+8Ll68GF3Xq3V9QoiK5Ba8ELX02muv0bZtW0wmE3FxcX4J5O7du8nNzfW7jXyyrKwsv+cn/1jWh9Mdv3379n7PTSYTCQkJfn38Dhw4wJ///Ge++OIL8vLy/OrXZuqptLQ0ADp27FhhW6dOnfjwww/9yux2OxEREX5l4eHhfrFkZWX59Qm12+0VbikDvn6UN954Y6WxnTxAa+LEifzhD39g0aJFvj6nn376KVdffbUv2Tp+vCFDhlR6vDZt2lRafiatW7f2ex4aGgrgl5SeXJ6Xl0fr1q3rNJ7qfC6PJ3VJSUnVPu7Jjn/eTnb8M3nyZ/Crr77iySefZOPGjTidTl95ZUnuqa8d+H9esrKyyM/P5x//+Af/+Mc/Ko0rKyuLTp06VSi/9NJLue6667jzzjv54x//yNChQxk3bhyTJk0iICDgjNcrhCgnCagQtdSvXz/fKPhT6bpObGws//73vyvdHhcX5/f85Nad6tA0rdLpnqoalFPT4596zOHDh5Obm8vDDz9Mp06dsNlsHD58mKlTpzZIK1B1Rl337dvXl9gCzJkzp9I5SY/H++KLL54xaWrZsiWXXHKJLwFdsWIF2dnZTJo0qcLxPvjgg0r/4Kjta1/VNVdVfvzzUJfxnM3npi6tXLmScePGMWTIEP7xj38QGxuL2Wxm4cKFfPDBBxXqV/c1mjp1apV/iFT12dA0jY8//pg1a9bw5Zdf8s0333DLLbfw/PPP89NPP1X6R48QoiJJQIWoB23btmXFihUMHjy4wm3J6jrd7cvw8PBKbz+fnIBV1/EWs+M8Hg+pqakMHz4cgC1btrBr1y7eeecdbrnlFl+9ZcuW1Sjmkx2fQ3Lnzp0Vtu3cubNWc0y+//77fqPSq2rpa9u2LVB+K3XYsGFnPO7EiRO5//772b9/P4sWLSI8PNz32px8vNjY2CaxylFN4qnu+1Wd86WkpFRoyayO45+3k9+v45/J45+DTz75hMDAQL755hu//08LFy6sVcxRUVEEBwejlKrWZ6Ay/fr1o1+/fjz55JMsWrSISZMm8dFHH3HbbbfV6nhCXGikD6gQ9WD8+PG4XC6eeeaZCtucTieFhYVnPIbVakXTNPLz8ytsa9u2LTt27PC7lb9p0yZ++umnGsf6zjvv+E0N9cEHH5CXl8fo0aOBE61JJ7e4KqV8Uy6dzGazAVQa88liY2Pp0aMHCxcu9LuF/+uvv/LLL79UGNVcHQMHDmTYsGG+R1UJaK9evWjTpg3PP/88JSUlFbaf2j1iwoQJGAwGPvjgAxYvXsy1117rmx0AyvtqhoSEMG/evErnb83Ozq7xtZyNmsRjs9nOevWu4cOHY7fbmTdvnt+tcaDaizK88sorvn/rus6rr76KzWZj8ODBQPlnUNM0vxb+1NTUWi8jazQaueaaa/joo48q/SPo1M/AyfLy8ipc1/E7IWVlZbWKR4gLkbSAClEPLr/8cm6//Xbmzp3LunXruOKKKzAYDOzatYtFixbx/vvvn7HlJSAggG7duvHRRx/RoUMHIiIiSEpKIikpiWnTpvHXv/6VkSNHctttt5GZmclrr71G165dq5Xcniw0NJQhQ4YwZcoU3zRMSUlJvluTnTp1om3btvzhD3/g8OHDhISE8Mknn1ToCwrQu3dvAP7v//6PkSNHYjQauf766ys97/z58xk1ahQDBw5k2rRpvmmYYmNjmT17do2uoSaMRiNvvPEGY8aMoVu3bkyZMoXY2FgOHTrE0qVLadeund80UnFxcQwcOJBnn32WoqIiv9vvUP76LViwgClTptCnTx8mTZpEs2bNSE1N5YsvvuB3v/sdTz75ZL1dz6lqEk/v3r15//33eeCBB+jbty92u52rrrqqxud74YUXuPPOO+nXrx/XX389ISEhrF+/HovF4pdcViYoKIjPP/+cvLw8evfuzVdffcW3337L008/7budPWbMGP76178yatQoJk+eTGZmJq+88grt2rVj8+bNtXqdnnnmGVasWEGfPn2444476NSpE9nZ2axdu5b169dXufb7O++8wz/+8Q9+97vf0bZtW4qLi3nzzTcJCQnhyiuvrFUsQlyQGmv4vRDnquPTsmzYsOG09XRdV6+++qrq2bOnCgwMVKGhoapHjx7q0UcfVVlZWb56gLrvvvsqPcbq1atVnz59lMViqTAl03vvvafatGmjAgICVI8ePdQ333xT5TRML774YoVjH59qJzk5WT300EMqKipKBQUFqWuuuUYdPnzYr+62bdvUsGHDlN1uV5GRkWr69Om+qW1OnlrJ6/Wq++67T0VFRSlN03xTMlU2DZNSSi1btkxdcsklvtfnuuuu85s+R6nyqXpCQ0MrxD9nzpxKp3yq6jqPT8N03Lp169TVV1+tIiIilMViUYmJierGG29Uq1atqnCMBQsWKEBFRkYqj8dT6XmWL1+uhg0bpkJCQpTValXt27dX06dPV1u2bPHVqck0TKd+JjZs2FDpa1jV57E68RQXF6ubbrpJhYeHK8D32Tn+mi1evLhCbKdOS3Tc4sWLVf/+/ZXValWhoaFqwIABvqmqqnL8vd29e7caNmyYslqtKi4uTj399NMV6v7rX/9S7du3VxaLRXXq1EktXLiw0s9AVf+f4uPj1ZQpU/zKjhw5ou666y7VsmVLZTabVWxsrBo5cqR69913q7ze9evXqxtuuEG1bt1aWSwWFR0drcaOHat+++23016rEMKfplQdLVwthBBC1MDUqVP57LPPzthlQwhx/pE+oJWo6S0oIYS4UMj3oxCiLkgCWgnpSC6EOBfdcccdtGjRgpCQEBISEpg3b55vW0JCAlar1Tc/ateuXWt1Dvl+FELUBUlAhRDiPHH//fezZ88eCgsLWblyJe+99x6LFi3ybV+8eDEOhwOHw8HWrVsbMVIhxIVORsELIcR5okuXLn7PDQYDe/bsaaRozuztt99u7BCEEI1EWkCFEOI88sc//hGbzUbr1q0pLi7mpptu8m2bMmUKUVFRXHbZZdWaM3b+/PlER0f7PSqbO1UIIWqqSSSgCxYsoE+fPlgslirnDDzuhx9+ICkpiaCgIPr27cumTZsqHKtFixbY7Xauu+66SucqFEKI89XTTz+Nw+Hg119/ZfLkyYSHhwPw3nvvkZqayoEDB5g0aRKjR48+48pZs2bNIjMz0+8RFBTUEJchhDjPNYkENC4ujscee4zp06eftl5OTg5XX301Dz30EHl5edxwww2MGzfOt/rGsmXLmDNnDl9++SVHjhzBaDRy11131WvsabklbE4vxO2t//WwhRCiOjRNo2/fvgQGBjJnzhwABg0ahNVqxWq1cvfdd9OzZ0+WLFlSbzG4vB4256Zz0JFfb+cQQpy7mkQf0GuvvRaAjRs3nnbZuk8//ZR27dr51qP+/e9/z4svvsjy5csZM2YMb7/9Nrfeeiu9evUC4KmnnqJLly4UFBQQGhpapzErpfjbyv18uuUIqbklmAwG/j35IgYmRtbpeYQQorY8Hg979+6tdJvBYKj2Upk1ddCRz1Obl5NRWoRZMzI0rh13dxpQL+cSQpybmkQLaHWlpKT41tyF8r/yu3fvTkpKSqXb27dvT0BAADt27KjymLXt47QpvZCPNqbzc2ouB/LL2JdbwqAFvzD1g99qfX1CCFFbeXl5vPvuuxQWFqLrOj/99BOvvvoqw4YN48CBA6xcuRKXy4XL5eKf//wna9euZcSIEfUSy1u7f+XHjH1szEnnt5yDvL93PdvyM+rlXEKIc9M5lYA6HA7CwsL8ysLCwigqKqrW9srUto9TRpGTnZlFeE658/7OuiNoD35ZresRQoi6omkaCxcuJD4+ntDQUG677TYefPBBZs6cicPh4N577yUiIoKYmBjeeecdvvrqK9q2bVsvsfyQsZcSjxsApeBQcQHrsg/Vy7mEEOemJnELvrrsdjsFBQV+ZQUFBQQHB1dre13qHhuCS6/69pX24JcUPTkSuzWgzs8thBCnCgsL47vvvqt0W5cuXdi4cWODxWIx+v+0aBpYTeYGO78Qouk7p1pAk5KS/L5ElVJs3ryZpKSkSrfv2bMHp9NJp06d6jyWuNBAbunV4rR1gh/7hn+s2Fbn5xZCiKZsbKsuRFvtGDSNQJOJDiFR9IuMb+ywhBBNSJNIQD0eD2VlZXg8HnRdp6ysDLfbXaHetddey+7du3nvvfdwuVy8/PLLAAwbNgyAqVOnsnDhQjZs2IDD4eCxxx7j2muvrfMBSMe9Mv4iukSf/nb9jK/2EvMHuSUvhLhw3NKuN9fEJ9E/Op5B0Yk8mHQprexhjR2WEKIJaRIJ6JNPPonVauWpp54iOTkZq9Xqm5LJbrezcuVKAJo1a8Znn33GM888Q2hoKO+//z5ffPEFFosFgOHDhzN37lzGjBlDTEwMLpeL1157rV5j3/rwFWesc1Qh/UKFEBeMYHMgf+x+Bf8cMJ7XBlzHiBYdGzskIUQTo6n6mofjHDZ8+HCWLVtWo32qm2CqF66qTUhCCNEk1Ob7UQghTtUkWkDPB9VNLLUHv6y3ufeEEEIIIc4FkoDWoeomoYY/fMX873bXczRCCNE4lFL89+B2/rz+f7y49UdSi3IbOyQhRBMjCWgdq24S+tDXO7jjw1/rORohhGh4Xx3cxnv71rOnKIe12Qd5ctNyitxljR2WEKIJkQS0HlQ3Cf3n2qOMXLAC72nmExVCiHPNqqP7/Z6XeN0yEb0Qwo8koPWkukno0v0ObLO+wuPVz1xZCCHOAXazpUJZsDmwESIRQjRVkoDWo+omoU7A/NDXHMw78xr0QgjR1F0Tn4RJO/Hz0j4kkh4RcY0YkRCiqTmnluI8F6kXriL4wS9xVKNu6ye/5Z5LWjNraHsSIs68Hr0QQjRFSeGxzO87lnXZhwi3WOkb2QqjQdo7hBAnyDdCAyh64Sq6hFSv7j9+OUCvv/7A8l2Z9RuUEELUo+bWYK5s1ZlLohMwGYyNHY4QoomRBLSBbJ1zFXf1iqpW3bxSDze9t47MImc9RyWEEEII0fAkAW1Ar97Yn8eHJlSr7tFiD/3/tooD0i9UCCGEEOcZSUAb2J/HdOPzqb2rVXd/bgk9X/ierUcK6zkqIYQQQoiGIwloIxjXLY5PbuxRrbq5pV6Gv/YLBSVyO14IIYQQ5wdJQBvJtb1a8b9pPatV94jDxdVv/yZryAshhBDivCAJaCMa2bUluX8ZUa26P+zNpfXj3/DJ5vR6jkoIIYQQon5JAtrIwoMsfDWten1CDxW5Gf/OOno+/z0Pf7WNDzccxuWRFZSEEEIIcW6RBLQJGNM1jmW39612/Y1Hili45gCfp2SwcO2BeoxMCCGEEKLuSQLaRAzrHIPruTHYq7k2VVaJm882p/PC9/tYsv1o/QYnhDgn3HHHHbRo0YKQkBASEhKYN2+eb1tKSgr9+/cnKCiILl268N133zVipEKIC50koE2I2Whg96PDCQvQqlW/TIfd2cX8/vOtrDuYX7/BCSGavPvvv589e/ZQWFjIypUree+991i0aBFut5urrrqKcePGkZeXx5w5c7jmmmvIzJQV14QQjUMS0CYmJiSQ3HljGNclstr77Mwq5tYPN+BweuoxMiFEU9elSxesVqvvucFgYM+ePXz//feUlJQwe/ZsLBYLkyZNIikpieTk5EaMVghxIZMEtAnSNI3Pb7uEzpFB1d5ne6aDGZ9spqDUXY+RCSGauj/+8Y/YbDZat25NcXExN910EykpKXTr1g2D4cRXfo8ePUhJSTntsebPn090dLTfo6REVmcTQpw9SUCbsG/vuYRLWodWq65Hh/fXH2bQglVszSiq58iEEE3V008/jcPh4Ndff2Xy5MmEh4fjcDgICwvzqxcWFkZR0em/K2bNmkVmZqbfIyio+n8YCyFEVSQBbcJiQ4P4+b4hHHhsaLXqexWkZDgYvGAln8l8oUJcsDRNo2/fvgQGBjJnzhzsdjsFBQV+dQoKCggODm6kCIUQF7omkYDm5+czceJEgoODiYuL46WXXqqy7rvvvkvHjh2x2+0MGTKEnTt3+rZ9//33GAwG7Ha773HyKNBzVatwG1/fVv1pmvJKvUx+fwNTPtiArsvqSUJcqDweD3v37iUpKYktW7ag6yfmDd64cSNJSUmNGJ0Q4kLWJBLQmTNn4nQ6OXz4MN988w3z5s1jyZIlFeqtWrWK++67j//85z/k5+czdOhQxo0bh8dzYvBNdHQ0DofD93jkkUca8lLqzZVdYripZ1y165d6dD7ceJgXfthbj1EJIZqKvLw83n33XQoLC9F1nZ9++olXX32VYcOGcdlll2G1WnnuuedwOp0kJyezZcsWJkyY0NhhCyEuUI2egBYXF5OcnMxTTz1FSEgI3bp1Y/r06bz11lsV6n7++eeMHz+eXr16YTKZeOyxx9i/fz8rV65shMgb3rs39WbhxO7VftNcXsVTy3bL6HghLgCaprFw4ULi4+MJDQ3ltttu48EHH2TmzJmYzWa++OILFi9eTFhYGH/+85/59NNPiY6ObuywhRAXqGpOe15/du3aha7rfreCevTowaefflqhrlIKpZTfc4DNmzdz+eWXA5CTk0NMTAwWi4VRo0bx9NNPExERUeX558+fz/z58/3K2rdvf1bXVJ+m9osnp8TNH77aXq36BU4PrZ9Yyt+u6UZMSCADEsIJCmj0t10IUcfCwsJOO7l8t27dWLNmTYPFU+gqY2PuYcICrHQLj0XTqje/sRDiwtDoLaAOh4PQUP+R3lWNzrzyyitJTk7m119/xeVy8fjjj+PxeHzTgnTq1ImNGzeSnp7OqlWrOHToEFOmTDnt+c/FUZ4PXt6O9yf3JNhirFb9vDIvD3y+lWe/28MfvtzG+kP5HC1y1nOUQogL1e6CLO7/9XNe27maZ7as4NktK/waD4QQotGbwux2O4WFhX5lVY3OHDp0KM899xy33HIL2dnZTJkyhS5dutCyZUsAYmJiiImJAaBVq1YsWLCAdu3aUVJS0uSTypqa3LslFrOBm95bR5n3zPWzStys2p/DmgN5bE4vpHV4EEPbNeP2/vH1H6wQ4oLycdpmyrwnuv5szjvCptx0ejRr0YhRCSGakkZvAe3QoQOaprF161Zf2elGZ95xxx3s2LGD7OxsHnvsMVJTU7n44osrrWswGCrctj+fXNc9jtfGX0SQuZpLd3oUDqeX7ZkOipwevtuTwzaZM1QIUcdynRUnq89zlTZCJEKIpqrRE1Cbzcb48eN59NFHKSoqIiUlhTfffJNp06ZVqOt0Otm0aRO6rnP06FFuv/12rr32Wjp27AjAihUrSE1NRSlFRkYG//d//8eIESOw2WwNfVkNZsrFrVk1YyBdm9sxVSMPVUCJ08PWjCK8uuJQgfwoCCHq1sWRrf2eBxiM9Iio/iweQojzX6MnoACvvPIKZrOZ2NhYhg8fzuzZsxk9ejRQfov++Ch3l8vFlClTCAkJISkpibi4OF5//XXfcTZs2MDgwYOx2Wz07t2byMhI3n333Ua5pobUs1U4KQ9dTvItvTEbzpyFlnkV+3NLWLwlnTdXp/HP1Wl8vycbr8wZKoSoA9fGd+N3rbsSYw2mc2g0s7sNJdxyfnWDEkKcHU2dr/enz8Lw4cNZtmxZY4dRK8Ne/Zlv9+RUu74BSGwWRKswK8M7RPLIsA71F5wQ4px3Ln8/CiGajibRAirqzuMjO2IzG6juhCc6cCCvhH05Jfxz9QHWHcyvx+iEEEIIISQBPe8MbNOMh4e2J9oeUK0+oQBuHTKLyjhS6GTGp1soLHXVb5BCCCGEuKBJAnoe+tOIDvx6/2DuGZhApyhbtRLRMq/CretsO1rEtI8213+QQojzmtPrYX32IfYXVb9LkBDiwtHo84CK+tE6PIhnxnYh2r6XF37YS17pmZfjVArK3Dob0wvYneWgfZS9ASIVQpxv0hy5zNv0HUWe8gUvLomKZ2bngbIakhDCR1pAz2NWs5FHh3fg9n7xRNkCMBs0NPD1Dw0wan6j5hXg1hXZDhdZxbJSkhCidhbt3+RLPgF+yUpja35GI0YkhGhqJAG9AAxtH8mAhHBah1uJsgUQbDESay9PSN2VTL1U6PRwzydb+Pfag7i9ut+2bIdTlvEUQpzW0VJHhbLMSsqEEBcuuQV/Abi8XTN+PZCPxWQgr9RNy1ArrcMC+ftPqZR4dE6diEsBW44U8YcvtvLltqPMGdGB9IIyfj2Yz7aj5T8iSTHBPHBpGwLN1VuPXghx4ejZrAXph04ssWzUNLrLRPRCiJNIAnoBsJiM/HlEB/ZmF+NVivaRNib8+zdKXV40yhPOU+kKSjxelu7MYntGEVaLkSyHiw5RNiJtFlIyivjfjkx+1y22oS9HCNHETUjojtPrYXVWGuEBViYl9iAy8PxdkU4IUXOSgF5A2kae+AGIDQ7kTJOFFrt0QGdvbjFBZhM6irS8UiJtFgBS82QZTyFERQFGE9M6XMy0Dhc3dihCiCZK+oBeoJ4c1QFbgBENDeMZPgUer6LU7cXh9JLlcKEfu2ffKVpGyQshhBCi5iQBvUCFBll49drutAyzEB5oxnSaT4JHgdujY9BAV4ptGYUUlbnZmVnE7iwZWCCEEEKImpFb8Bewa7vHsienmM1Hitid7WDT4cJKR8UDYIAwq5m44AD25JSSX+Zhe6aDFXtyePHqJBKbBTVs8EIIIYQ4Z0kCegEzGDQeHtqO7UcdFJS52ZZRxB++3Ep+mdevngbYAkz0bhnK6rQ83F5FTrEbTQOnR+edtQcZ3jGKni1CCAqQj5QQQgghTk+yhQucpml0iQkG4OLW4fx3RybLdmbhcJUnoQEmDaNmIDEiCKXA7VUcX8xEKTha5OSzlAx2ZRcTEmhizogOxIYENtblCCGaiCJ3GRtz0gm3WOkaFiOrIAkh/EgfUOFjNGiM7x7HpW0jSYiwEmY1E2Q2MTAhnNeu60bPFqEYDeDRddxeHbeugwaB5vKPUV6pixe+38u3u7IoLHM38tUIcWFxOp3cfvvtJCYmEhwcTNeuXfnggw982xMSErBardjtdux2O127dq23WPYUZnPfms95decvzNv8HfNTvkedOuGwEOKCJi2gws/oztGsPpBHcKAJh9NDc7uFF6/uwtpDBaTmlRITHOibT9Ro0AgwaoQGmvHoOluOFLLjaDHphU4+3JjO3JEdyCl28+O+HKxmI6M6RdEi1NrYlyjEecnj8RAXF8e3335LYmIiP/30E2PGjCExMZFLLrkEgMWLFzNq1Kh6jyU5dRNlXo/v+cbcdDbnHeEimYxeCHGMJKDCT5jVzPNXdWHj4UJMBo3ucSEYDRobDhcAEBRgxGYx4fYqLCYDRgO4vTpHi5xkOVxEBAWQV+ICAvjHT6mk5pb6Jrr/JTWP567qTERQQKNdnxDnK5vNxhNPPOF7PmjQIAYOHMjPP//sS0AbSq6zpEJZTllxg8YghGja5Ba8qMBsNNC3dRg9W4ZiNJT324oJLp98XlflSWpEkJmk2GB6twwjzGpia0YRBWUeUnNLWLk/h4N5xWxML/RbZSm3xMXbaw+SXyq354Wob8XFxfz2228kJSX5yqZMmUJUVBSXXXYZP/300xmPMX/+fKKjo/0eJSUVk8tTXRzZ2u+5WTPQs1mLml+EEOK8JQmoqJYxnZsTG2Ih0lbeehlmNfued20egltXKFWeoJa5FZuPFNH22NRMHl1nw6ECvtuTzRur07jr4038vD+3MS9HiPOarutMnTqVvn37MmLECADee+89UlNTOXDgAJMmTWL06NGkpaWd9jizZs0iMzPT7xEUdOYp166N70aXsGjSSwoo87i5o2N/wi0yVZsQ4gRJQEW1hFrNPDe2Cy9d3ZW5IztwVdfmtG9mIyTQRPLmdIpdXgwGMGjlD03TmNyrBUYD7Djq4GBBKV6lKHF52XbUwb/XHcJb1ZyjQohaU0px1113kZ6ezkcffeQbfT5o0CCsVitWq5W7776bnj17smTJknqJ4YeMvfyadZASt5tdhVnM3biUg478ejmXEOLcJAmoqDajQSMpNoSpfVsz78rOtA4PIr/UQ5Q9AA3QdTAZNcxGAy1CLGQWubimawxmo4GAYw+XRyen2E1abillbu8ZzymEqD6lFDNmzGDjxo0sWbIEu73q5XINBkO9jUx/d+861uccJiU/g31FufyadYC7fk7maGlRvZxPCHHuaRIJaH5+PhMnTiQ4OJi4uDheeumlKuu+++67dOzYEbvdzpAhQ9i5c6ff9gULFtCiRQvsdjvXXXcdeXl59Rz9hWv70SIO5pdQWOoh3GrGaACTQSM6OIBgi4kPN6bz4aYjFDndeHSdEqeHnBI3WQ4nB/JLeOGHfZRKEipEnZk5cyarV6/mm2++ISQkxFd+4MABVq5cicvlwuVy8c9//pO1a9f6bs/XtW35R8kpK8ale3DpHsq8HrKcJXybvrtezieEOPc0iQR05syZOJ1ODh8+zDfffMO8efMqvTW0atUq7rvvPv7zn/+Qn5/P0KFDGTduHB5P+XQfy5YtY86cOXz55ZccOXIEo9HIXXfd1dCXc8FIyy/lYH4ZBU4PQQFGWodbmTuiI/1bh9Mm0gaUJ6SFZR6MmoZbV3h1haJ8UNOOTAf/25HZuBchxHkiLS2Nf/zjH2zbto1WrVr55vucN28eDoeDe++9l4iICGJiYnjnnXf46quvaNu2bb3E4tF1vEpHVwpdKby6jga4dPmDUwhRrtGnYSouLiY5OZl169YREhJCt27dmD59Om+99RajR4/2q/v5558zfvx4evXqBcBjjz3GvHnzWLlyJZdffjlvv/02t956q2/7U089RZcuXSgoKCA0NLTBr+18VljmRkPDFmCk2OVF0yAmOJBbL25FSkYRZR4dKL8laDYa0BVYzUY8usKgaWQUOWkeHMiBvNJGvhIhzg/x8fGnvaW+cePGBoslwGjEq5RvFgwv5UnokJg2DRaDEKJpa/QW0F27dqHrut9UIT169CAlJaVCXaWU3xfs8X9v3rwZgJSUFHr06OHb3r59ewICAtixY0eV56/tNCMXOpNBIyjAwEVxofRsEUrflmEkRFgxGw30jw/31XN7dUwGDV0prGYjBk3DZNTwHBuAlHRsGVAhxPmj0FWG2WBAA9+jzOtBl9WQhBDHNHoC6nA4KrROhoWFUVRUsbP6lVdeSXJyMr/++isul4vHH38cj8fjSxgdDgdhYWHVOtZxtZ1m5EIXFGBieIcooLxl02Q0cHnbSOwWE1P6tmRM52ia2cykFzrpEGnDaNDw6DpR9gAijs0jOqpTFEPbRzbylQgh6prZYMSt6yhAAUbNQIDRxI8Z+xo7NCFEE9Hot+DtdjuFhYV+ZQUFBQQHV2wZGzp0KM899xy33HIL2dnZTJkyhS5dutCyZUvfsQoKCqp1LHH2bu7dks7RwezOdtC2mY2LW4cBYDEZubF3S0ICTeQUl0863zUmhAN5JYQGmrm2ewz3DEgg/KQVkXJLXHy97SjZxS76tApjcJtmjXFJQog64NI96CctQ+FROsFmC1aTuRGjEkI0JTVuAfV4PFx//fWkpqbWSQAdOnRA0zS2bt3qK9u4caPfLfmT3XHHHezYsYPs7Gwee+wxUlNTufjiiwFISkry6+e0Z88enE4nnTp1qpNYhT9N0+jbOozJvVrSLz7cN9/gcSdP8xkTbKFPqzAm9ojl0WEd/JJPt1dn7je7WLIji7UHC3j15zS+3na0oS5DCFHHCpxlGPH/Psh3ljIsrn0jRSSEaGpqnICaTCby8/PrLACbzcb48eN59NFHKSoqIiUlhTfffJNp06ZVqOt0Otm0aRO6rnP06FFuv/12rr32Wjp27AjA1KlTWbhwIRs2bMDhcPDYY49x7bXXygCkRjIwIRyr+cRHzKhpjO7UvEK9TemFZBe7fM8Lytz8a80BDhfIACUhzkUGgwEFvhRUAyIDgwg1BzZiVEKIpqRWfUDHjx/Pu+++i9dbN1NqvPLKK5jNZmJjYxk+fDizZ8/2jYC32+2sXLkSAJfLxZQpUwgJCSEpKYm4uDhef/1133GGDx/O3LlzGTNmDDExMbhcLl577bU6iVHUXKTdwhOjOjKsfSSD20Tw6LD2dI8LYePhAj7ccJhfUnPRdYXJcKKlZF9OMVszitie6eChL7fzw97sRrwCIURttA1uhuHYHRGN8j6gXqX47siexg1MCNFk1KoP6O7du1m9ejWrV6+mXbt2WK1Wv+3PPPNMjY4XFhZGcnJypdscDofv38HBwWecSmTmzJnMnDmzRucX9adFqJVp/Vr7nn+yOZ1PNmf4nq9PKOCuAQnEh1vZmekgo8h5bL9AFPDRxnQubVv1QKUDeSVoaLQKt1ZZRwjRsHpHtmRvYQ5ZzmKMmkaw2YLJYCDNIQuDCCHK1SoBDQoKYujQoXUdizjPebw6X2/zn3j+59Q8JvWI488jOvCf9YdJLywj0GwkIqh8sEJhmQddVxgM/v3Jytxenluxlx2Z5X+gJMUE8+BlbbCYjA1zMUKcJZfLxc6dOwkJCSE+Pt5vm9PpZNmyZYwdO7aRojs7w+M6sDHnMI4cJyVeN/muUrLKiukcFt3YoQkhmohaJaBz5syp6zjEBUBX4Nb18vXgS1wYNY1mtgCcXp1Is5FeLUIpKHOzP7eE3VkanaPtXJ0Ui8GgkV/qZu2BfIICjPRtFcbSnVm+5BMgJaOIFXtyGNVJfuBE03fw4EFmzpxJeno6mqbRu3dv/vKXvxAZWd7a73A4eOKJJ87ZBHRUy06sPLKX1VkHgPJb8OklBRwpLjzDnkKIC0WjT8MkLhwBJgOdou289etBvMeGyBe7vIRbzSileP6HvRS7vLi9ijK3zt6cEm7qFUdabglPLNtFqbt8daVWYYHEBlsqHP90qyrpumJfbgkhFhPRlewrREP629/+RmJiIu+88w5FRUX89a9/5bbbbuP1118nJiamscOrE2kl+UQF2n0Lhmiaxn8P72BS256NHJkQoilo9InoxYUlNNBMi5BAAowGXF6dI0Vl/OGLbWQ5nGxOL8SgaYQEmggPMqOANQcL+GLrUV/yCXAwv4zK1lOpalWlzCInD365lT//byf3f76VN1en1c/FCVFNmzZtYubMmYSFhdGqVStefPFFBgwYwO23386hQ4caO7w6EWQKoMTrwql7fMPhm1vtjRuUEKLJkARUNKhSt5foYAtuXSfAWP7x23ykkE+3ZGAxG1AKil0e8kvdlLq96Lqi0OmucJzEiCDGdI7GYjIQaDJwddfmXJIQXqEelA9kOlp0Ypqn7/bksDWj6tWxhKhvTqcTo9G/v/LDDz/MkCFDuPPOOzlw4EAjRVY3Nuemk1nmwKPrFLmdFLjKiAy0MbPzoMYOTQjRRMgteNGgBiZEsGxXFseXhNaASFsAm9ILue3i1sz9ZhclLh2jQSPAaGDl/lxGdoxia8aJ/p4mg0a/+HBiQwK5oWcLgAqDlE52qJL5RA/ml9JV1qEXjSQ+Pp5t27aRmJjoV/7QQw+hlOKBBx5opMjqxmcHUggyBXBpTFsOFeejafBkr9G0tlf+R6IQ4sIjLaCiQQ1IjOD2fq0JtpgIs5rpEhNMUICJuNBAbuzVkg7RNlqFBdK2WRA9W4SSlldK5+bBTLu4Fe0ig+geG8zsoe2IDSmf0Npg0E6bfELFW/Maij3Zxdz20UbuTN7El1szqthTiPpx+eWX880331S67eGHH2bkyJG+vpPnomJ3+R0Hi9FEu5BI2oVEYTFKe4cQ4oRafyOsXr2adevWkZubi67rfttklLw4nYk9WlDm0fnfjiwAgi1GbuzVAovJQNtmNlqGnpjTUwMsJgPDOkQxrENUrc434aI4Cso8rDmQR7DFREJ4EMt3ZRFoNmLQNP6zIZ1WYVZ6tJAVs0TDuPXWW7n11lur3D579mxmz57dgBHVrb5RrVm26VtyXSWYDUY6hETSu1nLxg5LCNGE1CoBffXVV1m4cCFJSUlERkZWWANciDO5pU8rRnSIIqvYRccoOwGm8sb4sV2i/Saqv6J9JGFW81mdK9BsZOagRO7RE9h6tIg7kjdxpNCJyaDRLtJGRFAAm48USgIqmow9e/Zwxx138N133zV2KLVS4ColNCAQj/ISYDARYDBS5HZiN8sMFEKIcrVKQBcvXswTTzzBqFGj6joecQGJCQkkJsR/bejrusfRMcrOf7dn4vLqdIiyVToRfW0o4PVf0jBoGkopSt06OzMd9IsPJy5E1qgWTYeu636rwJ1rtuUfJTYohNigEACUUmzOS/c9F0KIWt+C79KlS13GIYTPpvRCNqaXT1i97aiDlIwi7hqQcNbHLShzk1viJjTQxDanB5dHgQYlLg9D2jQ76+MLIcq1soVxpLQIpRQ78zPJdDoo9jjZVZDN9A79CDSd3V0NIcS5r1aDkG644QY+/vjjuo5FCJweL0t3ZfmVrdyXS16Jq4o9qi/caibKHsCh/DJCLGbCrCaibAHYAkxkOpxnfXwhRLkb2vQkxGTh64Pb2JiXzpGSIlZm7OeHjL0sSt3U2OEJIZqAWrWA7tixgzVr1rBq1SratGmDyeR/mGeeeaZOghMXHq+u8Hj9R/8qwKOf/YhgTdO4d1Ai3+3ORtPAbjHRPtKGpmkcLiijZZj1zAcRog6UlZWddrvTeW7/QdTcGkyBq5QiT/kfjgpFvruM9TmHSAyOaOTohBBNQa0S0KCgIC6//PK6jkUIggJMXJIQzs+peb6ypJhgoux1M3ihXaSN6f1bs3JfLmajBmiYjRqdomWFFtFwBg8efNrBm0qpGg/udDqdzJgxg2+//Zbs7Gxat27No48+yuTJkwFISUnh9ttvZ/PmzSQkJLBgwQKGDh16VtdxOiuP7q9QdrTUQStbWL2dUwhx7qhVAirTLIn6dOcl8cSHW9mR6aBNsyDGdG5eo/13ZTk4lF9G15hgmley7vutfVvj9io2HC4gyh7A5J4tWLk/l/WHyp9f2y220v2EqCuvvfZanR/T4/EQFxfHt99+S2JiIj/99BNjxowhMTGRPn36cNVVVzF9+nR++OEHPvvsM6655hp2795NdHR0nccCEGgyo4HfsrkWo4lJiT3q5XxCiHOLps5ituNDhw6xf3/5X7mJiYm0bHl+zPM2fPhwli1b1thhiFp4a80Blu/OBsrnEL1nYAIDEyu/5Xe8lemjDYdZnJKBw+nBbDQQFxLIE6M6EhJowmyUtRrEuevKK6/kiiuuoHv37tx0000cOXIEg6H8Mz1w4EAmT57MjBkzanTM6n4/Pr5hKU9uWo5Hlc8TbUTj8R4jebTnsJpfiBDivFOrFtCioiKeeOIJvv/+e8zm8tGMHo+HSy+9lDlz5mC3y+1M0fAyi5x8eyz5hPKWl/9sOFxlAnr8Fuf/dmSy4XABTo+OrhRrFezJduD2KmwWI52ig7kmKYYBVRxHiKaouLiY3377jfvuu4+UlBS6devmSz4BevToQUpKymmPMX/+fObPn+9X1r59+2qdf0hMIlE7bRS6ytBRWI0BtA+teraJlRn7WJGxh0CjmTEtO9M1PKZa5xFCnJtq1bzz3HPPcfDgQd5++21+/vlnfv75Z9566y0OHjxY4ctKiIaSX+bm1Ob8/FL3GZc0TM0rxekpb6UpdnrJK3GTXlDGvtwSthwpYne2g1d+SiUtt6SeIheibum6ztSpU+nbty8jRozA4XAQFhbmVycsLIyioqLTHmfWrFlkZmb6PYKCgs54/lxnCc9t+R6DZkAHTJqRIJOZv25dyYacwxXqr8lK49Wdv7CjIIuNuek8s+U7Djryq3/BQohzTq0S0FWrVvHoo4/StWtXX1lSUhJ//OMf+fHHH+ssOCFqom0zG5G2AL+yi1uHnXEwx8n9Pd26ToBRo9jl9ZUVlXlQwPrDBXUarxD1QSnFXXfdRXp6Oh999BGapmG32yko8P/8FhQUEBwcXC8xvLtnHbsKssgqc1DmdVPsdVHmcWMyGPjy4NYK9VedMmDJqxSrs9LqJTYhRNNQqwTU4/EQGFhx5ZjAwEC8Xm8lewhR/4wGjYcvb0tChJUQi4lh7SOZ3j/+jPuN7BjFRXEhtA6zEhsSSKjVTMhJy3/aLeU9VaLraCS+EPVFKcWMGTPYuHEjS5Ys8XWHSkpKYsuWLei67qu7ceNGkpKS6iWObw7voMjtxKV7UYCuFDmuEoyahquS3wi7qeL/Lbs5oEKZEOL8UasEtE+fPrzwwgvk5OT4yrKzs3nppZfo06dPnQUnRE3klrh4aeV+UnNLKXR68OiKQNOZP+I39GzB8A5RxEdYGZQYwcCEcBIjggg0GWgREsjhgjK2ZxaxaGM6aw/k1/+FCFFLM2fOZPXq1XzzzTeEhJxY9vKyyy7DarXy3HPP4XQ6SU5OZsuWLUyYMKFe4nDpXorc/nOZKmBd9iF6N6s4WHVMq84EGU/80dc80M6Q5m3qJTYhRNNQq1HwR44c4YEHHiA1NZXY2FhfWWJiIn/961+JialZ5/H8/HzuuOMOlixZQnBwMA899BD3339/pXUXLVrE3LlzOXjwIDExMcyePZvbbrsNgNTUVBITE7HZbL76N910U42nPJFR8Oem139J5Ye9uX5lDw9ty0VxodXa/+S5F48UlhFoMvDSj/tZdyifkEAToGE0wIvjuhIpraGijqxZs4bly5eTkZGB2+3221aT7660tDQSEhKwWCx+i4M88sgjPPLII2zZsqXCPKBXXHFFjeOtzvfjExuWMm/ztzh1/9ZOuymAZ/uM4Z7OAyvsk+8sZXVWGoFGE/2j4mW5TiHOc7UaBR8bG8sHH3zAmjVrSE1NBSAhIYF+/frVePJkKP+r3el0cvjwYdLS0rjiiivo2LEjo0eP9qt34MABbrrpJj755BPGjh3LmjVrGDZsGL169aJnz56+etnZ2ZV2ERDnt4P5FVeXOZBXWu0E9OTPbmxI+ecnv8xNSOCJH0KvDluPFnHpsQTU5dH5928H+Tk1j5BAE5N6xHFJgoyWF9Xz2WefMX/+fEaMGMG6desYNmwYBw8eZN++fVx99dU1OlZ8fPxpB9x169aNNWvWnG3I1XJr+4t5a/evpBXn+5WHma3sd+RWuk+Yxcqolp0aIDohRFNQqwQUyn+s+/fvT//+/c8qgOLiYpKTk1m3bh0hISF069aN6dOn89Zbb1VIQA8ePEhYWBhXXXUVAP3796dz586kpKT4JaDiwtS1eTD7cvxHqneNObtBFnEhgWQ5XBXKjvtk8xG+21PeFaXM4WLBqlQSIoJ8CawQp/P+++/zxz/+kbFjx/Ltt99yxx130LJlS15++WU8Hk9jh1drrexhTEy4iL9v/4kyvfw6jGiEWqx0Cq2fie+FEOeWWvUBXbRoEStWrPA9f+qpp7jkkkuYNGmSr0W0unbt2oWu636d4auan65fv3507NiRxYsXo+s6q1atYv/+/QwZMsSvXrt27YiLi+P666/n4MGDNbs4cc66tnsMAxPCMRrAbjEytW9L2jSznXnH07ihZxyhgSf+Thvarhnto07Mc7sx3X9ksQI2pxee1TnFhSM9PZ3evXsDEBAQQElJ+R9Q11xzDUuWLGnM0M5alNWO2Wjk5HtiLYNCuF5WQhJCUMsE9IMPPiA0tPy25m+//cbSpUt54oknaNeuHX/9619rdCyHw+E71nFVzU9nMpmYMmUKt9xyCwEBAVx++eU8++yzxMeXj3SOjIxk7dq1pKamsnnzZmw2G1ddddVpR+bPnz+f6Ohov8fxHwFxbrGYjMwYlMjCST14fXx3RnQ80dKyK8vBqn055Jf697FTSvHj3hxe+zmVJduP4vLofttbhwfxt2uSeOSKdjx/VRduP2VUfVwlLZ3S+imqq1mzZr7pkWJjY9m2bRtQ3qf+5BHr56ItuRkUu0/cPdBR7HfkklF6+rlHhRAXhlrdgs/KyiIuLg6AH374gWHDhjF8+HDat2/PtGnTanQsu91OYaF/i1FV89MtXbqUWbNmsXTpUvr168f27dsZO3YsMTExjBkzBrvd7huFHxkZyT/+8Q+Cg4PZvXs3nTpV3rdo1qxZzJo1y69s+PDhNboG0bSYTlk+89WfU1m5r7zfmdmo8eClbekeVz5C+N11h/jfjixf3S1HinhoaDu//c1GA0mxIVRmYo84dmcXk1tSntgOSAinW2z9zK0ozj99+vRh5cqVdOrUiauuuornn3+e7777jpSUFC699NLGDu+s7CjMRANOTqMzSor4w9ovmdquLyNadMQqA42EuGDVKgG12+1kZmYSExPDzz//zJ133gmU9wut6TygHTp0QNM0tm7d6pvYvqr56TZv3szAgQO55JJLAOjatStXXnklS5YsYcyYMRXqa5qGpmlnXAlHnL/255T4kk8At1fx0cbDdI8Lwe3V/ZbuBNiYXsjRIqff5PSnExsSyItXd2VnpoPQQDOtwq11Gr84vz3yyCO+ls6JEycSGhrK5s2bGTBgANddd10jR3d2bCYz6qS1yRTg8LjYkpvBR6mbWJdziCd6jWq8AIUQjapWt+CHDh3Ko48+yj333ENeXh4DBgwAyvtztmjRokbHstlsjB8/nkcffZSioiJSUlJ48803K21Jvfjii/n5559Zu3YtADt37uS///0vF110EVA+ncn27dvRdZ38/HxmzpxJu3bt6NChQ20uU5wHckpcFcqyiyuWnaymf7AcbyGV5FPUVHZ2NmbziVbAkSNHMmvWLCZNmuQ3z/K5aHDzNvj3AC3/wfGo8kaKPUU57C3MrmRPIcSFoFYJ6IMPPsikSZNo06YNr7zyim+1jaysrFpNbPzKK69gNpuJjY1l+PDhzJ492zcC3m63s3LlSgCGDBnCvHnzuPHGGwkODmb48OHccMMNvnlA9+3bx5gxYwgODqZTp07k5uby1VdfYTQaa3OZ4jzQNSaYILP/+9+3VRhQnjhe3jbSb1v32GBipA+naCDjxo0jLy+vQnlBQQHjxo1rhIjqTmtbOJZTbrHrKMzaiZ8dQy2m7RNCnB9qfAve4/Hw17/+lZtuusnXD/S4yZMn1yqIsLAwkpOTK93mcDj8nt91113cddddlda94YYbuOGGG2oVgzg/Wc1G/nhFO/6z4TCZDhe9WoRyQ68Tn9spfVuSEGFle6aD+HArV7SPPM3RhKhbVbW2l5WVERBwbi9FuTR9J2Ve/6mkdMpvwwN0Co0iMbhZI0QmhGgKapyAmkwmvv76a2688cb6iEeIOtc20sZjwyvvhqFpGpe1i+SydpJ4iobz8ssvA+Wfv3/+859+C2fous6WLVvO6a5D/zu0gzVZBzi1fVMD0DRuatOLK+LaN0JkQoimolaDkIYMGcLKlSu5/vrr6zoeIYQ47x2fbkkpxc6dO/36gZrNZtq2bcvNN9/cWOGdtR8y9hEVaONQsf88uWbNSLjFypWtOjdSZEKIpqJWCWibNm1444032LRpE507d8Zq9R98UZt+oEIIcaF4/fXXAXj88cd58MEHff3ozxcWo5EAQ8WfF7PBwJDoNo0QkRCiqalVArp48WLsdjtbt25l69atfts0TZMEVAghqmHOnDmNHUK9GNeqK2/u+rXCLfgAg4nxCd0bJSYhRNNSqwT0iy++qOs4hBDigjB79uxq133mmWfqMZL60yuyJQm2cDJLivCeNBdonruUP2/4htCAQPpGtW7ECIUQja1WCagQQojaObXL0vnqqlZdWJt9AE4Z6P9r1gHmblzKV8NuQztlGqb9RTn899AOSjwuLo1py8WSpApx3qp1Apqamsp3331HRkYGbrf/+trn620lIYQ4WxfK92OX8OZQ4SY8eFH8mnmAMq8bqymAvYXZ7C3KIcISxCvbf8Kpl09UvyE3nfu6DKJfVHwDRy6EaAi1SkB/+OEHZs+eTffu3dm0aRM9evTg4MGDFBcX079//7qOUYhG49UVBo0KLTVC1KVt27Zx6NAhBg8ejNVqxeFwYLFY/EbHn2s25aZj1LQKLaAARcfmAv0kdTOfpG0B4FBxARrQwhbqq/f9kb2SgApxnqpVAvr6668zY8YMbrrpJoYMGcJjjz1GTEwMTzzxBAkJCXUcohANz+XR+devB/g5NRer2ci13WIZ1Sm63s+rlOKX1Dx2ZxfTplkQAxMiMBgk+T1fZWdn88ADD7Bz506UUnz66ae0bNmSBQsWYDAYeOihhxo7xFpLdeRSpnuq2KrIKCkief8mDIbylZGMmkaaI48YazDGY2WBxnM3ARdCnF6tluI8cOAAl19+OVA+Z11ZWRkmk4kpU6ZUuaKREOeSz7dmsHJfLl4dHE4v//7tELuzHGfe8Sy9s/YQC35K5ZudWbz6cxpvrE6r93OKxjN//nxiY2NZsWKF32T0V1xxBatXr27EyM7e1rzMyho/AdDQuP2nRazOOsD2/KN4lU5UoA2zwYhX6QCYNQNjZL5QIc5btWoBDQkJobS0FICoqCj27dtHu3btKC0tpaSkpE4DFKIxbM0oqlCWklFE+6j6m6+xxOXh2z1ZfmUr9+UyuVcLQgKlJeh8tG7dOv75z38SFBTkV96iRQuOHj3aSFHVjQJX1b8FbuVlb1EupV43qkxxxFxIS1sYExK6M6h5IiUeN5dExxMbFNKAEQshGlKtEtAePXrw66+/0q5dO6644gpeeOEF1q1bx+rVq7n44ovrOkYhGlyrMCu7sor9ylqH1e/oZV2BrvuXKcCjV9WOJM51Hk/lt6gzMzMrJKXnmubWYLYXZlW5XUdhMRgJMpkxahqXxbTl+sQehAQEVrmPEOL8Uatb8A899BBDhw4F4LbbbuOGG24gMzOTyy67jD/96U91GqAQjeHabjG0DD3xQzgoMYJeLUNPs8fZs1tM9I8P9yvr2SKEiKCAej2vaDz9+/fno48+8j3XNI3S0lL++c9/MmDAgBofb8GCBfTp0weLxVJhqeSEhASsVit2ux273U7Xrl3POv7TuTo+CWMlo+Ch/Ba8Bhg0A9HWYB5Muow7OvaX5FOIC0itWkDDwsJ8/zYYDEydOrWOwhGiaQgPCuDZsZ3Zn1uCLcBE82BLg5z3rgHxJERY2Z1dTNtmNkZ1imqQ84rG8fvf/557772X66+/HqfTyZw5czhw4AB2u53HH3+8xseLi4vjscceY/ny5WRnZ1fYvnjxYkaNGlUXoZ9R17AYtCp6gUYG2tCObb0ith0jW3RskJiEEE1HrecBzc7OZsmSJRw6dIi7776bsLAwNmzYQFRUFC1btqzLGIVoFJqm0aaZrUHPaTYauKprTIOeUzSe5s2b88EHH7B06VJ2795NaWkpY8aMYfTo0X6Dkqrr2muvBWDjxo2VJqANaXn6LrxVbCvzuLEHBhBvDyfaaq+inVQIcT6rVQKakpLCjBkzaNOmDdu3b+fmm28mLCyMdevWsXfvXp5++um6jlMIIc5LJpOJK6+8skHONWXKFHRdp2vXrjz11FMMHDjwtPXnz5/P/Pnz/crat29frXOlFedVOQq+0OMkWLeQ7yrjh4x9hJqtNLfaibbay1tOZd5dIc57tUpAX3rpJaZMmcK0adMYMmSIr7x///4sXry4zoITQojz2a+//sqKFSs4cuQIUH4LfejQofTp06fOz/Xee+/Ru3dvAN5++21Gjx7Nli1biI+veqL3WbNmMWvWLL+y4cOHV+t8waaqu63oSqGhkV1WTIDByAtbv6dDaPk8u/0iW3Nf18HVOocQ4txVq0FIu3btYuTIkRXKIyIiyMvLO+ughBDifPfkk08yY8YMNm3aRFBQEFarlQ0bNnDPPffw1FNP1fn5Bg0ahNVqxWq1cvfdd9OzZ0+WLFlS5+c5zmQwYjzN9lJP+RLOB4vzCTKdGGi3JvsAewsbt/uAEKL+1aoF1GazkZubS4sWLfzKd+3aRVSUDJoQQojTWb58OUuXLuWVV16pMHXd6tWreeihh+jfvz9XXHFFvcVgMBhQqv6m+IoKtBFoDKDY66p0e46zGKvJTIQliFir/3yf+a6yeotLCNE01KoFdMSIEfz9738nLy/P11cnJSWFl19+mdGjR9dpgEIIcb758ssvmTp1aqXzJvfv359bbrmFL7/8ssbH9Xg8lJWV4fF40HWdsrIy3G43Bw4cYOXKlbhcLlwuF//85z9Zu3YtI0aMqIvLqdSQmDZYTVW3cSgg3hbO/V0G+5beBLCbAugSGo1br2oIkxDifFCrBHTGjBm0bt2a0aNHU1JSwsSJE5k2bRqdO3fm9ttvr+sYhRDivLJz587TzvM5aNAgduzYUePjPvnkk1itVp566imSk5OxWq1Mnz4dh8PBvffeS0REBDExMbzzzjt89dVXtG3b9mwu47Quj2lHu+DI09bZ78illS2MsS07E2cNoWdEHFe36srD6/7LlJUf8uTG5eQ6ZXU9Ic5HtboFHxAQwGOPPcbtt9/Onj17KC0tpUOHDqftzC6EEKJcQUEBkZFVJ2fNmjWjsLCwxsedO3cuc+fOrXTbxo0ba3y8s2E0GPz6dp5KAUVuJz9k7OMvvUcxuW0vHG4nM1cvxnWs9XNbwVHe3LWGh7pd3kBRCyEaSo1aQHVd5+2332batGnccsstJCcn06dPH4YPH35WyWd+fj4TJ04kODiYuLg4XnrppSrrLlq0iC5duhAcHEz79u3517/+5bf9448/pm3btgQFBXHFFVeQlpZW67iEEKI+eL1ejMaqh+gYDIYql+k8l+wuqHopTgC37sVw0oxLewqzfcnncVvzMuojNCFEI6tRC+hbb73FwoULufLKK7FYLHz22Wfk5uYyZ86cswpi5syZOJ1ODh8+TFpaGldccQUdO3as0J/0wIED3HTTTXzyySeMHTuWNWvWMGzYMHr16kXPnj3Zvn07U6dO5dNPP2Xw4ME88sgjTJw4kTVr1pxVfEIIUZeUUvzlL38hIKDyFkKXq/KBO+eaYLMFSqvebjMG0D0izve8pS0MDfzmD21tD6+wX00ppWRuUSGamBoloF9//TV/+tOffEu5jRo1iunTp/OnP/0Jg6FW3UkpLi4mOTmZdevWERISQrdu3Zg+fTpvvfVWhQT04MGDhIWFcdVVVwHlnfU7d+5MSkoKPXv25L333mPUqFG+jvVPPPEEUVFRbN26td7XPRZCiOoaO3bsabdbrVbGjBnTQNHUn/ahzdhWmFnldi86n+zfjMVgYmzrLkQG2rg+sQeLUjfhVYpQcyBT29V+TtQSj4vXd65mXfYhwi1WJrfpySXRCbU+nhCi7tQoAc3IyKBXr16+50lJSRgMBrKysmjevHmtAti1axe6rpOUlOQr69GjB59++mmFuv369aNjx44sXryYq6++mp9//pn9+/f7JsNPSUmhb9++vvrBwcG0bduWlJSUKhPQs1npQwghauNs7xqdKzJKi0+7PddVyidpm/kuYw+POq7g3i6Duap1V4bEtCGrrJgEezgmw+lmEz299/auZ232QQBynCW8sv1n2odEERnYsEvsCiEqqlGzpdfrxWw2+5UZjcaz6qvkcDgIDQ31KwsLC6OoqKhCXZPJxJQpU7jlllsICAjg8ssv59lnn/X1P3U4HISFhVXrWMfNmjWLzMxMv0dQUFCtr0cIIUS5iIAzr2fvUjpZZcX8fftPvrLQACvtQiLPKvkESDml/6iOYlu+9CkVoimoUQtoZf2WnE4nzz33HFar1Vf2zDPPVPuYdru9wmjPgoICgoODK9RdunQps2bNYunSpfTr14/t27czduxYYmJiGDNmDHa7nYKCgmodSwghRP3qE9WaJem7zlhPR5FWlMt7e9cxrlVXQqqRuFZHS1so2U7/VtgWQaFV1BZCNKQatYCOHTuW0NBQ33JuVquVK6+8koiICL+ymujQoQOaprF161Zf2caNG/1uyR+3efNmBg4cyCWXXILBYKBr165ceeWVvuXkkpKS/KYacTgc7N27t9JjCSGEqF9rsw5Wu65HKeZuWMrtPy3Cq+t1cv7JbXoSEVD+m6QBo1p0pG3I6ecmhfLGlq15GazPPiQT4gtRT2rUAlof/ZZsNhvjx4/n0Ucf5d133yUtLY0333yThQsXVqh78cUXM2/ePNauXUvfvn3ZuXMn//3vf3nkkUcAuOmmm+jbty/Lly9n0KBBzJkzh+7du8sAJCGEaASHi/OrXdeoaRS5y1iTdYAF21cRbQ2mR0QcncOqN74gvaSATbnpxFhD6BERh6ZptLSFcX2bHiw5tIOWQWGMbdXljMdxeT08tflbdh9bjz7SYuPPPYbXSb/RjTmH+e7IHswGI6NbdqJdNZJhIc5XtZqIvq698sorTJ8+ndjYWIKDg5k9e7ZvBLzdbmfJkiUMHjyYIUOGMG/ePG688UaOHDlCeHg4N910E7fddhsAnTt3ZuHChdxxxx1kZGTQv39/Fi1a1JiXJoQQF6zIQBsUnLkegFfp5DlLMWpOFu3fRHxwBF8e3Ma09n0ZFtfhtPuuyUrj79t+Qj82gVPfyFb8vusQlh7eydt7fgNgvyOP7QWZzO87FovR/6dvX1EOuc4SksJi+Dkz1Zd8AmQ7i/ny4FZubV9x2dSa2JybzvyU731TTP2WfZB5va+khU26BIgLU5NIQMPCwkhOTq50m8Ph8Ht+1113cdddd1V5rAkTJjBhwoQ6jU8IIUTN9YuOZ8XRfdWqqwO60vEonSDTicGuH+zbwCVR8djMlkr3yywt4uG1X5NeWkiw2UKCPYK12QfZV5TDslP6n2Y7i9mYe5h+UScWTlmwbRU/Z5UvWGIzBdAtPKbCObLKTj+avzq+z9jrN7+pW+msOrqfSW16nPWxhTgX1W7yTiGEEOIMejdrWeN9FHCouBCn18OmnHR+zNjHPb98yuK0LZXWfz7lBw6VFODyeskpK2HXsdWXCl1lmLSKo+hPLtuef9SXfAIUe1ykOfI4dcr62lzHqSyGiu09gcYm0QYkRKOQBFQIIUS9aGELJSbQXu36GmDCQJ67lK15GeS7SokMtOFWOsmpm0lz5PrVzygp5FBJgV//zCK3EwCbOYArW3byjycohIsiYn3PM8v877AB6EpxV8dLaBEUQqTFxoSE7gyNbVfta6jKqJYdsZw0rVRYQCCXxrQ96+MKca6SP7+EEELUiz6RregU2pyMShK9ypg1I1GBNsIsVg4W56Or8rk804sL6RQWzb6iXOLtEb76IQGBmDUDCfYITJqBPGcJJV43DreTORuWkmiP4P4ug9mef5RmgTaGxrbzm1u0W3gsZs2AW50Ydd+rWUsGx7RhcEybunshgHh7BM/2GcNPmamYDUYGN08kNKBms8YIcT6RFlAhhBD1wmwwMuik/pZnrG800MoWRlZZMbpSuHUPSilyXSXsKsikWYD/IiFBpgDGJ3THoGm0todjNQWglCLNkUeOs5jtBUfZVZjFlPZ9GduqC0GmAL/9IyxBPJB0KQn2cELMFobFtmdS4kV1cu2VibYGc018N8a26iLJp7jgSQuoEOepzCInJqNGRFDAmSsLUU+Ups5c6ZgSj5u12QcJPjbgSFcKhY5ZMxIbFEquq6TCPle17kqvZi35LC3l2Oh1jYzSIvYW5RBmtpJVWswVse2JDQqp9JwXRcRxUURcra5NCFF70gIqxHmmzO1l3vLd3P/5Vu79NIVXf05F16ufBAhRl25o06vadRXgReFWXoyaAbPBiNlgIMYaTLw9nLgqVjFqYQul2OsiKtCGV+k4PE6UUjh1D5oGb+3+tY6uRghRVyQBFeI88/X2o6RkFAHlP+gr9+XyS1pe4wYlLlhO3UtgJaPRT6fU48ale3DpXnQFzQJtXBQew4f7N3LTD+/zt20rKXKX+e0TYrYQFmAlKtCOhoamQXiAlbigUPYW5dTlJQkh6oDcghfiPLMvp+Jtyv25JQxMjKikthD1K9Bowmg0gKf6S1rqlK9IZNQMmDSNEreTf+5ag0fXcSudXzLTOFScz3N9r/Ltc1Wrrqw4sheDpmE1mgkLCKRHsxYYNI12wbLikBBNjbSACnGe6RRdcdqbjlHVnwpHiLqkaeCtRRcQL8cnp1dklBaR6sgns9RBoauMo2UOklM3+63TXuAuBcpHxre2h6ErRa6zmFhrMNPa962jqxFC1BVJQIU4z4zqFM3AhHAMGpiNGmO7RNOnlSz3JxpHnrMUg3bq1O6nZ0BDozz5dCudIrcLdaxvKIBSijxnKbpePn1SicfFn9d/w285h8gsLULXFR5dp8TjJt9VRr6r7DRnq9r+ohxWZ6ZVuN0vhDh7koAKcZ4xGw3MGJTIPydexBsTujO5V0u0GiYA4ty0YMEC+vTpg8Vi4frrr/fblpKSQv/+/QkKCqJLly589913DRJTsNmCvYplNKuio1Dge3iPzdOpAyVeN27lxaQZeOi3r5i/ZQV/2biMfUW5eHWdAlcZu4uyUYDJYKDU6+bdvetqHPebu9bw6Pr/8fK2ldy2ahEbcw7X+BhCiKpJAirEecpqNmIx1Wzwhzi3xcXF8dhjjzF9+nS/crfbzVVXXcW4cePIy8tjzpw5XHPNNWRmZtZ7TJ3CmnNxZKta7atQuHUvnpMmigdw6+Uj3T8/sI2Fu9fy3t71FLpKyXYWk+cqxat0dBRRx1ZhOlJaWKPzpjly+e7IHvKdpfyWfZBfsw5y66qP2CBJqBB1RhJQIYQ4T1x77bX87ne/IzLSf9DN999/T0lJCbNnz8ZisTBp0iSSkpJITk6u95jMBiMvnDRYqK64dC/ZZQ5Si3I5UlJIanE+Js1AgMEICjqFRpf/G+gUEs3Swzv5Ln03ha5SvjywlWc2f8f7e9dXens9q6wYpRR7irJxH7vNX+h28uSm5azPPoR+SkIshKg5GQUvhBDnuZSUFLp164bBcKLNoUePHqSkpJx2v/nz5zN//ny/svbt29f4/B3CotEov51eVxTlt+OP/9ugoNjjIshoJjrQzp7CbDbkHCI60E5uWTEbcg/j1r2klxYSaw3BYjSxOe8IW/MzmNymJ7nOEi6KiCM0wEqXsOYYNQMub3mfU13plLndrMs+xF82LadLWHP+dNEwAk3mOrwiIS4skoAKIcR5zuFwEBYW5lcWFhZGWlraafebNWsWs2bN8isbPnx4rWIwaAZfX866cnJCa6B8cJJL93KkrBDTsblHdxdls6swC6PBgFkzoqM4YMmnX1RrLAYjSw7tICUvA7vZgsVg5OFuQ2lhC2FI80R+zTqAR+kYNSNepTAZDFiMJvY7cvk+Yy+jWnaq0+sR4kIiCagQQpzn7HY7BQUFfmUFBQUEBwc3WAzx9jD2FeXWy7GNaHgBg1Y+gElDw6V7fSPpFaB0HS/lCXCRq4yDxflEBtrIc5aSYC+fI9epe3lv73ryXSXkukrpGBbNroIscpzF5a2hGqSXFNDSFkZmmeOMcXl1nZVH9/H1we0ooE9kS65unYRVWk6FkD6gQghxvktKSmLLli2+aYsANm7cSFJSUoPFMK/X6Do9nnbsAeVJp1Er72+aYA/HpBkAhX5SG6ni2LyigMPjYn9RDg63k2aBQX4JYUr+EXJd5XOKBpstWI0mwgOshAdYMWtGDjjyKfG46NWsxWnjU0rx7JYV/P7XL0hO3czHaZt5a/daXt62si5fBiHOWZKACiHEecLj8VBWVobH40HXdcrKynC73Vx22WVYrVaee+45nE4nycnJbNmyhQkTJjRYbJPa9DzrYxyfHxTAoGmYNSNmzYBB01CqvLUz0mIjLMCKQdMwUPn0YwZNw6BpJNojSAqL8dvWxt7M73mJ102wOZCE4AgsRhOBJhNDY9uRFB572lh3FGSyPucQec7yZBYFh4rz2Zx3hDxnxdXKhLjQSAIqhBDniSeffBKr1cpTTz1FcnIyVquV6dOnYzab+eKLL1i8eDFhYWH8+c9/5tNPPyU6OrrBYjtSUoixlvPRGtAwoWE1mbEYTJg1AygwahpepUCBpmkYNQNb8jPoHh5LnDUUi8GIUTP4JaIa5fsZNAMHi/N5IGkIXUKbE2MN5upWXXmg6xBMaOQ5S8kpK8ZmCiDaaqeFLZTekS3p3awl1yf28Iuv0FXGJ6mbeXPXGlLyjgBQ4nFj0MrXpD+uvD+p5hudL8SFTPqACiHEeWLu3LnMnTu30m3dunVjzZo1DRvQSV7f+QtBRjNFHleN99VR6IBR9xJislCqeyj2uHAp77Ep6zU0pfDoXrxKZ13uIdy613cD3qBp5cmvAu+xgUr5zlLSHHl8c3gXj/UY5juX0+shJCCQbQWZeJVOgj2cDqFRHCwuwG4K4IY2PYm2BvvV//OGb3x9Qr87sod7Og2gb2QrIgNtxFpDSC8pn4c0KtDO8LgO2Go4Mb8Q5yNJQIUQQtS73YXZhFuCapWAHufUvWS5Tty+9g0wOpagakqhVPk8nuX9QI/XUQRoRtyqPCk1HmsR9eg6G3PTOVxcQAtb+XK1q47uJ9dVSofQKN95OoZE86eLhhNoNGE6pfVyfc4hjpYWAfhWHPvfoR0Map7Iny8azuIDKWzMSScsIJCJiT3oHdmy1tcvxPmkSdyCz8/PZ+LEiQQHBxMXF8dLL71Uab33338fu93ue9hsNjRN49NPPwXKJ1s2GAx+debNm9eAVyKEEKIyCfYIOoREnbliDZw6r+jJzz1Kx6vKF/XU0Agwmgg1BWIzBGA2lqegWWXF7Mw/yvb8o77J5bPLiskuK2ZTbjrrcw5xuLiAbGcxdrOlQvKplOKbwztZk3WANVkHSC3KhWNJMUBMUAh3dxrA6wPH82zfsZJ8CnGSJtECOnPmTJxOJ4cPHyYtLY0rrriCjh07Mnq0/6jJG2+8kRtvvNH3fMmSJVx//fWMGjXKVxYdHU1GRkaDxS6EEOLMbu9wMUVuJ6sy9lOmPA1yzuOJoEZ5slisl59X6QojGkpT7CvK5bWdv7AsfRd/6jGcyEAbuwqzfDunOfJ8ramn+jkzle0FmZgNRpxeD+klhQSZzNzV6ZL6vzghznGN3gJaXFxMcnIyTz31FCEhIXTr1o3p06fz1ltvnXHft956i0mTJhEUFNQAkQohhKitxOBmvNRvHO1CI89cuYbO/EOm8Oo6ZoMRpdSxPqUKm8lCcEAgTt3LwZIC/nd4B9llxcRZQyj1unB4nIQFBOLWvRWO6NV1tuYfxagZ6BYeQwtbKFFWO30iWzEkpm2dX6MQ55tGbwHdtWsXuq77zUfXo0cP3231quTk5PDFF1/w/fffVyiPiYnBYrEwatQonn76aSIiIqo8Tl0tNSeEEOL0DJqBvpEtScmv27tUZ1riU0Mj1GKlwFnq6xPqQeFFx2I0YjcFALC7IJtij5O04jysxvKyfFcZhSetF5+Sd4S3dq8lo7QIEwacXg8Wo4l4ezgAA6MTKPO42ZibTpDJTLfwWF/fUCHECY3eAupwOAgNDfUrCwsLo6io6LT7vf/++7Rt25ZLLjlxq6NTp05s3LiR9PR0Vq1axaFDh5gyZcppjzNr1iwyMzP9HtKiKoQQ9ePOjgOwGuq27ePUJTlPdXx1JKdevpqRQitfS97tokNIFArIKSvmh4y9fHtkDwWuMvJcpYDCbDRiPtb3s8zj5oWUH1iffYi9hdkcKS2gyO30nadLaHOSwmK4/9fP+dv2VTyzZQWPb1xaaQuqEBe6Rm8BtdvtFBYW+pVVZ4m4hQsXcuutt/qVxcTEEBNTPqlwq1atWLBgAe3ataOkpESSSiGEaAL6Rbfm2tZJvJ+6sV6ObzaYcB3r63l8gnqTwUBbewSZZUWgypfs1DCgK0VKbgaluhunx4PNFECxx4VH1/EoL8psoWtoc9zHpm1KLy1gXc4hCl3lSefRUget7GE82/tKQKOVPYxXd/xM4bGk1KN7WXFkL2UeDze07clFEXE1uha37mVddvmUUr0jWxJ0rKVWiPNBoyegHTp0QNM0tm7dSteuXYEzLxG3YcMGUlJSuPnmm097bIPBgFIKpc50g0YIIURDubfrYD5O3YwT/cyVa0AD2tojyCpzkOcuK59ySYMEWzjFXtex/p+gKTjebppeWojZYMCl6xQfmyJKAZoGWc4S1mYfRFeKe9cs5tLmiRS6yij1enx9SgtcpbQ6dvsdykfRQ/mgp615Ryn2uFAoDpTkM619X4bFdajWtZR4XPx5/Tekl5Y30ITtD+TxniOJCrTXzYslRCNr9FvwNpuN8ePH8+ijj1JUVERKSgpvvvkm06ZNq3KfhQsXMnr0aF9r53ErVqwgNTUVpRQZGRn83//9HyNGjMBms9X3ZQghhKim3pEtuSaxe50fVwEOrwsdCDAYMWkGjJrGoZICtuVn+hbyPN4kYdQ0FAqn14uudLwovMcGKBnQ0JWOS/eyOe8I67IP8WnaVgCKPU5KvW4cHmeFVsm+ka0AKHCVlie0GoRbyu/A/ffQjmpfy4oje3zJJ5T3Rf364PbavTBCNEGNnoACvPLKK5jNZmJjYxk+fDizZ8/2TcFkt9tZuXKlr67L5eKDDz6oNEHdsGEDgwcPxmaz0bt3byIjI3n33Xcb7DqEEEKcmclg5ImeIwkzB9b5sQ8U55PjKsGte/HoXpy6lxKvG6fuwXtSb9GTGkFRlQxj0pXyHQPKl9bcXpBJgMFIREAQYQFWIixBBBnNfvuNbNGRCQndibAEYTMH0DEkCtuxJNWrV7/FN99VVqGsvF+qEOeHRr8FD+WDjpKTkyvd5nA4/J4HBASQnZ1dad0HHniABx54oM7jE0IIUbfah0ZxY9tevLLj53o5vludPtkzoKFp5W2ieiUpqPdYK6hT9+LWvZR63bh0L2FmCzZzIMFmC2EBgbSyhQHlg5g+3L+RNEcenUKj+Uf/a3l80zKOlJ4YUDu8RfVuvwP0i2rNfw9t94urf1Trau8vRFPXJBJQIYQQF56/XjwOs2bgpe2rGvzcGhomg5FQs4UCd/ktdf/tEBoQiFfpZDmLQSkMmkaes5Q8VxktbaG0CY5gWFwHlFI8t2UFB0sKADhUUkCOs5g5PUaw5NAOMssc9G7WkgHNE6odX7uQSO7rMpivD23HrXu5IrY9l0RXf//64vR6+M++DWzIOUxsUAjXJ/YgIbjqqQ6FqIokoEKIJi2jsIy31x5kX24J7SNtTO3biii7pbHDEnUgwGjixf6/Y9HeDaS7iuvtPMdn4Ty5NVGho+sakYF2vEpR5nX7bdeALmHN2Z5/FJOmgWZAVzoKsBrNlHk9tA1uxuQ2PTlcUuBLPo/bmJtOgMHIpDY9ah33xVGtubiJtXq+vXstPxzdB0CWs5j9Rbn8rf/vsBglnRA10yT6gAohRGWUUrzwwz42HynC4fSy4XAhL6/c39hhiTr289X31evxT729bkDDoBlwKS/b84+S6yqpkHyaNCNGzUC4JYh4Wzg2kxmjZkDTNMwGI+1CIjEZjGiaht1kwYBGvrOEfUU5ONxOrEazb/7QzbnpfJK6mc256fV6nQ3h1+yDfs+LPE625x9tpGjEuUz+ZBFCNFmZDheHC/wHY+zLKSG/1E2Y1VzFXuJcE2+P4NFuQ3lqy3f1do7yKZm0k56X9/v0ovwyVAMaRk3DZg4g2Gwh0V5+e9mLTonHDWhEBdqIDrTT5tit5zCLlUJ3Gd8d2YOuFEbNwPSO/TAaDHywdz1fHToxen1My87c2LZXvV1nfYuwWDlc4t9doZlFZpoRNSctoEKIJis00ITF5P81FWQ2YgswNlJEor480mMYF0e2rN+TqPK+n0qBt4r5oQ2aRoDBSIw1mGCzhRCzhUMlBcRaQ+keEUuvZi3oFhFLh9Aorosvn0oqvbiAHzL2oo7tr2nwceoWCpwl/O/wTr/j/+/wDopPWj3JrXtJyTvC4WL/W/hN1Q2JPf0S+SHNE2llD2u8gMQ5S1pAhRBNVqDZyPU94vj3b4eO/bjDjb1bYDae/d/OTo+X5buyOZBfStfmwQxuEyFrdjeiIFMAP4yewdhl/+LbjD31cg6FwmYKKB9wpB+blN63rbxFJsRsITE4gnhbOIXuMnYVZFHgKiO7zMEt7fpwW/uLaRZoo7m1fLU+pRSz133tm8QeNAI0jUJ3GYdLCvEonRKPC4fbid1sIcgUgFP3YgMOFxcwb/O3vumVhjRP5K5OA+rl2utKr8iWvNzvd2zJO0KsNYQOoVGNHZI4R0kCKoRo0kZ2iqZXy1D25ZTQtlkQkXU0AOmF7/eRklE+Rc7KfbkcyC/lpt713AInTivQZGb56LuYu/5/PL5peZ0fXwcCDSbCAqy0DAplbfZBPMcGFkF5EprvKqPY7cKpH+sf6izBresYNQNfHNhKgauMv/e/xnfMTbnpFLlc5RPXowCFR9eJsATSLiQSi8HIzzkn+n5eGtOGiGMT03+4f4Pf3J4/Ht3PkJg2dAnzX2SlqYmwBHFpTNvGDkOc4+QWvBCiyYuyW+gXH15nyefhglJf8nnc8l1ZeHVZtrcpmNtrFAnH5tesSxbNiNFgwOFxsSkvHZOhfKCRUSvv96mhYdQMpJcWsjorjTRHvm+KJqPBgFvXyXGWsOmkwUT5rlKsZjMJwRGYNSMaGgFGEw92vRRF+XrwcbYQggMsxAWFoCudMk95H8ojJUUVYkwvKaxQJsT5SFpAhRBCNDmP9xzFXzYtY09RTp0dU1G+Vrvn2CT12rHFOY//2WHQNEyahsvrPTbg6MR+xR4XkYFBBBpNBJtP/CHUI6IFGhAVaMdA+Qj5iYkX8Ydul5HnKsULJNhPzJPpBRweF4EmM90jYkk/fCLhNKDRLTy2zq5XiKZMWkCFEBecFqFWkmKC/cqGd4jCaDh/+4BOnTqVgIAA7Ha773HgwIHGDqtKN7XrxXN9x3JDQo86O6ZbeXGr4zfKTyzBqZ30rHy0vAGFIsBgxHgsSfXoXopcTg468sguO7FCn9VUPt2Sw+1E0zTCAgK5NKYtmqYRYQmibXAzvxgS7RFEBpaPGp+U2IMhzRPJc5awpzAbk2ZgV0FWnV2vEE2ZJKBCiAvSg5e14abeLRjSJoK7B8QzuVeLxg6p3j3wwAM4HA7fo3XrpjXJ+ckMmoFr4rvxweU3kXzpzQRw9jMfVNrBQoOwgCBsJgsmTcNoMNDcai9POpWO0k7sp6GR6yrltZ2r2ZafAcD6nEO4dC9dw2Po2awFrezhLEvf5Tv877sOoV9kayItNvpFtuaBrkN82yxGE/2j4gm3BNEuJBKX8vLqzl/Oi/lChTgTuQUvhLggWUxGruzcvLHDENUwvs1FXNWqC//cvYY565eQe9I0RmdDAZoqX3Kza3gMuwqyOFJaRICxvN3z5PXkzZoBTSu/FZ/vKuW37EN0CYtBr2Q6p5PLIixB3Nd1cJUxnDqxO8DqrAN0j4g7u4sToomTFlAhhLhAvPHGG0RERHDRRRfx1ltvnbH+/PnziY6O9nuUlJQ0QKQVWcxmZnYZxMHr53Bf50GYqJvuEh4U2c5iDhbnk1nmINBoJM9ZigJMGDBQPlWTW+kUuMoodDtxej38fDSVu37+mOXpu7EYTrTOKqXw6F5uXfkh9/zyCd+m7z7t+cMDrNUqq4pX10kvKcCte6u9jxBNgaZUFbPxXsCGDx/OsmXLGjsMIYSoM+vXr6d169aEhYWxcuVKJkyYwOuvv851111Xo+M0le/HImcpj6xfwk+ZqWyog1vWNpOZsAArLq8Xt+6lyOPErBlRqPJb8UphMZhQQHOrnc5hzQk/Np2S3WTh0uaJFLjLKPG6WZ9z2O/YT/QcSbuQyErPm+8sZc6Gb8hyFgMQZbHxRK+RhFYjCd2Rn8nft68iz1WK3RTA9A796NvE1o4XoipyC14IIS4AvXqdWP7x8ssvZ8aMGSQnJ9c4AW0qgi1WXur3O9bnHGJNVhrPb/metJLaryZU7HFT4nEfG3JUTjOUT8ukvArNUH7D0GIwUuh2sjUvg67hMYRbgnB4nIRbrOwszGLFkb14lU4rWxjGY/tsyTtSZQIaZrHyXN+xrM85BECvZi2xGM/806yU4rWdv/jmEXV4XLy+czXdI+Kqtb8QjU0+pUIIcQEyGAyc6zfAjAYDfaNa0zeqNTO7DKbQVcaywzuZ9MN7VS61eTrHx8JrlA84ahvSjP2FObiUjkGVT2RfpnsIMprJ1b2kOfIItwSxpzCbB379kjKvhxKPiwCjiTKvh05h0QDEBYWc9rwWo4lLohNqFGuxx0XmSaPxAUq8bjJKC4k/adonIZoq6QMqhBAXgEWLFlFUVISu66xatYoFCxZwzTXXnHnHc0hIQCDXJV7EWwMnkmgPx2IwYkIjQKvZT53FYMJiMGLRTBg1AwGa0W8EvVcplFLku8o4WlpEjrOYHGcJHl3HYjTh8nrIdZbg1XX6RraiT7NWHCkp5L8Ht/Nb9kH0kwY31ZbdbCHO6p/Y2k0BxAWFnvWxhWgI0gIqhBAXgAULFnDHHXfg9Xpp3bo1Tz75JNdff31jh1UvbmnflwmJPchzlrAu5xDJqZs4VFzAoeIC9hZlU1n6d/zGu6ZpGI49wiyBGAwGNB00pfnmDdVRBJssdAtvjkvXcXu9lHhdlHrdhJmthFusdAuP5Zk+V5IY3Iz12Yd4cduPvlbZ7uGxzO4+9Kyv857OA1iw/ScySouICLByR8f+mA1nP12VEA1BElAhhLgA/Pjjj40dQoOymsxYTaHE2UK5qnVXX3l2mYPZv33Np6kpeJQXj65jMhgwaQYKPU4CjSYsRjMXR7aiXXAkh4sL2evIQdPKp2wyoBFkNBMbFEyPiJbsKcoi1WDAajRT6nFT5nXT2h7G1PZ9STw2CX1y6ma/LgGb846wIz/Td4v+cHEBKzL2AHB5TDta2KrXitkmuBkv9L2KAlcZIQEWDDVs6RWiMUkCKoQQ4oIRGWjnzUGTeHPQJLblHaWZJYh9RTmkOfIIMBj4NecQ7YKbMSHxIjJKi1AoSg+6ySwrwmYKINwSRIDByLT2FxNuCeJgST4dQ6NJc+TicLtoHxLJ3/r/jqSTltQscpdViON42UFHPn/e8D+cuhddKf69Zx1XxLZjQHQCg5onommVTzfl1r14dZ09RdnsKMgkwR5B72Ytq6wvRFMjCagQQogLUpfw8oUImgcFc0nzBACuTbzItz00wMrf+v+OP180nM8OpLAqMxWjpjGqRScmtenBAUce36bvJsISRIQlCJNm4PGeI3wtn8cNbJ7Ilwe3+Z6X374vT1C/O7Ib57E5PHcWZJLnLKXIXcamvCOkFedxU9veFeJenLaFLw5sZU9hDiUeF+1CIjEaDFwe05bpHfvX6WskRH2RBFQIIYSogkEzEB0UzB2dLuHWDhdjQPNNr9TaHs7jPUey/MhulFJcEde+QvIJMDHhIgKNJn7LPkR0oJ1r47sRaDIDJ5b5LPO4yXOW+pUtT9/N9Yk9MJ3Ur3NrXgbJqZvRleJwSQG6UlhLCmhtD+f7jL1cl9CdiGPzk9YHt+4lu6yY6EC773UQojaaxKcnPz+fiRMnEhwcTFxcHC+99FKl9d5//33sdrvvYbPZ0DSNTz/91FdnwYIFtGjRArvdznXXXUdeXl4DXYUQQojzmdlgrJB0JQRHcHuHfkzv2J82lSSfUD5d1DXx3Xiq92ju6zqYVvYw37bLY9ti1gwn1pvXNJpbg4HyuT5PnUxqR0Gmb5t+bGvhsdv5CnB6PWd1jaezPvsQM375lAfXfsm9axazPf9ovZ1LnP+aRAI6c+ZMnE4nhw8f5ptvvmHevHksWbKkQr0bb7wRh8Phe3z88ceEhIQwatQoAJYtW8acOXP48ssvOXLkCEajkbvuuquhL0cIIYSolnh7BE/0GslVrbrQq1kLuofHYjMFAHB5bLsKo9oTjs3xaTQYaHaspfN4/fYhkcSeYc7R2nJ5Pby28xccHhcA+a4yXtvxyzk/l6xoPI1+C764uJjk5GTWrVtHSEgI3bp1Y/r06bz11luMHj36tPu+9dZbTJo0iaCg8v+Eb7/9NrfeeqtvxY+nnnqKLl26UFBQQGiozI0mhAC3V2d3VjFhVjNxoYGNHY4QxNsjmNbhYm5s24tl6btIc+TROTSay2LbVqjbq1kLhsa2Y8WRPbQLiSTRrtMpNJpOYdH8rnVSvcWYWebwJZ/HZTmLKXI7CQmQ/0fnG6UUe4tyAKpcxetsNXoCumvXLnRdJynpxH+cHj16+N1Wr0xOTg5ffPEF33//va8sJSXFL2lt3749AQEB7Nixg379+lV6nPnz5zN//ny/svbt29fiSoQQTd3hglLmLd9DXqkbgMvaNuOOS+IbOSohylmMJsa26nLaOpqmcXuHflwX340yr6feWjxP1dwaTIjZQqHbeaIs0E6w2dIg5xcNp9Tj5unN37LneAIa3IxHul/h67dcVxr9FrzD4ajQOhkWFkZRUdFp93v//fdp27Ytl1xyid+xwsLCanSsWbNmkZmZ6fc43qIqhDi/fLgh3Zd8Any/N4cdR0//XSNEUxRuCWqw5BPK+7/O7DyQiAArANGBdmZ0HijTPp2Hlh7e6Us+AfYU5bAsfVedn6fRW0DtdjuFhYV+ZQUFBQQHB592v4ULF3LrrbdWOFZBQUGNjyWEuDCkF1acjzG90Emn5vIdIcSZJIXH8rf+vyPfVUZ4gFWSz/NUemlhxbKSimVnq9FbQDt06ICmaWzdutVXtnHjRr9b8qfasGEDKSkp3HzzzX7lSUlJbNy40fd8z549OJ1OOnXqVOdxCyHOPd1j/VuMDBokxUjyKUR1GTQDEZYgST7PY91PWkTBVxZRsexsNXoCarPZGD9+PI8++ihFRUWkpKTw5ptvMm3atCr3WbhwIaNHjyYmJsavfOrUqSxcuJANGzbgcDh47LHHuPbaa2UAkhACgEk94hjcJgKzUSPKHsC9gxKJDpY+bEIIcdzA5olcF98NmykAuymA8fHduSQ6oc7P0+i34AFeeeUVpk+fTmxsLMHBwcyePds3mMhut7NkyRIGDx4MgMvl4oMPPuDNN9+scJzhw4czd+5cxowZQ2FhISNGjOBf//pXg16LEKLpCjQbuXtAAncPSGjsUIQQosm6LqE71yV0r9dzaEom8apg+PDhLFu2rLHDEEKIJke+H4UQdaHRb8ELIYQQQogLiySgQgghhBCiQUkCKoQQQgghGpQkoEIIIYQQokFJAiqEEEIIIRqUjIKvRL9+/QgJqd4SZyUlJefc0p0Sc8OQmOtfY8QbGBjIl19+2aDnbErO9+/Hs3GhXS/INV8IanK9Nfl+lAT0LEVHR5OZmdnYYdSIxNwwJOb6d67Fe6G50N6fC+16Qa75QlBf1yu34IUQQgghRIOSBFQIIYQQQjQoSUCFEEIIIUSDkgT0LM2aNauxQ6gxiblhSMz171yL90Jzob0/F9r1glzzhaC+rlcGIQkhhBBCiAYlLaBCCCGEEKJBSQIqhBBCCCEalCSgQgghhBCiQUkCKoQQQgghGpQkoEIIIYQQokFJAiqEEEIIIRqUJKCnyM/PZ+LEiQQHBxMXF8dLL71UZd0ffviBpKQkgoKC6Nu3L5s2bfLbvmDBAlq0aIHdbue6664jLy+vScf8/fffYzAYsNvtvse8efMaNWaXy8X48eNJSEhA0zT+97//VajT1F7nM8XcFF/n1atXM3LkSJo1a0azZs0YM2YMu3fv9qvT1F7nM8XckK+zOKEm30fnoqlTpxIQEOD3uTpw4IBv+8GDBxkxYgQ2m43ExEQ+/PDDRoy29hYsWECfPn2wWCxcf/31fttSUlLo378/QUFBdOnShe+++85v+8cff0zbtm0JCgriiiuuIC0trSFDr5XTXW9CQgJWq9X3fnft2tVv+5lygabI6XRy++23k5iYSHBwMF27duWDDz7wbW+Q91gJPzfeeKMaN26cKigoUJs3b1ZRUVHqv//9b4V62dnZKjQ0VL3zzjuqrKxMvfDCC6p169aqrKxMKaXU0qVLVUREhFq3bp0qLCxUEyZMUBMnTmzSMa9YsUI1b968XmKsbcxOp1O9+OKL6scff1QtW7ZUS5Ys8dveFF/nM8XcFF/n//73v+rDDz9U+fn5yul0qoceekh16tTJt70pvs5nirkhX2dxQnXfv3PVlClT1MMPP1zl9oEDB6q7775blZSUqBUrVii73a62bNnSgBHWjU8++UQtXrxYzZgxQ02aNMlX7nK5VEJCgnrqqadUWVmZ+vDDD1VISIg6evSoUkqpbdu2KZvNpr755htVUlKi7r//fnXxxRc31mVUW1XXq5RS8fHxFb7HjzvT72pT5XA41J/+9Ce1d+9epeu6WrlypQoJCVE///xzg73HkoCexOFwqICAAL8vi0ceeUSNHz++Qt033nhD9e7d2/dc13XVsmVL9dVXXymllJo8ebJ68MEHfdt37dqlTCaTys/Pb7IxN9QPdk1iPlllXwJN8XU+U8xN/XVWSqmjR48qQGVnZyulmv7rXFnMkoA2vLN5/84Vp0tAj/+/yM3N9ZVNnjxZ/eEPf2io8OrcnDlz/BKypUuXqujoaOX1en1lAwYMUAsWLFBKlb/f1113nW9bYWGhslgsKiUlpeGCPgunXq9Sp09Az/S7ei4ZPXq0ev755xvsPZZb8CfZtWsXuq6TlJTkK+vRowcpKSkV6qakpNCjRw/fc03T6N69u6/uqdvbt29PQEAAO3bsaLIxA+Tk5BATE0N8fDx33nknubm5dRpvTWM+k6b4OldHU3+df/jhB2JiYmjWrBlwbrzOp8YMDfM6ixPq+v9JU/XGG28QERHBRRddxFtvveUrT0lJIT4+nvDwcF/Z+Xb9KSkpdOvWDYPhRPpw8jWe+l0RHBxM27Ztz/nXYMqUKURFRXHZZZfx008/+cqr87t6LiguLua3334jKSmpwd5jSUBP4nA4CA0N9SsLCwujqKio0rphYWFV1j3T9qYYc6dOndi4cSPp6emsWrWKQ4cOMWXKlDqNt6YxV+dYTe11PpOm/jrv27ePmTNn+vXda+qvc2UxN9TrLE6oy/8nTdX//d//sWvXLjIzM3nppZd46KGH+OSTT4CG+3/SmJrKb19Deu+990hNTeXAgQNMmjSJ0aNH+/o8ng/Xq+s6U6dOpW/fvowYMaLB3mNJQE9it9spLCz0KysoKCA4OLjSugUFBVXWPdP2phhzTEwMXbt2xWAw0KpVKxYsWMB///tfSkpKGi3m6hyrqb3OZ9KUX+eDBw8ybNgwHn74YSZNmuR3rKb6OlcVc0O9zuKEuvx/0lT16tWLyMhITCYTl19+OTNmzCA5ORlouP8njamp/PY1pEGDBmG1WrFardx999307NmTJUuWAOf+9SqluOuuu0hPT+ejjz5C07QGe48lAT1Jhw4d0DSNrVu3+so2btzodzvpuKSkJDZu3Oh7rpRi8+bNvrqnbt+zZw9Op5NOnTo12ZhPZTAYUOX9hBst5jNpiq9zTTWV1/nQoUMMHTqUO+64gwceeMBvW1N9nU8X86nq63UWJ9Tn/5Om6vjnCsr/n6SlpZGfn+/bfr5df1JSElu2bEHXdV/Zydd46neFw+Fg796959VrcOp7XpPf1aZEKcWMGTPYuHEjS5YswW63Aw34Hte6t+p5avLkyerqq69WhYWFasuWLap58+anHVH+7rvv+kY9t2rVym8UfLNmzdT69etVUVGRmjRpUr2NGq6rmL/77ju1f/9+peu6OnLkiBo3bpwaOXJko8aslFJlZWWqtLRUtW7dWn3xxReqtLTU1zm6Kb7OZ4q5Kb7Ohw8fVu3atVNz586t9DhN8XU+U8wN+TqLE2ry/+Rc9NFHH6nCwkLl9XrVypUrVWRkpPrPf/7j2z5gwAA1Y8YMVVJSon744QcVHBx8To6Cd7vdqrS0VD366KNqwoQJqrS0VLlcLt8I6aefflqVlZWpRYsWVTpCetmyZaq0tFQ98MAD58Qo+KquNy0tTf3444/K6XQqp9Op3njjDWWz2dSePXuUUmf+XW3K7rnnHtWzZ0+/QXNKqQZ7jyUBPUVeXp4aP368stlsKiYmRr344ou+bTabTf3444++5ytWrFBdu3ZVgYGBqk+fPmrDhg1+x/r73/+uYmNjlc1mU9dcc02FN7mpxfzCCy+oli1bKqvVquLi4tS0adNUZmZmo8ccHx+vAL/HihUrfNub4ut8upib4us8d+5cBSibzeb3SEtL89Vvaq/zmWJuyNdZnHC69+98MHjwYBUaGqrsdrvq0qWLeu211/y2HzhwQA0bNkxZrVYVHx+vPvjgg0aK9OzMmTOnwnfYlClTlFJKbd68WV188cUqMDBQderUSS1fvtxv30WLFqnExERltVrV5ZdfrlJTUxvhCmqmquvdunWruuiii5TNZlPh4eFq4MCBfr8/Sp05F2iKUlNTFaAsFovf9+dTTz2llGqY91hTSu5HCSGEEEKIhiN9QIUQQgghRIOSBFQIIYQQQjQoSUCFEEIIIUSDkgRUCCGEEEI0KElAhRBCCCFEg5IEVAghhBBCNChJQIUQQgghRIOSBFQIIYQQQjQoSUDFBe3RRx/lvffea+wwztptt93Gd99919hhCCFOcscdd/DSSy/5nl911VV89NFHjRfQeei3336jT58+lJSU1PmxX3/9dW6++eY6P64oJwmoaHBz586lT58+PPvssxW2/elPf6JPnz5+X9oHDx5k9uzZjBw5kgEDBjB27FgefvhhcnNzAUhPT6dPnz6VPlJTU6uMY+fOnfz666+MHz++ri+xwd12220sWLAAXdcbOxQhzguZmZnMmzePsWPHcskll3DVVVcxe/Zstm3b1tihndOOf1/v2bOnsUPx06dPH1auXOlXdvPNN/P3v/+9kSI6/5kaOwBxYWrevDlLly7l97//PQEBAQA4HA5WrFhB8+bNffXcbjczZ86kTZs2vPDCC0RERJCens6PP/5IaWmp3zFff/114uPj/crCw8OrjGHRokUMHz6cwMDAOryyxtG/f39KS0tZvXo1AwYMaOxwhDinHTp0iGnTptGiRQtmz55NQkICxcXF/PDDD7z00ku88cYbjR1inVBK4fV6MZkaJhXweDxN+ninCgoKIigoqF7PcSGTFlDRKLp27Up4eDg//vijr2zp0qV06tSJFi1a+Mr27dvH4cOH+eMf/0hSUhJxcXH06dOHBx54wK8eQGhoKJGRkX4Po9FY6fm9Xi/ffvstgwcP9ivv06cPixcv5t5772XgwIFMmjSJHTt2sHv3bqZOncqgQYOYMWOGr/X1uMWLF3PttdcyYMAAJkyYwBdffOG3/aWXXuKaa65h4MCBXH311fzrX//ya62cO3cuDz30EG+99RbDhw9n+PDh/OMf//BtV0rx+uuvM2bMGC655BKuvPJKXnnlFd92g8HAwIEDWbZs2ZleeiHEGTzzzDNERkby5ptvMmjQIFq2bEnHjh254447eOGFF3z19uzZw7333sugQYMYOXIkf/nLX3A4HNU6x5n+T5/q+O3g5ORkRo8ezaBBg/jTn/5EWVmZr46u67z11ltcddVVDBw4kBtvvJFVq1b5th+/Xf3zzz8zefJk+vfvz86dOyuc63gr5dKlS5n6/+3dfVzNd/8H8NfpRp066T4W3ZDbOqVrCnPbZW4yKkzuLjeNGHPPlbncrBhmurYxxkWGTTNkITZmMblnNtHB3IYZRqSc0u35/v7w6ztHRenbOZ16PR+P83g4n+/N5/1Jvc/7fG8+37AwtG3bFoMHD4ZKpdJa7/Tp0xgxYgTatWuHXr16YenSpcjLyxOXBwUFYe3atZgzZw46duyITz75BMHBwQCAgQMHws/PD6NHjwZQ/HIF4OkRyFWrVonv/fz8EB8fj8mTJ6Ndu3bYuHGjViwDBgxA27ZtER4ejps3b4rLbt68iSlTpqBbt27o2LEj3nnnHSQnJ2vFCQBTpkyBn5+f+P75U/AajQarVq1Cjx498MYbb2Do0KH49ddfi/18T548icGDB6N9+/Z47733cO/evWI/Y2IBSnoUFBSkVajt3LlT/MMvYmtrCyMjI+zbtw+FhYWS9X358mWo1Wo0b9682LI1a9YgJCQE33zzDRwdHfHBBx/gk08+wcSJE7F27VrcvXtXqzjcvXs3YmJiMHHiRMTFxYkfUgcOHBDXUSgUmDt3LuLi4jBp0iRs3LgR27dv1+r3xIkTSE9PR0xMDKZNm4Z169bh2LFjAIB9+/Zh48aNmDlzJrZt24ZFixYVO9rbvHlznD59WrKfEVFN9OjRI5w4cQLDhg0r8QuslZUVAODx48cYO3YsvLy8EBsbiyVLluDmzZuIiooqUz9l+Zt+3o0bN3Dw4EF8/vnn+Oyzz3D69GksXbpUXL5u3Trs2bMHs2bNwubNm/H2228jIiICv//+u9Z+li9fjilTpmDr1q1wdXUttb9ly5YhLCwMsbGx8PDwwJQpU8QzT7du3cKkSZPQrVs3bNq0CR9++CGOHDlS7JT1119/jebNm2Pjxo3417/+ha+++grA0+Juz549iI6OLtPPq8iqVavQpUsXbNmyBd26ddOK9d///je++uor1KpVCxEREeKX/OzsbHTo0AErVqzAhg0b4OPjg0mTJuHRo0dijAAwb9487NmzR3z/vI0bN+Lbb7/FtGnT8O2336JFixaYNGlSsQJz9erVeP/997F27VqkpaUVK6zpKRagpDc9e/bEqVOncP/+fVy/fh1XrlxB165dtdZxcnLC1KlTsXz5cnTu3Bnjxo3D+vXrcf/+/WL7Gz58ODp06CC+unfvXmrfd+/eRa1atWBjY1NsWUhICLp06QJ3d3cMHToU165dw8CBA/H666+jSZMm6N27N06dOiWuv2rVKkydOhUBAQGoV68eunbtin79+iE+Pl5cJzw8HD4+PnB2dkbnzp3Rv39/JCYmavVra2uLqVOnwt3dHYGBgfD29hb7uXv3Luzt7dG6dWvUrVsXPj4+6NWrl9b2Dg4OuHv3buk/cCJ6qVu3bkEQBLi7u79wvc2bN8PT0xNjxoyBu7s7mjdvjlmzZuHAgQPFzpCUpCx/08/Ly8tDVFQUGjduDH9/f0yePBnbt29HdnY28vLysG7dOkRGRqJNmzaoX78++vbti4CAgGJfdt977z34+/vDxcVFLKhLMnDgQAQEBKBhw4aYPXs2BEHA7t27ATwtdnv27ImBAwfCxcUF//jHPzB58mRs27YNgiCI+2jdujUGDx6M+vXro379+uJlUUVnrKytrV/6s3rWW2+9hV69eqFevXqoW7eu2P7uu+/C398fjRs3xrx583Djxg2cPHkSANCsWTP07dsXjRo1gpubG6ZMmYLatWvj6NGjAP6+VMvKygoODg6lXroVGxuLd955R/x8mDZtGpydnREXF1fs59uiRQs0adIEAwcO1DpKSn/jNaCkNw4ODmjdujW+//57ZGZmonPnziVebzNw4ED06tULJ0+eREpKCnbs2IH169dj9erVaNKkibje4sWLtb7Ny2SyUvvOycmBqalpicsaN24s/tve3h4A0LBhQ7HNzs4O6enpAIAnT57g1q1biIyMxNy5c8V1CgoK8Nprr4nv9+7di02bNuHWrVt48uRJseVFfTwbs729vfhB1qVLF2zcuBEhISFo27Yt2rdvjw4dOsDI6O/vkGZmZigoKEBBQYHOrukiqqkuX76MEydOFLuMB3haxNrZ2b1w+7L8TT/P2dlZzEkA4OPjg/z8fNy6dQvGxsbIycnBmDFjtLbJz8+Hn5+fVltJZ35KolQqxX+bmZmhSZMmSE1NBfB0/JcvX8auXbvEdTQaDXJzc/HgwQM4ODiUq6+yKm1/z8bq4OCA1157DampqWjTpg2ys7OxatUqHD58GA8ePEBhYSFyc3PL9YVdrVYjLS0NLVq0ENtkMhl8fHyK3ezaqFEjrVjK8oWkJuKnFOlVUFAQli9fjuzsbMyfP7/U9RQKBTp37iweBR08eDBiY2Mxb948cZ06derAxcWlTP3a2NggOzsbhYWFxU6zlVS8Pdsmk8m0Tu0AQGRkZLHEWLTN2bNnMWfOHIwdOxatW7eGpaUlduzYgb17976wX5lMJh5JqFu3LuLj43H8+HGcOHECH374IZo1a4Zly5aJRWtmZiYUCgWLT6IKqF+/PmQyGa5fv45mzZqVut6TJ08QEBCAcePGFVvm6Oj40n7K8jddHkWnxpctW6ZVpAJPi8dnyeXycu//ednZ2QgNDUVoaGixZc+eWSprX0ZGRlpHToGSbzJ6ldiXLFmCX375BZMmTUL9+vVhZmaGSZMmIT8/v9z7KovnPy+eHxc9xU8q0quOHTti4cKFsLS0RMuWLcu0jYmJCerVq1fsLvjyaNq0KQRBQGpqqta31fKyt7eHg4MD/vzzT63rkZ519uxZ1KtXD2FhYWLb7du3y92Xubk5AgICEBAQgJ49eyIsLAx//fWXeBoqNTUVTZs2faVxENFTNjY2aNWqFb7++mt07dq12BfUx48fw8rKCk2bNkVSUhKcnZ1LvdnxZV72N/2827dv4+HDh+LR1ZSUFJiamqJ+/foQBAGmpqa4e/cufH19Xyme56lUKvGIX25uLi5duoQuXboAeJpDr127VuYv/UWKzjw9P2Wcra0tHjx4IL7Pzs7GrVu3yhXrP//5TwBAWloa7ty5gwYNGgAAzpw5g+DgYAQEBAB4+mX9r7/+0trexMTkhfcZKBQKODg44MyZM+LPVxAEnD17VtwvlQ8LUNIrExMTbN++HTKZrMRv/RcvXsTq1avRq1cvNGjQADKZDIcOHcLRo0cxZ84crXUzMjKQlpam1WZtbV3iqXZbW1s0bdoUycnJFSpAgafXdy5duhQWFhZo06YN8vLykJKSAo1Gg379+sHFxQW3b9/GTz/9hObNm+Pnn3/GsWPHXnjt1fN27twJjUYDpVIJMzMz/Pjjj7CystI6zZecnIw2bdpUaCxEBLz//vsYOXIkRo0ahREjRsDd3R1PnjzBwYMHceLECaxevRqhoaGIj4/H7NmzMXToUNSuXRvXr19HUlISZs2a9dI+yvI3/bxatWohKioKEyZMQEZGBpYuXYrg4GDx0qXBgwfjk08+QWFhIVq0aIHHjx/jt99+Q506dcTCsTy2bNmC+vXrw9XVFevWrQMABAYGAgCGDRuGESNG4L///S+Cg4NhZmaGq1evIiUlBZMmTSp1n7a2tjAzM8PRo0fh4OCAWrVqQaFQoGXLlli2bBmOHTuGOnXqICYmplxHglevXg0rKytYW1tj6dKlcHV1RatWrQAArq6u2L9/P9q1aweNRoMvvvii2JcGZ2dnnDx5EkqlErVq1ULt2rWL9TFkyBCsWbMG9erVQ6NGjbB161bcvn27xKPA9HIsQEnvFApFqcvq1KmD1157DStXrsSdO3dgZGQEFxcX/Oc//yl2wf67775bbPvPP/+81Hkxe/fujR9//LHCE9H369cPcrlcvBPWwsICTZo0wbBhwwAAnTp1wqBBg7Bo0SIUFBSgY8eOGD58uNZNSi9jZWWFdevW4dNPP4UgCGjSpAmWLFkizqGalpaG5ORkrUsSiOjVuLq6YsOGDfjyyy/x0Ucf4eHDh7C3t4e3tzemTp0K4OkNkl9++SWWLVuG9957D/n5+ahXrx46depUpj5e9jddEjc3N7Rr1w4TJkyAWq1GQECAVrE3fvx42Nra4ssvv8Tt27dRu3ZteHp6Ijw8/JV+Du+99x6+/PJLXL58Ge7u7vjss8/EYrdp06b43//+h5UrV2LEiBEwNjZG/fr1X3ojlYmJCSIiIhATE4MVK1bA19cXq1evRkhICC5evIiZM2fC3Nwco0ePLtcR0HHjxmHx4sX4448/4OnpicWLF4vX006ZMgVz587FO++8Azs7O4wcOVK8A77I5MmT8dlnn+G7776Dk5MTdu7cWayPwYMHIysrC//973+RkZGBRo0aYenSpWW65IKKkwm8OIFqqJycHLz99tv473//K/mF8rr2xRdf4NGjR2U68kJEhqfoJpoNGzZUel+3b99GcHAwNm3aVOEzRESl4RFQqrHMzc0xd+7canGHorW1NQYOHKjvMIiIiMqEBSjVaM9PT2KohgwZou8QiIiIyoyn4ImIiIhIp/gkJCIiIiLSKRagRERERKRTLECJiIiISKdYgBIRERGRTrEAJSIiIiKdYgFKRERERDrFApSIiIiIdIoFKBERERHpFAtQIiIiItIpFqBERAZk9OjRqFevHmrXrg13d3csXLhQXObu7g65XA6FQgGFQgEvLy+tbZOSkqBUKmFhYQF/f3+cOXNG1+ETEQFgAVqioKAgfYdARFSiyZMn48qVK8jMzMShQ4cQGxuLLVu2iMu3bdsGtVoNtVqNc+fOie0PHjxASEgIpk+fjvT0dAwaNAjBwcHIzc0tV//Mj0QkBRagJcjJydF3CEREJfL09IRcLhffGxkZ4cqVKy/dLj4+Ho0aNcKwYcNgZmaGKVOmQKPRIDExsVz9Mz8SkRRYgBIRGZj//Oc/sLS0hKurK7KysjBkyBBx2fDhw+Ho6IiAgAAcOXJEbFepVPD19RXfy2Qy+Pj4QKVSldpPdHQ0nJyctF7Z2dmVMiYiqllYgBIRGZiPPvoIarUaJ0+exODBg2FrawsAiI2NxfXr13Hz5k0MGDAAPXr0wI0bNwAAarUaNjY2WvuxsbHB48ePS+0nIiIC9+7d03pZWFhU2riIqOZgAUpEZIBkMhn8/f1hbm6OyMhIAED79u0hl8shl8sxduxY/OMf/8Du3bsBAAqFAhkZGVr7yMjIgJWVlc5jJyJiAUpEZMAKCgpw9erVEpcZGRlBEAQAgFKpRHJysrhMEAScPXsWSqVSF2ESEWlhAUpEZCDS09OxYcMGZGZmQqPR4MiRI1i5ciW6dOmCmzdv4tChQ8jLy0NeXh5iYmLwyy+/oFu3bgCAvn374vLly4iNjUVeXh6WLl0KAOjSpYs+h0RENZSJvgMgIqKykclkWLduHSZOnIiCggLUq1cP06ZNw/jx43HhwgVMmDABV65cQa1ateDp6Yldu3bBw8MDAGBvb4/t27dj/PjxGDVqFJRKJRISEmBmZqbnURFRTSQTis7PkKhr16746aef9B0GEVGVw/xIRFLgKXgiIiIi0qkqWYAuX74cfn5+MDMzw8CBA0tc58CBA5DJZJgxY4ZW+9atW+Hh4QELCwu8+eab4hQkJJ2Mk1PxcH8frVfGyan6DouISDR1xzn0WXdS6zV1x7mXb0hEOlElrwF1dnbG7NmzkZiYiLS0tGLL8/LyMHHiRLRu3Vqr/cKFCwgLC0N8fDw6dOiAmTNnon///jhx4oSuQq8RCtWpKHh0Xt9hEBGVKvVhFs7/pdZ3GERUiipZgPbt2xcAkJycXGIB+tFHH6Fnz564c+eOVntsbCwCAwPFuz7nzZsHR0dHnDt3Dl5eXpUfOBERERG9VJU8Bf8ily5dwsaNGzFnzpxiy55/1JyVlRU8PDz4qDkiIiKiKsTgCtCxY8di8eLFJT4Ojo+aIyIiIqr6DKoAjY2NhVwuR0hISInL+ag5IiIioqqvSl4DWprExEQcPnwYdevWBfC0uDQ2NsapU6eQmJhY7FFzarUaV69e5aPmiIiIiKqQKnkEtKCgADk5OSgoKIBGo0FOTg7y8/OxdOlS/P7770hOTkZycjKCg4PxzjvvYPPmzQCAIUOGYPfu3UhMTEROTg4iIyPh4+PDG5CIiIiIqpAqWYDOnz8fcrkcCxYsQFxcHORyOUaNGgVra2vUrVtXfMnlclhaWsLe3h4A0Lx5c6xbtw6jR4+GnZ0dTp8+jS1btuh5NERERET0rCp5Cj4qKgpRUVEvXW/9+vXF2kJDQxEaGip9UEREREQkiSp5BJSIiIiIqi8WoERERESkUyxAiYiIiEinWIASERERkU6xACUiIiIinWIBSkREREQ6xQKUiIhqBBMjmb5DIKL/VyXnASUiIpKaq60cU3ecQ+rDLK32BnaW+DSET8wj0iUeASUiMiCjR49GvXr1ULt2bbi7u2PhwoXiMpVKhTZt2sDCwgKenp7Yv3+/1rZbt26Fh4cHLCws8Oabb+LGjRu6Dl/vUh9m4fxfaq3X8wUpEVU+FqBERAZk8uTJuHLlCjIzM3Ho0CHExsZiy5YtyM/PR1BQEIKDg5Geno7IyEj06dMH9+7dAwBcuHABYWFhWLlyJR48eAAfHx/0799fz6MhopqKBSgRkQHx9PSEXC4X3xsZGeHKlSs4cOAAsrOzMWPGDJiZmWHAgAFQKpWIi4sDAMTGxiIwMBDdunWDXC7HvHnzcObMGZw7d05fQyGiGowFKBGRgfnPf/4DS0tLuLq6IisrC0OGDIFKpYK3tzeMjP5O676+vlCpVACenp739fUVl1lZWcHDw0NcXpLo6Gg4OTlpvbKzsyttXERUc/AmJCIiA/PRRx9h4cKFOHXqFLZv3w5bW1uo1WrY2NhorWdjYyNe51na8sePH5faT0REBCIiIrTaunbtKskYpFLSTUXtG9jrKRoiKiseASUiMkAymQz+/v4wNzdHZGQkFAoFMjIytNbJyMiAlZUVALx0uaEq6aai25lP9B0WEb0EC1AiIgNWUFCAq1evQqlUIiUlBRqNRlyWnJwMpVIJAFAqlUhOThaXqdVqcTsiIl1jAUqVLuPkVDzc30frlXFyqr7DIjI46enp2LBhAzIzM6HRaHDkyBGsXLkSXbp0QUBAAORyORYvXozc3FzExcUhJSUFoaGhAIAhQ4Zg9+7dSExMRE5ODiIjI+Hj4wMvL85/SUS6x2tAqdIVqlNR8Oi8vsMgMngymQzr1q3DxIkTUVBQgHr16mHatGkYP348ZDIZEhISEB4ejrlz58Ld3R3x8fFwcnICADRv3hzr1q3D6NGjcffuXbRp0wZbtmzR84iIqKZiAUpEZCBsbGyKTS7/LG9vb5w4caLU5aGhoeIRUSIifaqSp+CXL18OPz8/mJmZYeDAgWL7pUuX0Lt3b9SpUwc2Njbo1KkTTp06pbUtn/RBREREVLVVyQLU2dkZs2fPxqhRo7TaHz16hLfeegvnzp3DgwcP0K9fP/To0QNZWU+n4OCTPoiIiIiqvip5Cr5v374Ant7BmZaWJra3atUKrVq1Et9PmDAB77//Pn7//Xe0bNlS60kfADBv3jw4Ojri3LlzvNC+ssmk+1XKODkVhepUrTZjRQNYt/pUsj6IiIhIf6pkAVpWv/zyCzQaDRo1agTg6ZM+/P39xeXPPumjtAI0Ojoa0dHRWm2NGzeuvKCrKWNL1xILR1On9uXeF29aIiIiqt4MtgB98OABhgwZgnnz5sHa2hpA9X3Sh76Ut6AsqXA0VjSslNiIiIjIcBlkAZqRkYHu3bsjMDAQ06dPF9ur65M+9IUFJREREVWGKnkT0osUFZ/+/v5YunSp1jI+6YOIiIio6quSBWhBQQFycnJQUFAAjUaDnJwc5OfnIzMzE4GBgfD09MSKFSuKbccnfRARERFVfVWyAJ0/fz7kcjkWLFiAuLg4yOVyjBo1Ctu2bcPx48exefNmWFlZQaFQQKFQ4NChQwC0n/RhZ2eH06dP80kfRERERFVMlbwGNCoqClFRUSUuGz58+Au35ZM+iIiIiKq2KnkElIiIiIiqLxagRERERKRTLECJiIiISKdYgBIRERGRTrEAJSIiIiKdYgFKRERERDrFApSIiIiIdIoFKBERERHpFAtQIiIiItIpFqBEREREpFMsQImIDERubi7Cw8PRoEEDWFlZwcvLCxs3bhSXu7u7Qy6XQ6FQQKFQwMvLS2v7pKQkKJVKWFhYwN/fH2fOnNH1EIiIALAAJSIyGAUFBXB2dsa+ffuQmZmJVatWYezYsTh27Ji4zrZt26BWq6FWq3Hu3Dmx/cGDBwgJCcH06dORnp6OQYMGITg4GLm5ufoYChHVcCxAiYgMhKWlJebNm4eGDRtCJpOhffv2aNeuHY4ePfrSbePj49GoUSMMGzYMZmZmmDJlCjQaDRITE3UQORGRNhagREQGKisrC6dOnYJSqRTbhg8fDkdHRwQEBODIkSNiu0qlgq+vr/heJpPBx8cHKpWq1P1HR0fDyclJ65WdnV0pYyGimoUFKBGRAdJoNAgLC4O/vz+6desGAIiNjcX169dx8+ZNDBgwAD169MCNGzcAAGq1GjY2Nlr7sLGxwePHj0vtIyIiAvfu3dN6WVhYVNqYiKjmkLQA3bJlC37++Wfx/YIFC/DGG29gwIABuH79upRdEREZDKlzoyAIGDNmDG7fvo3NmzdDJpMBANq3bw+5XA65XI6xY8fiH//4B3bv3g0AUCgUyMjI0NpPRkYGrKysXn1gRESvSNICdOPGjbC2tgYAnDp1Cnv37sW8efPQqFEjfPrpp1J2RURkMKTMjYIgYNy4cUhOTsbu3buhUChKXdfIyAiCIAAAlEolkpOTtfZz9uxZrdP3RES6YiLlzu7fvw9nZ2cAT6f76NKlC7p27YrGjRtjxIgRUnZFJMo4ORWF6lStNmNFA1i34pceqhqkzI3jx4/H8ePHsW/fPtSuXVtsv3nzJm7cuIHWrVsDAL766iv88ssvWLNmDQCgb9++iIiIQGxsLPr3748VK1YAALp06SLFEImIykXSI6AKhQL37t0DABw9elRMhDKZDIWFhWXez/Lly+Hn5wczMzMMHDhQa5lKpUKbNm1gYWEBT09P7N+/X2v51q1b4eHhAQsLC7z55pvi9U9UfRWqU1Hw6LzW6/mClEifpMqNN27cwIoVK3D+/Hm4uLiI830uXLgQarUaEyZMgJ2dHerWrYuvvvoKu3btgoeHBwDA3t4e27dvx6JFi2BtbY1vvvkGCQkJMDMzk37AREQvIekR0M6dO2PWrFlwcXFBeno62rZtCwC4dOkS6tWrV+b9ODs7Y/bs2UhMTERaWprYnp+fj6CgIIwaNQpJSUnYvn07+vTpg8uXL8PJyQkXLlxAWFgY4uPj0aFDB8ycORP9+/fHiRMnpBwmEVG5SJUb3dzcxFPqJXn2FHtJAgICXnjXOxGRrkh6BHTatGno378/GjZsiC+++EK8Nun+/fsIDQ0t83769u2L3r17w8HBQav9wIEDyM7OxowZM2BmZoYBAwZAqVQiLi4OwNM7QAMDA9GtWzfI5XLMmzcPZ86c0ZqMmaoImaTffXS/f6JykCo3UuUwMZLpOwSiGkfST2kTExMMHTq0WPvgwYMl2b9KpYK3tzeMjP6um319fcVv9CqVCv7+/uIyKysreHh4QKVSFXskXZHo6GhER0drtTVu3FiSeKl0xpauJV67aerUvlL3z2tDSR8qOzdSxbjayjF1xzmkPszSam9gZ4lPQ0r+7CCiipH8MNH169fx22+/4eHDh8VOFY0aNapC+y5tHruKznMXERGh1da1a9cKxUllU3Tt5rOMFQ0rdf9E+lKZuZEqLvVhFs7/pdZ3GEQ1hqQFaFxcHKKjo2Fvbw8HBwdxbroiFU2yL5vHjvPcEVFVVNm5kYjI0EhagK5fvx4TJ07EkCFDpNytSKlU4uOPP4ZGoxFPwycnJ2PQoEHi8mcvwler1bh69SrnuSMivars3EhEZGgkLUCzs7MREBBQ4f0UFBSIL41Gg5ycHBgbGyMgIAByuRyLFy/GlClTkJCQgJSUFMTHxwMAhgwZAn9/fyQmJqJ9+/aIjIyEj49Pqdd/0lOVeS0mEUmXG4mIqgtJ74IPCgrCvn37Kryf+fPnQy6XY8GCBYiLi4NcLseoUaNgamqKhIQEbNu2DTY2Nvjggw8QHx8PJycnAEDz5s2xbt06jB49GnZ2djh9+jS2bNlS4Xiqu5Lm0dRk39Z3WETVhlS5kYioupD8JqT169fjxIkT8PDwgImJ9u4nTZpUpn1ERUUhKiqqxGXe3t4vnNczNDSU05oQUZUjRW4kIqouJC1AL168iCZNmqCwsBCXLl3SWvb8RfdERDUFcyMRkTZJC9BVq1ZJuTuSGK/1JNIP5kYiIm2V8riYnJwc3Lp1CwBQv359mJubV0Y3VE6VPe8mEb0YcyMR0VOSFqD5+flYtmwZvvvuO+Tl5QEAatWqhX79+mHChAnFrnsiIqoJmBuJiLRJmvWWLFmCAwcOYO7cuWjRogWAp/N0LlmyBBqNBtOmTZOyOyIig8DcSESkTdJpmPbu3YsPPvgAXbp0gaOjIxwdHdG1a1fMmTMHP/74o5RdEREZDOZGIiJtkhag2dnZ4pycz3JyckJWVpaUXRERGQzmRiIibZIWoEqlEmvWrEF+fr7YlpeXhzVr1sDb21vKroiIDAZzIxGRNkmvAZ02bRomTJiAt956C02bNgXwdP47Y2NjLF++XMquiIgMBnMjEZE2SQvQJk2aYMeOHfjhhx9w/fp1AMCbb76JHj16cLoRIqqxmBuJiLRJPveHubk5+vbtK/VuiYgMGnMjEdHfKlyAHj58GG3atIGJiQkOHz78wnXbt+dTd4ioZmBuJCIqXYUL0ClTpuDHH3+EnZ0dpkyZUup6MpkMJ0+erGh3REQGoTJyY25uLsaNG4d9+/YhLS0Nrq6umDVrFgYPHgwAUKlUCA8Px9mzZ+Hu7o7ly5ejc+fO4vZbt27F+++/jzt37uCNN97A2rVr4ebmVrGBEhG9ggoXoL/88kuJ/yb94TPfifSvMnJjQUEBnJ2dsW/fPjRo0ABHjhxBz5490aBBA/j5+SEoKAijRo1CUlIStm/fjj59+uDy5ctwcnLChQsXEBYWhvj4eHTo0AEzZ85E//79ceLECUliIyIqD0mnYdq1a5f4mLln5efnY9euXVJ2RS9Q9Mz3Z1+a7Nv6DqtiZHxUIRkuqXKjpaUl5s2bh4YNG0Imk6F9+/Zo164djh49igMHDiA7OxszZsyAmZkZBgwYAKVSibi4OABAbGwsAgMD0a1bN8jlcsybNw9nzpzBuXPnJBsnEVFZSVqAzps3D2q1ulh7VlYW5s2bJ2VXVMMYW7oi4+RUPNzfR+v1WPWJvkMjeqnKyo1ZWVk4deoUlEolVCoVvL29YWT0d1r39fWFSqUC8PT0vK+vr7jMysoKHh4e4vKSREdHw8nJSeuVnZ39yvESERWRtAAVBAEymaxY+/3796FQKKTsimqganlkl2qEysiNGo0GYWFh8Pf3R7du3aBWq2FjY6O1jo2NDR4/fgwAL11ekoiICNy7d0/rZWFh8UrxEhE9S5LzmsOGDQPw9GL68ePHw9jYWFym0Whw69YttGnTRoquiIgMRmXlRkEQMGbMGNy+fRs//vgjZDIZFAoFMjIytNbLyMiAlZUVALx0ORGRLklSgBZNIXLhwgW0atUKcrlcXGZqaorXXnsNb775phRdAQCuX7+OcePG4dixYzAxMUFgYCC++OILWFlZ4Y8//sDIkSNx5MgRODk54aOPPsLAgQMl65uIqKwqIzcKgoBx48YhOTkZiYmJ4hFUpVKJjz/+GBqNRjwNn5ycjEGDBonLk5OTxf2o1WpcvXoVSqWyIkMkInolkhSgo0ePBgA4Ozuja9euMDMzk2K3L+yvTp06+PPPP5GTk4O3334bc+bMwZIlSzBo0CD4+Phgx44dOHHiBIKCgqBUKplkiUjnKiM3jh8/HsePH8e+fftQu3ZtsT0gIAByuRyLFy/GlClTkJCQgJSUFMTHxwMAhgwZAn9/fyQmJqJ9+/aIjIyEj48PvLy8KhwTEVF5SXoNaK9evSq9+ASA1NRUDBo0CHK5HLa2tnj77beRkpKCy5cv48SJE1iwYAHkcjkCAgIQHByMr776qtJjIiIqjVS58caNG1ixYgXOnz8PFxcXKBQKKBQKLFy4EKampkhISMC2bdtgY2ODDz74APHx8XBycgIANG/eHOvWrcPo0aNhZ2eH06dPY8uWLRWOiYjoVUg6t01+fj5iYmKQmJiIu3fvoqCgQGu5VBPRT548GRs3bkTHjh2Rk5ODuLg49OrVCyqVCm5ubrC1tRXX9fX1xf79+0vdV3R0NKKjo7XaGjduLEmcVAVxOifSA6lyo5ubGwRBKHW5t7f3C+f1DA0NRWhoaNmCJiKqRJIeAV2+fDn27t2LESNGQCaT4d///jeGDh0KW1tbzJo1S7J+AgIC8Pvvv8Pa2hqOjo4wMzPDxIkTeZcnvVRp0zllnJyq79CoGtNVbiQiMhSSFqD79u3Df/7zH/Tq1QvGxsZ44403MGHCBLz33ns4cOCAJH0UFhYiMDAQPXv2RFZWFjIyMlCvXj0MGTKEd3lSmZQ0ndPzT44ikpIuciMRkSGRtAB99OiR+FxhS0tLZGZmAgD8/Pxw6tQpSfpIT0/HrVu3MGHCBJibm6N27doYO3YsfvjhByiVSty4cQOPHj0S109OTuYNSESkV7rIjUREhkTSArRevXq4c+cOAMDd3R379u0DABw5ckSyo5AODg5o2LAhVqxYgby8PGRlZWH16tVo0aIFGjduDH9/f8yePRtPnjzBwYMHkZCQgOHDh0vSNxHRq9BFbiQiMiSS3wX/+++/AwDCwsKwefNmtGvXDv/9738xdOhQyfrZtm0bkpKSULduXbi6uuL27dv4+uuvAQCbNm3CxYsXYW9vj2HDhmHVqlU8AkpEeqWr3EjSMjEq/vQqIpKGpLcEP5tIW7dujbi4OPz+++9wcXGR9M5yHx8f8QjC81xcXPDTTz9J1hcRUUXpKjeStFxt5Zi64xxSH2ZptTews8SnIZw/lagiKnVOGmdnZzg7O1dmF0REBoe50XCkPszC+b/U+g6DqNqRvADdt28fNm7ciOvXrwN4er3ToEGD0KVLF6m7IiIyGMyNRER/k7QAXb9+PWJiYhASEiJOdnz27FlERUXh1q1bCAsLk7I7IiKDwNxIRKRN0gJ048aN4lx3RQIDA+Hp6YnPP/+cSZaIaiTmRiIibZLeBZ+XlwcfH59i7T4+PsjNzZWyKyIig8HcSESkTdICtEePHtixY0ex9p07d6J79+5SdkVEZDCYG4mItFX4FPzSpUv/3pmJCbZu3Ypjx46Jc2+eO3cOt27dQnBwcEW7IiIyGMyNRESlq3ABev78ea33zZo1AwDcuHEDAKBQKNCsWTNcvny5ol0RERkM5kYiotJVuABdtWqVFHEQEVUrzI1ERKWT7BrQgoICdOrUCVevXpVql0S6I6vUZzJQDcbcSERUnGSfuiYmJrC2toYgCFLtkkhnjC1dkXFyKgrVqVrtpk7tocn+s1i7saIBrFt9qssQyUAxNxIRFSfpXfBhYWH44osvoFbzsWVkeArVqSh4dF7rpcm+XWL78wUp0YswNxIRaZP0vOP27dtx/fp1dO/eHfXq1YO5ubnW8q+//lrK7oiIDAJzIxGRNkkL0Pbt26N9+/ZS7pKIyOBJlRuXL1+O9evXIyUlBX369MGmTZvEZe7u7vjrr79gbGwMAHBzc8O5c+fE5UlJSRg3bhyuXbsGLy8vrFmzBi1atKhwTEREr0LSAnT06NFS7o5eorRrFomoapEqNzo7O2P27NlITExEWlpaseXbtm1DYGBgsfYHDx4gJCQEn3/+OQYMGIAvvvgCwcHBuHTpEszMzCSJjYioPCS9BpR0q7RrFomoeurbty969+4NBweHcm0XHx+PRo0aYdiwYTAzM8OUKVOg0WiQmJhYSZESEb2YpEdA/f39IZPJSl1+8uRJKbsjIjIIusqNw4cPh0ajgZeXFxYsWIB27doBAFQqFXx9fcX1ZDIZfHx8oFKp0LNnz1L3Fx0djejoaK22xo0bSxIrEdVskhagn332mdb7goICXLx4ETt37sSoUaOk7IqIyGDoIjfGxsaiZcuWAID169ejR48eSElJgZubG9RqNWxtbbXWt7GxwePHj1+4z4iICERERGi1de3aVZJ4iahmk/wmpOcFBATAw8MDCQkJCAkJkayv7777DpGRkUhNTYWDgwM+++wz9O3bFyqVCuHh4Th79izc3d2xfPlydO7cWbJ+iYjKSxe58dk+xo4di02bNmH37t0YM2YMFAoFMjIytNbPyMiAlZVVhfslInoVOrkGtGnTpvjtt98k29/+/fsxefJkrFq1Co8fP8Yvv/wCX19f5OfnIygoCMHBwUhPT0dkZCT69OmDe/fuSdY3EQA+OYkkIXVufJaRkZE4+b1SqURycrK4TBAEnD17FkqlslL6JiJ6GUkL0JycHK3XkydPcOvWLcTExKBevXqS9fPBBx/ggw8+QLt27WBkZAQnJyc0bNgQBw4cQHZ2NmbMmAEzMzMMGDAASqUScXFxkvVNBPz95KSH+/tovTJOTtV3aFQFSZUbCwoKkJOTg4KCAmg0GuTk5CA/Px83b97EoUOHkJeXh7y8PMTExOCXX35Bt27dADy9eeny5cuIjY1FXl4eli5dCgDo0qVLpYyXiOhlJD2M06FDh2IX2guCACcnJyxYsECSPgoLC3Hy5EkEBQWhSZMmUKvV6N69O5YsWQKVSgVvb28YGf1dV/v6+kKlUpW6P15kT6+qaBYCopeRKjfOnz8fc+fOFd/HxcVh+PDhmD59OiZMmIArV66gVq1a8PT0xK5du+Dh4QEAsLe3x/bt2zF+/HiMGjUKSqUSCQkJnIKJiPRG0gL0f//7n9Z7IyMj2Nraon79+jAxkaarv/76C/n5+di0aRP2798PhUKBwYMHY/LkyWjYsCFsbGy01rexscGNGzdK3R8vsieiyiZVboyKikJUVFSJy549xV6SgICAF34ZJyLSJUkL0KKpPZ5PqAUFBfjtt9/w+uuvV7gPCwsLAMD48eNRv359AMCsWbPQu3dvzJw5kxfaE1GVo4vcSERkSCS9BnTMmDHIzMws1q5WqzFmzBhJ+rCxsYGLi0uJc+oplUqkpKRAo9GIbcnJybzQnoj0She5kYjIkEh6BFQQhBILw8zMTMjlcsn6CQ8Px/Lly/HWW2/B0tISixYtQnBwMAICAiCXy7F48WJMmTIFCQkJSElJQXx8vGR9ExGVl65yY3U2dcc5pD7M0mpr38BeT9EQUUVJUoDOmDEDwNPTTB9++CFq1aolLissLMTly5fh4+MjRVcAgJkzZyItLQ2enp4wMTFBz5498dlnn8HU1BQJCQkIDw/H3Llz4e7ujvj4eDg5OUnWNxFRWek6N1ZnqQ+zcP4vtVZbQ3sLPUVDRBUlSQFa9A1eEASYmZnB3Nz87w5MTBASEoI+ffpI0ZW4z88//xyff/55sWXe3t44ceKEZH0REb0qXedGIiJDIUkBGhkZCQAwNzfH+PHjYWlpKcVuiYgMGnMjEVHJJLsJqbCwEPHx8bh//75UuyQiMnjMjURExUlWgBobG8PFxQVZWVkvX5mIqIZgbiQiKk7SaZjGjx+PpUuX4vr161LulojIoDE3Vi8mRsVnNCCi8pF0Gqb58+cjOzsb/fv3L3bBPQD89NNPUnZHRGQQmBurF1dbeYnTQjWws8SnIV56iorIsEhagE6ePFnK3RERVQvMjdVPSdNCEVHZSVqA9urVS8rdERFVC8yNRETaJL0GFADS0tKwYcMGfPTRR3j06BEA4PTp07h165bUXRERGQzmRiKiv0lagKpUKrz99tvYv38/duzYAbX66emJX3/9FV988YWUXRERGQzmRiIibZKegl+yZAmGDx+OESNGoGPHjmJ7mzZtsG3bNim7qlEyTk5FoTpVq83Uqb2eoiGi8mJuJCLSJukR0EuXLqF79+7F2u3s7JCeni5lVzVKoToVBY/Oa7002bf1HRaVRCbpdzqqJpgbiYi0SfppaWlpiYcPH6JevXpa7ZcuXYKjo6OUXRFVScaWriUesTZWNIB1q0/1FBXpG3MjEZE2SY+AduvWDcuWLUN6ejpksqcT9apUKixduhQ9evSQsiuiKqukI9bPF6RUszA3EhFpk7QAHTduHNzc3PDWW2+Jky6PGDECzZs3R3h4uJRdEREZDOZGIiJtkpyC12g0+Prrr3Hw4EEUFBSgR48e6Ny5M548eYImTZrAzc1Nim6IiAwKcyMRUckkOQK6du1axMTEwMPDAz4+PkhKSsK+ffvQtWtXJlgiqrGkzo3Lly+Hn58fzMzMMHDgQK1lKpUKbdq0gYWFBTw9PbF//36t5Vu3boWHhwcsLCzw5ptv4saNGxUaGxFRRUhSgH7//feYM2cOZs2ahX//+99YunQp9uzZA41GI8XuiYgMktS50dnZGbNnz8aoUaO02vPz8xEUFITg4GCkp6cjMjISffr0wb179wAAFy5cQFhYGFauXIkHDx7Ax8cH/fv3r/D4iIhelSQF6N27d/H666+L75VKJYyMjHD//n0pdk9UI2WcnIqH+/tovTJOTtV3WFQOUufGvn37onfv3nBwcNBqP3DgALKzszFjxgyYmZlhwIABUCqViIuLAwDExsYiMDAQ3bp1g1wux7x583DmzBmcO3fu1QdHRFQBkhSghYWFMDU11WozNjZGQUGBFLsvVVpaGhwcHNCmTRux7WWnoYgMBe+mN3y6yo0qlQre3t4wMvo7pfv6+kKlUonLfX19xWVWVlbw8PAQl5cmOjoaTk5OWq/s7GxJYyeimkmSm5AEQcCHH36IWrVqiW25ublYvHgx5HK52LZo0SIpuhNFRETA09MTeXl5AP4+DTVq1CgkJSVh+/bt6NOnDy5fvgwnJydJ+yYiehld5Ua1Wg0bGxutNhsbG/E6z9KWP378+IX7jYiIQEREhFZb165dKxQrEREg0RHQXr16wdraGnK5XHy99dZbsLOz02qTUlJSEi5fvox33nlHbHvZaSgiveETkmokXeVGhUKBjIwMrbaMjAxYWVmVaTkRka5J8qkYGRkpxW7KLC8vD+PHj0dsbCxOnz4ttr/sNFRJoqOjER0drdXWuHFj6YOmGo1PSKqZdJUblUolPv74Y2g0GjH/JScnY9CgQeLy5ORkcX21Wo2rV69CqVTqJD4ioudJOhG9rixatAhdunRBixYttNpf5TRTREQE7t27p/WysLCojLCphuM1nVRRBQUFyMnJQUFBATQaDXJycpCfn4+AgADI5XIsXrwYubm5iIuLQ0pKCkJDQwEAQ4YMwe7du5GYmIicnBxERkbCx8cHXl5eeh4REdVUBleAXrlyBevXr8fcuXOLLeNpJjI4PDVP5TB//nzI5XIsWLAAcXFxkMvlGDVqFExNTZGQkIBt27bBxsYGH3zwAeLj48Vr35s3b45169Zh9OjRsLOzw+nTp7FlyxY9j4aIajKD+/Q7fPgw7t69iyZNmgAAnjx5gidPnqBu3bpYtWoVUlJSSj0NRVTVlHZq3tSpvZ4ioqosKioKUVFRJS7z9vbGiRMnSt02NDRUPCJKRKRvBleADhgwAIGBgeL7zZs34+uvv8b3338Pe3t78TTUlClTkJCQgJSUFMTHx+sxYqIXKzo1/yxjRUM9RUNERFT5DK4Aff6uUWtra5iamqJu3boAgISEBISHh2Pu3Llwd3fXOg1FRERERPpncAXo88LCwhAWFia+f9lpKCIiIiLSL4O7CYmIiIiIDBsLUCIiIiLSKRagRIaE0zYREVE1wE8zIgNS6hOVrBrB2j+6lK2ISBdMjGT6DoHIYLAAJTIwpU3bxEd9EumXq60cU3ecQ+rDLK32BnaW+DSET50iehYLUKJqoqTClIh0K/VhFs7/pdZ3GERVHq8BJSIiIiKdYgFKRERERDrFApSIiKgS8eYkouJ4DSgREVEl4s1JRMWxACUiIqpkvDmJSBtPwRMRERGRTrEAJSIiIiKdYgFKRERERDrFa0CrkJKeZGPq1F5P0RARERFVDhagVUhpj1gkIiIiqk54Cp6IiEgPOD8o1WQ8AkpEVA2EhYVh48aNqFWrlth2/vx5uLq6AgD++OMPjBw5EkeOHIGTkxM++ugjDBw4UF/hlqqk+TLbN7DXUzSVi/ODUk1mcEdAc3NzER4ejgYNGsDKygpeXl7YuHGjuFylUqFNmzawsLCAp6cn9u/fr8doiYh0Z+rUqVCr1eKrqPgEgEGDBqFRo0ZIS0vDunXrMGrUKKhUKj1GW7Ki+TKffd3OfKLvsCpNSeN9viAlqo4MrgAtKCiAs7Mz9u3bh8zMTKxatQpjx47FsWPHkJ+fj6CgIAQHByM9PR2RkZHo06cP7t27p++wiYj05vLlyzhx4gQWLFgAuVyOgIAABAcH46uvvtJ3aERUQxlcAWppaYl58+ahYcOGkMlkaN++Pdq1a4ejR4/iwIEDyM7OxowZM2BmZoYBAwZAqVQiLi5O32ETGYyMk1PxcH8frVfGyan6DovKYPXq1bCzs0OLFi2wdu1asV2lUsHNzQ22trZim6+v70uPgEZHR8PJyUnrlZ2dXWnxE1HNYXAF6POysrJw6tQpKJVKqFQqeHt7w8jo72G9LMkywRJpK5qN4dnX89ODUdUzceJEXLp0Cffu3cOSJUswffp0fPfddwAAtVoNGxsbrfVtbGzw+PHjF+4zIiIC9+7d03pZWFhU1hCIqAYx6AJUo9EgLCwM/v7+6Nat2yslWSZYIqoOXn/9dTg4OMDExAT//Oc/MW7cOPHsj0KhQEZGhtb6GRkZsLKy0keoRESGexe8IAgYM2YMbt++jR9//BEymcxgkiwnnCeiymZkZARBEAAASqUSN27cwKNHj8Qv6cnJyVAqlXqMkIhqMoM8AioIAsaNG4fk5GTs3r0bCoUCwNMkm5KSAo1GI65bFZNsSac4Ndm39R0WERmwLVu24PHjx9BoNDh8+DCWL1+OPn36AAAaN24Mf39/zJ49G0+ePMHBgweRkJCA4cOH6zlqIqqpDLIAHT9+PI4fP44ff/wRtWvXFtsDAgIgl8uxePFi5ObmIi4uDikpKQgNDdVjtERElW/58uVwcXGBtbU13n33XcyfP19rns9Nmzbh4sWLsLe3x7Bhw7Bq1aoq9+WciGoOgzsFf+PGDaxYsQJmZmZwcXER22fOnImZM2ciISEB4eHhmDt3Ltzd3REfHw8nJyc9RkxEVPkOHjz4wuUuLi746aefdBQNEdGLGVwB6ubmJl7XVBJvb2+cOHFChxERVWGy0v/EeS0yERHpi8EVoERUdsaWrqUWmkXXImutr2ioy/CIiKiGYgFKVM2x0CQyLCZGMn2HQFTpWIASERFVIa62ckzdca7YM+HbN7DHnxk5xdob2Fni0xAvXYZIVGEsQImIiKqY1IdZOP+XWqutob0FUh9mF2snMkQGOQ0TERERPcVT9mSIeASUiIjIgJV2yp6n5qkqYwFKRERk4Eo6ZU9UlfEUPBERERHpFAtQInq5F0xoT0REVF78VCGil3rRhPaa7D+LtRsrGsC61ae6DJGIiAwIC1AiKpPSJrQvqZ2IiOhFeAqeiIiIiHSKR0CJiIhqGE7bRPrGApSIiKiG4bRNpG8sQCtRaTdtEBERVTY+IYmqMhaglai0mzaIqj1O20Skd6U9Ial9A3s9RUT0N35KSIBHOom0lTZt06tMz1Te/ZS4vlUjWPtHl6tfouqgpFPtDe0tSlyXR0xJl1iASoBHOomKK3F6plc4Mlre/ZT29yhVQUzS4JG5qudFR0z/zMgpcztvZqKyqHYF6KNHjzB69Gjs3r0bVlZWmD59OiZPnqzvsIgI0h0ZfdHE+KXhfKVVKz+W58gc6U5p/y+pD7PL3P6iI6klFbiN7BWIDvasYOQvpq9+qXTVrgAdP348cnNz8eeff+LGjRt488030bRpU/To0UPfoRERKvfIKM88vBjzI+nCi46kllbglndaqPIWlFL1S9KpVgVoVlYW4uLi8Ouvv6J27drw9vbGqFGjsHbt2kpNsMaKBsXajCycAQhsZzvby9Bu6tgGWReWofDJXe12W+9K/fsqad/Vlb7y47LDqbibmaPV5v1abTSwsyy2rnNtOQSh+NEzthte+58ZOcXaAZT6/17a+uVRt7ZZuX/fpOi3Oijp51a3tjkmtK+8HCkTBKF4VjZQp0+fRqtWrZCfny+2xcXF4YMPPsCFCxdK3CY6OhrR0do3J7i5ucHGxqbUfrKzs2FhUTNOFdWksQIcb3UnxXjNzc2xc+dOiSLSHV3lx+qkpv19lKSm/ww4/vKNvzz5sVodAVWr1bC2ttZqs7GxwePHj0vdJiIiAhEREeXqx8nJCffu3XulGA1NTRorwPFWdzVtvM/SVX6sTmry70uRmv4z4Pgrb/zV6lnwCoUCmZmZWm0ZGRmwsrLSU0RERFUD8yMRVSXVqgBt0qQJZDIZzp07J7YlJydDqVTqMSoiIv1jfiSiqqRaFaCWlpbo168fZs2ahcePH0OlUmHNmjUYMWKEvkMjItIr5kciqkqqVQEKAF988QVMTU3x2muvoWvXrpgxY4bkd3jWpGuiatJYAY63uqtp432eLvJjdVLTf18A/gw4/sobf7W6C56IiIiIqr5qdwSUiIiIiKo2FqBEREREpFMsQImIiIhIp1iAEhEREZFOsQAlIiIiIp1iAUpEREREOsUCtAzmzJkDR0dHWFtbIzw8HLm5uSWud+/ePQwePBj16tVD7dq14efnh7179+o42vJ59OgR+vfvDysrKzg7O2PJkiWlrpuUlASlUgkLCwv4+/vjzJkzugtUImUd7/Hjx9G9e3fY29vD3t4ePXv2xOXLl3UbrATK8/9bZP369ZDJZPjf//5X+QFKrDzjzcnJwaRJk+Dk5ITatWujZcuWL3wuOlU/Zf19ycvLQ79+/eDu7g6ZTIY9e/boNtBKVNNy4vPKOv5r166hdevWsLOzg42NDdq2bYvDhw/rNthKoNfPCIFeKCYmRmjQoIFw9epVIS0tTWjXrp0wffr0Ete9evWqEB0dLdy6dUsoLCwUvvvuO8HS0lK4fv26jqMuu3/9619CcHCwkJGRIZw9e1ZwdHQUfvjhh2LrpaWlCdbW1sJXX30l5OTkCJ988ong6uoq5OTk6CHqV1fW8f7www/Cpk2bhEePHgm5ubnC9OnThWbNmukh4oop63iLpKWlCU2aNBG8vLyElStX6jBSaZRnvMOHDxf69esn3L17VygsLBSSk5MN7veZKqasvy+5ubnCZ599Jhw8eFCoX7++sHv3bj1EWzlqWk58XlnHn5mZKVy+fFkoLCwUNBqN8N133wk2NjZCbm6uHqKWjj4/I1iAvkTbtm2FZcuWie/37t0rODo6lnn75s2bC1u3bq2M0CpMrVYLtWrVElJSUsS2mTNnCv369Su27urVq4WWLVuK7zUajVC/fn1h165dOolVCuUZ7/P++usvAYCQlpZWmSFK6lXGGxYWJvzvf/8TOnXqZHAFaHnG+/vvvwsKhUJIT0/XYYRUlbxqPnBzc6s2BWhNy4nPe9XxFxYWCtu3bxcACH/++Wdlh1lp9P0ZwVPwL6FSqeDr6yu+9/X1xf379/HXX3+9dNvbt2/jypUr8PLyqsQIX92lS5eg0WigVCrFNl9fX6hUqmLrPv9zkMlk8PHxKXHdqqo8431eUlIS6tatC3t7+8oMUVLlHW9SUhIuXLiAUaNG6SpESZVnvCdPnoS7uzs+/PBDODo6onnz5vjyyy91GS7pWUXyQXVR03Li815l/G5ubjAzM0Pv3r3xzjvvwNnZWRehVgp9f0aYSLKXakytVsPGxkZ8X/Tvx48fo06dOqVul5OTg/79+yM8PBzNmjWr5ChfjVqthrW1tVabjY1NidfBqdVq2Nralmndqqo8433WtWvXMH78eHz++eeVGZ7kyjPevLw8jBs3Dl9//TWMjAzze2l5xvvHH39ApVIhODgYt27dwtmzZ9G1a1c0atQInTp10lXIpEevmg+qk5qWE5/3KuO/ceMGcnJysGnTJshkssoOsVLp+zPCMD9pJNKvXz/IZLJSXwCgUCiQkZEhblP0bysrq1L3m5eXh7fffht16tSp0n+gCoUCmZmZWm0ZGRklju35n8OL1q2qyjPeIn/88Qe6dOmC999/HwMGDKjsECVVnvEuXrwYAQEBeP3113UVnuTKM14LCwsYGxsjMjISZmZm8Pf3R2hoKHbt2qWrcEnPXiUfVDc1LSc+71V/B8zNzREWFob58+cb5M24RfT9GVGjC9CtW7dCeHodbIkvAFAqlUhOTha3SU5OhqOjY6lHP/Py8hAaGgpjY2Ns2rQJJiZV9yBzkyZNIJPJcO7cObEtOTlZ63B8ked/DoIg4OzZsyWuW1WVZ7wAcOvWLXTu3BmjR4/G1KlTdRWmZMoz3sTERHz77beoW7cu6tati6NHj+L999/HO++8o8uQK6Q84/Xx8dFlaFQFlTcfVEc1LSc+r6K/A3l5ebh27VplhVfp9P4ZUaErSGuA1atXCx4eHsK1a9eEBw8eCB06dCj1Lvi8vDwhJCRE6N69u8HcTTt48GAhJCREyMzMFFJSUoQ6deq88C74DRs2iHeEuri4GMw4i5R1vH/++afQqFEjISoqSg9RSqes433w4IFw584d8fXGG28IH3/8scHdpFPW8ebn5wuNGzcWPvjgAyE/P1/47bffBBsbGyEpKUkPUZO+lPX3RRAEIScnR3jy5Ing6uoqJCQkCE+ePBEKCwt1HLH0alpOfF5Zx79v3z7h5MmTQn5+vpCVlSXMnTtXUCgUBn0TkiDo9zOCBehLaDQaYdasWYK9vb1Qu3ZtYcSIEVpFV2BgoLBgwQJBEAThwIEDAgBBLpcLlpaW4is2NlZf4b9Uenq60K9fP8HS0lKoW7eu8Nlnn4nLLC0thYMHD4rvf/75Z8HLy0swNzcX/Pz8hNOnT+s+4Aoq63ijoqIEAFr/j5aWlsKNGzf0FPmrKc//77MM8S54QSjfeC9cuCC0b99esLCwEBo1aiSsXbtWDxGTPpXn98XNzU0AoPX6+eefdR+0xGpaTnxeWce/fft2wdPTU7C0tBTs7OyEf/7zn8KhQ4f0FLV09PkZIROE/z/XTERERESkAzX6GlAiIiIi0j0WoERERESkUyxAiYiIiEinWIASERERkU6xACUiIiIinWIBSkREREQ6xQKUiIiIiHSKBSgRERER6RQLUKJKkpeXh5CQEFy4cKHUdQ4dOgQ/P79y7dfPzw+HDh2qUGwZGRno3r077t27V6H9EFH1VJb8ZQiY66ouFqBUYVFRUfDz84Ofnx/atGmD3r17IyYmBgUFBfoOTa++++47uLu7o3nz5pXWx+3bt+Hn54crV66Uaztra2v06tULq1atqqTIiKi8inLpxx9/XGzZnDlz4OfnhyVLlhRblpCQgFatWuHzzz8vtmznzp1ifn721b179xfGoov8pQvMdVUXC1CSRIcOHbBnzx5s374d7777LtavX48NGzZUWn9VpbjNz88vsV0QBMTFxSE4OFjHEZVdUFAQ9uzZg8ePH+s7FCL6f3Xq1MHevXuRl5cntqnVavz888+oU6dOidskJCRg6NCh+OGHH1BYWFhsubW1Nfbs2aP12rRpU6kxGEL+Kg/muqrJRN8BUPVgamoKBwcHAECPHj3w66+/4uDBg3jnnXeQm5uLFStW4Mcff0RWVhYaN26MKVOmwNvbGwCQnp6OxYsXIzk5GZmZmXBzc8Po0aMREBAg7j8oKAh9+vRBamoqkpKS0LNnT0ydOhWffvop9u/fj8ePH8PR0RGDBg3CwIEDATw9OhgdHY1ffvkFJiYm6NChAyIiIlC7dm0AT482ZGdno1mzZvj2228BAH369MF7771X6jiLtmncuDG2bt0Ka2trbNmypdh6Fy5cwJ9//ol27dpptR86dAiffvop7t27B19fX3Tu3LnYtgcOHMDq1atx/fp1ODk5oU+fPhg6dCiMjIp/Xyz6gCga8+uvv47Vq1dDpVJhxYoVuHjxIgoLC+Hp6Ylp06bBw8ND3Nbd3R1OTk5ISkpCr169Sh0zEemOl5cXrl27hoMHD6JLly4AgL1796JZs2Yl5oCbN2/i8uXLWL58OQ4dOoQjR46gY8eOxdYrys9lUVL+OnXqFMaMGYNly5Zh6dKl+OOPP9CyZUssWLAAx48fx/Lly5GRkYHAwEBERETA2NgYACTL/2+//TauXbuGn3/+GXZ2dpg0aZKYPzMzM/Hxxx/j+PHjyMnJQd26dTF27Fjx58dcVzXxCChVCjMzM/HoYHR0NM6dO4dFixbh22+/Rdu2bTFu3Djxmpzc3Fx4eXlhyZIl2Lx5M3r06IH3338fqampWvv8+uuv0bx5c2zcuBH/+te/sGnTJhw8eBAff/wxvvvuO8yZMweOjo4AAI1Gg2nTpiErKwtr1qzB0qVLcenSJcydO1drnydOnEB6ejpiYmIwbdo0rFu3DseOHXvh2I4fP47bt29j5cqVWLRoUYnrnD59Gu7u7jA3Nxfb7ty5g+nTpyMgIADffPMNunfvjpUrVxbbLioqCv/617+wZcsWREREYPPmzdi8eXOJ/Xz11VcAgFWrVmHPnj2Ijo4GAGRlZSEoKAhffvkl1qxZA3t7e0ydOlXrqAoANG/eHKdPn37heIlIt4KCgpCQkCC+37lzJ4KCgkpcNyEhAQEBATA3N0dgYKDWdq+qpPxVJCYmBjNnzsTq1atx9epVTJ8+HXv27MEnn3yCBQsWYOfOndi7d6+4vlT5PzY2Fq+//jo2btyIgIAAREZG4tGjRwCAlStXIjU1FcuWLcOWLVswbdo0KBQKre2Z66oeHgElSQmCgHPnzmH37t0ICQnB3bt3sXPnTvzwww+wt7cHAISHh+Pw4cPYvXs3hg8fjrp162LIkCHiPoYOHYpDhw5h3759CA8PF9tbt26NwYMHi+/v3r0LV1dXtGjRAjKZDK+99pq47OTJk7h27Rp27dolFqVz5sxBWFgYbt68CVdXVwCAra0tpk6dCplMBnd3d2zZsgWnTp3CG2+8UeoYLS0tMWvWLJiYlP7nc/fuXbHfIkXXVE2aNAnA02/lFy5cQFxcnLhOTEwMRowYgZ49ewIA6tevjxEjRmDz5s0YNGhQsX5sbW0BPD3F9uwRjtatW2utN2fOHHTq1Annz5+Hr6+v2O7g4FDu60eJqHL17NkTq1atwv3795GVlYUrV66ga9eu+P7777XWKywsxPfff4/IyEgAT88+xcTE4OHDh7CzsxPXy8jIQIcOHbS2bdu2bYnXmgIl568i48aNg4+Pjxjn+vXrsXfvXtjY2KBRo0bw9/fHqVOn0KNHD0nzf4cOHdC7d28AwNixY/Htt9/i/PnzaNu2Le7evYumTZvC09MTAFCvXr1icTPXVT0sQEkSSUlJ6NChAwoLC1FYWIjAwECMHj0av/76KwoLC8XEUSQvLw+NGzcG8DSJrl27FomJibh//z7y8/ORl5cHNzc3rW2evxi+Z8+eGDduHN5++220a9cOHTt2hL+/PwAgNTUVzs7OWknUy8sLpqamSE1NFQvQhg0bQiaTievY29vj4cOHLxxr48aNX1h8AkBOTg5q1aql1Xb9+nUolUqtNh8fH60C9NKlSzhz5gxiYmLENo1GA41G88L+nvfgwQOsWLECv/32Gx4+fAiNRoP8/HzcvXtXaz0zMzPk5OSUa99EVLkcHBzQunVrfP/998jMzETnzp1hYWFRbL1jx45Bo9GIee+1116Dl5cXfvjhB62irnbt2uLZkiJyubzU/kvKX0WK8jYA2NnZwd7eHjY2NlptRTn0ypUrkuX/Ro0aif82NzeHlZWV2E/fvn0xY8YMXLx4EW+88QY6d+4MLy8vre2Z66oeFqAkidatWyMiIkK8FrSoQMvOzoaJiQm++eYbrUIPeHokEQA2bNggnjZp2LAh5HI55s+fX+wGn+cTpqenJxISEnDkyBGcOHEC06ZNQ7du3TB79uwyx/18ISmTySAIwgu3Kem01PNsbGxw/fr1MsdR5MmTJxg7diw6depU7m2fFRUVhczMTERERKBu3bowNTXFoEGDiv1MMzMzxaOoRFR1BAUFYfny5cjOzsb8+fNLXGfHjh14+PAh2rZtK7ZpNBpkZmZqFaAymQwuLi5l7vtF+evZnCmTyV6YQ6XM/yV96S/qp0OHDkhISMDhw4dx/PhxhIeHIzw8HCNHjhTXZa6reliAkiTMzc1LTHBNmjRBQUEBHj16JJ62ed6ZM2cQEBCAwMBAAE/vcL9165bWKfXSWFlZITAwEIGBgXjjjTcwZ84czJw5Ew0aNMDt27dx//598SjouXPnkJ+fjwYNGlRgpGXTtGlTbNu2TavN3d0dR48e1WpLSUkptt2NGzfK/GFhamoKAMWOkJ45cwYzZ84UP5iuX79e4rf/1NTUcs9DSkSVr2PHjli4cCEsLS3RsmXLYssfPXqEQ4cO4aOPPoK7u7vYnpubi5EjR0KlUhU741JWJeWvV1HZ+f9Z9vb2CAkJQUhICNavX49t27ZpFaDMdVUPC1CqVO7u7ujatSvmzJmDKVOmoHHjxkhPT8exY8fw+uuvo2XLlnBxccHPP/+MlJQUWFhYYP369VCr1S/d9zfffANHR0c0adIEwNO7x11dXWFkZIRWrVqhYcOGmD17NqZMmYLc3FwsXLgQnTp1Ek+/VyY/Pz+o1Wpcv35d/HDo27cvYmNjsWzZMgQHByMlJQU//fST1nYjR47E1KlTUadOHfEOz4sXL+L27dtaybSIra0tzMzMcPToUTg4OKBWrVpQKBRwcXHBDz/8gGbNmiEjIwNLly4Vi9Uiubm5uHDhAsaPH185PwQiemUmJibYvn07ZDJZsaOHAPD999/D3t4eb775ZrHlfn5+2LFjh1YBmpaWVmwfpd0ZX1L+ehWVmf+ftWrVKjRr1gweHh548uQJjh8/rnUKn7muamIBSpVu3rx5iImJwSeffIL79+/Dzs4OPj4+4kTII0eOxJ9//on33nsPFhYWCA0NLXYTTUnkcjnWr1+PP/74A8bGxvD29sbixYsBAEZGRvjkk08QHR2N8PBwGBsbo2PHjoiIiKjUsRaxsbFBQEAA9uzZgzFjxgAAnJ2dsWjRIixZsgTffvstfH198e6772rdSd+uXTt8+umnWLNmDdatWwdTU1M0bNgQoaGhJfZjYmKCiIgIxMTEYMWKFfD19cXq1asxZ84cLFy4EIMHD8Zrr72GSZMmFZsB4NChQ6hbt26pRyaISL+ev5P7WTt37sQ///nPEovTzp07Y+nSpZg2bRoAiNMjPe/IkSMwMzMr1l5S/npVlZX/n2VsbIxly5bhzp07MDc3h5+fH/7973+Ly5nrqiaZ8LIL3ojolVy6dAkTJ07E9u3by3TdqK6NHDkSoaGhJX4wEVHNVtXzV3kw11VNnAeUqJI0adIE7733Hu7cuaPvUIrJyMhAx44dX/o4PiKqmapy/ioP5rqqi0dAiYiIiEineASUiIiIiHSKBSgRERER6RQLUCIiIiLSKRagRERERKRTLECJiIiISKdYgBIRERGRTrEAJSIiIiKdYgFKRERERDrFApSIiIiIdIoFKBERERHp1P8Bpxv8UNO4TKwAAAAASUVORK5CYII=", + "text/plain": [ + "
" ] + }, + "metadata": {}, + "output_type": "display_data" }, { - "cell_type": "code", - "execution_count": 31, - "id": "bio-extended-code", - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n", - "\u001b[0m" - ] - }, - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWwAAAFsCAYAAADon4O5AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAQ6wAAEOsBUJTofAAAb19JREFUeJzt3Xd4VMX6wPHvppfdJKQXktBDr6FXgSBFQFCqBSyACldEL4qKAl4uIlhAQUEU/AmoFxQRVJAiVekQJLQESEJ6SM9ms31+f0RWYwIky0IK83mePE92zuycd89u3pydM2dGIYQQSJIkSdWeXVUHIEmSJFWMTNiSJEk1hEzYkiRJNYRM2JIkSTWETNiSJEk1hEzYkiRJNYRM2JIkSTWETNiSJEk1hEzYkiRJNYRM2JIkSTWETNg1yLZt2xg8eDB+fn44OjoSEBDAkCFD+PrrrzGbzVUd3m1LSEhAoVDw7bffVnUo94wvvvgChUJBVlYWUP57UK9ePaZNm3ZH44iOjmbu3LloNJqbxnevkwm7hnjttdcYPHgwLi4uLFu2jN27d7Ns2TK8vLx49NFH2blzZ1WHeNuCgoI4dOgQffv2repQ7llV9R5ER0czb968Mgl7yJAhHDp0CC8vr7saT3XlUNUBSLf2008/8fbbbzNnzhzmzp1batuoUaOYPn06jo6OVROcDTk7O9OlS5eqDuOeZsv3oLi4GFdX19tqw8/PDz8/P5vEUxvIM+wa4P333ycoKIjZs2eXu71Tp060a9euVNlPP/1E586dcXV1xc/Pj2effZaioiLL9r1796JQKNi5cyfjx49HpVIRHh7OokWLyrR//YzL3d0dT09Pxo8fT2Zm5k1jPnDgAAqFgsuXL1vKhg4dikKh4OzZs5aycePGMWTIEKD8r+NbtmwhMjISpVKJl5cXkZGR/Pzzz6X29cUXX9C6dWtcXFwICQnh9ddfx2Qy3TS+uXPnolQqOXXqFF27dsXV1ZX27dtz6tQptFotzz77LHXq1KFu3bosWbLEqmMya9YsWrVqhVKpJCQkhHHjxpGWllaqTp8+fXjggQf49ttviYiIQKlU0rdv31LH7Uby8vL417/+Rd26dXF2dqZ+/fq8+uqrlu0//fQTUVFR+Pv74+HhQefOndm+fftN27xZt9TixYsJCQnBzc2N4cOHl3ot15/3xRdfMGnSJHx8fOjUqVOF4vjiiy944okngJIErVAoqFevnmXbP7tEcnJyePLJJ/H19cXV1ZVu3bqxf/9+mx3Xak1I1ZrBYBDOzs7ikUceqfBzNm7cKOzs7MRTTz0ltm3bJlavXi38/f3FmDFjLHX27NkjANGgQQMxZ84csXPnTjF16lQBiG3btlnq/f7778LJyUk8+OCDYuvWreKbb74RjRo1El26dLlpDFqtVri4uIjVq1cLIYQwmUzCy8tLuLi4iI8//thSLzg4WCxcuFAIIUR8fLwAxMaNG4UQQly6dEk4OjqKxx9/XOzYsUNs375dvPPOO2LdunWW57/33nvC3t5e/Pvf/xY7duwQS5cuFUqlUrzyyis3jW/OnDnCyclJtGrVSqxevVr89NNPomXLlqJevXrikUceETNmzBA7duywHJPffvut0sfkiSeeEF999ZXYu3ev2Lhxo+jSpYto3LixMBgMljq9e/cWISEholOnTmLTpk1i48aNIjQ0tELHt127dqJOnTpiyZIlYvfu3eKLL74QTz/9tKXORx99JJYuXSq2b98uduzYIWbMmCEUCoXYs2ePpc6aNWsEIK5du1bueyCEEOHh4SI4OFh07dpV/PDDD+KLL74QAQEBpWK8/rzAwEDx1FNPiV9++cXyObpVHJmZmWL27NkCENu3bxeHDh0SJ0+eLDc+o9EoOnXqJHx9fcVnn30mtm7dKqKiooSTk5M4fvz4bR/X6k4m7GouPT1dAGLWrFmlys1mszAYDJYfk8lkKQ8PDxfjxo0rVX/btm1CoVCImJgYIcRfCXvmzJml2qxXr5546qmnLGW9evUS3bp1E2az2VJ29uxZoVAoxE8//XTT2Hv16iUmTpwohBDi1KlTwtHRUUyZMsXyjyMuLk4A4vfffxdClE0WGzduFIAoKCgot/2CggKhVCrFq6++Wqr8k08+Ea6uriIrK+uGsc2ZM0cA4ueff7aUbd26VQCl/rEZjUbh7+8vXnjhhds6JkajUSQnJwtA/PLLL5by3r17C3d3d5GZmWkpu56kkpKSbhj/p59+WurY3YrJZBIGg0EMGDCg1GejoglbpVKJvLw8S9nu3bstCfbvzxs4cKBN4rhR+Q8//FBqv0IIodfrRVhYmBg5cqSlzNrjWt3JLpEaQqFQlHr83Xff4ejoaPl5/vnnAYiNjSUxMZHRo0djNBotP71798bOzo7jx4+XamfAgAGl9tGsWTOSk5MB0Gg0/Pbbb4waNQqTyWRpq0mTJoSGhnLs2DGAUtuMRiPizzUxevXqZfmqun//fiIjIxk0aFCpMjc3NyIjI8t9za1bt8be3p7x48ezdetW8vPzS23//fffUavVjBo1qtT++/fvT3FxMTExMTc9pnZ2dvTr18/yuEmTJgD079/fUmZvb0/Dhg1JSkqq1DGBklE93bp1w9PTEwcHB+rWrWt5j/6ubdu2pfppmzdvDmB5H8qze/dumjVrRteuXW9YJzk5mQkTJhASEoKDgwOOjo7s2LGjzP4r4r777sPT09PyuG/fvnh7e3PkyJFS9a53b92pOA4cOICHhwf333+/pczR0ZGRI0dy8ODBUnWtOa7VnUzY1ZyPjw/Ozs5lPmT9+vXj2LFjHDt2jKCgIEv59b6+ESNGlErobm5umEwmS+K57p9X352cnNBqtQDk5uZiMpmYMWNGqbYcHR25evWqpa1+/fqV2rZv3z4AevfuzZUrV0hJSWH//v306tWLnj17kp6eTlxcHPv376dLly43vGDapEkTfvzxR/Lz8xkxYgR+fn4MGzaMq1evlnqt7du3L7X/xo0bA5R5rf/k6uqKk5NTqdduq2Ny7Ngxhg0bRnBwMGvXruXQoUMcPnwYwNLWzd6D8ur9XXZ2NsHBwTfcbjabGTZsGAcPHuStt95iz549HDt2jEGDBt203Rvx9/cvt+yfffIBAQF3NI7c3NxyYwkICCAnJ6dUmTXHtbqTo0SqOQcHB7p3787u3bsxmUzY29sDUKdOHcuZ6d+Tjre3NwDLli2jc+fOZdq72R/5P3l5eaFQKHjttdd48MEHy2z39fUFYOXKlRQWFlrKIyIiAOjatSuOjo7s37+fAwcO8OSTT+Lt7U2LFi3Yt28f+/fv5/HHH79pDAMHDmTgwIEUFBSwfft2ZsyYwRNPPMHu3bstr3XTpk2EhoaWeW79+vUr/ForqqLH5Pvvv8fT05MNGzZgZ1dyXpSYmGizOHx8fPjjjz9uuP3SpUucOnWKzZs3M3z4cEt5cXGxVfsr7yJzZmZmqZMFKPtN0NZxeHt7lxtLRkaG5fNQm8mEXQO8+OKLPPDAAyxYsIA33njjpnWbNm1K3bp1uXLlClOnTr2t/bq7u9O1a1fOnz/P/Pnzb1jveoIu7/nt27dn5cqVZGdn06NHD6DkzHv9+vXEx8fTq1evCsXi4eHB6NGjOXLkCF9//TVQ8g/Bzc2N5ORkRowYUclXZ52KHpPi4mIcHR1LJbD169fbLI7+/fvzv//9jyNHjpT7j/l6Qvz7P/PExER+++03S9dPZezZs4f8/HxLt8ivv/5KTk5Oufu2Jo6Knv326NGDxYsXs2PHDkt3ntFo5Pvvv7d8vmozmbBrgCFDhjBr1izefPNNoqOjGTNmDEFBQeTn53PgwAHS09NRqVRAyRnO+++/z/jx4ykqKmLIkCG4u7uTmJjITz/9xIIFCyr1B7t48WL69u3LmDFjGDt2LHXq1CE5OZmdO3fyxBNP0KdPn5s+v1evXixevJj27dvj4eFhKVu+fDmOjo437YNduXIlhw4dYuDAgQQFBREfH8+6dessf6heXl689dZbvPzyyyQnJ9OnTx/s7e25cuUKP/zwA9999x1ubm4Vfq0VVZFjEhUVxZIlS/jXv/7FiBEjOHToEGvXrrVZDI899hgff/wxQ4YMYc6cObRs2dLS9fTpp59a/nHPmjULk8mEWq1mzpw5hISEWLU/lUrFoEGDmDVrFnl5ebzyyit06tSpVF9yeSoaR7NmzQBYvnw5Dz74IG5ubrRq1apMe0OGDKFTp048+uijLFy4kICAAD766CPS0tJ47bXXrHptNUpVX/WUKu7HH38UgwYNEj4+PsLBwUH4+/uLQYMGifXr11tGiVy3Y8cOy5Vyd3d30aJFC/HSSy9ZrvRfHyVy7NixUs8bPny46N27d6myY8eOicGDBwtPT0/h6uoqGjduLJ555pkKXW3/8ccfBVBqlEVaWpoARNeuXUvV/ecIhd9//10MGTJEBAUFCScnJxEWFiamT59eZtTI119/LTp27ChcXV2Fh4eHaNeunXjjjTdKDZ/7pzlz5gh3d/eb7v+63r17iyFDhlT6mLzzzjuibt26ws3NTURFRYnY2FgBiMWLF9+07VOnTgmg1PC78uTk5Ihnn31WBAYGCicnJ9GgQQPx+uuvW7YfPXpUdOzYUbi4uIjGjRuL//u//xMTJkwQLVq0sNSp6CiRqVOnirffflsEBQUJFxcXMXToUJGSknLLY1fROIQQYu7cuaJu3brCzs5OhIeHlxufEEJkZWWJiRMnCm9vb+Hs7Cy6du0q9u7dW6qt2zmu1ZlCiD8v6UuSJEnVmhwlIkmSVEPIhC1JklRDyIQtSZJUQ8iELUmSVEPIhC1JklRDyIQtSZJUQ8iELUmSVEPIhF2OoUOHVnUIkiRJZVR5wl62bBmRkZE4OzszduzYG9Zbv349SqXS8uPu7o5CoWDTpk1AyQoqdnZ2peosWLDAqphq8mxekiTVXlU+l0hwcDCzZ89m165dN10Z+ZFHHuGRRx6xPN62bRtjx45l4MCBljJ/f3/S09PvaLySJElVpcoT9siRI4GSVZMrs5T96tWrGTNmzB2Z3EeSJKk6qvIuEWtkZ2ezZcsWy8Kdfy8PDAwkPDycKVOmlJnQvDyLFy/G39+/1I9Go7lToUuSJFmtRibs9evX07Bhw1JTczZt2pTo6GhSU1M5ePCgZVmiW5k5cyaZmZmlfuRZuyRJ1VGVd4lYY82aNWXOrgMDAwkMDAQgNDSUZcuW0ahRIzQajUzAkiTVCjXuDPvUqVPExMTw2GOP3bSenZ0domRV+LsUmSRJ0p1V5QnbaDSi1WoxGo2YzWa0Wi0Gg+GG9desWcOgQYMsZ9PX7dmzh4SEBIQQpKen8/zzzzNgwADc3d3v9EuQJEm6K6o8Yc+fPx9XV1f++9//snHjRlxdXZk0aRIASqWSAwcOWOrq9Xq++uornnzyyTLtnDp1ip49e+Lu7k6HDh3w9fW16ZJMkiRJVU2uOFOOqKgodu7cWdVhSJIklVLlZ9iSJEm1QX6xgdOp+WSpdXdsHzJhS5Ik3SaTWfDz+XROX7zCnsvZmMx3puNCJmxJkiQrGE1m0gu06AwmdlzMIPXscYrjjiN0RSju0D5r5DhsSZKkqrbnUjZZRXpcHSDn4klU2iwMdVsxoFU97OzuTMqWCVuSJMkKKfnFnM1Q450Ti48+C99mnejQvDFero53bJ8yYUuSJFnBx80Jjd5IkZ0vqpAQ2jZrRLj3nb2rWiZsSZKkSsgu0mOPGbfcKzTy9uWqvR3uPh537ELj38mELUmSVEEXMgo5eTUH7eUToMljzKAh5OKKvZ2CRr53/q5qmbAlSZIqKKtQQ3HcMUyaArr06UdwUADBd3H/clifJElSBZjNZq7FHEJoC4no3JtWDcPuegzyDFuSJKkcJrPg17gscjR6OoZ50cDHnaYREfj5+eHr61slMVU4YQ8bNgyFomJjC3/44QerA5IkSaoOstU6zmcU4oyR6DOpNOjTkWbNmlVpTBVO2I8++qjld7VazZdffkm7du1o1aoVAGfOnOHUqVMVWuVFkiSpuruaryUzrxD/a2fIszej1bbCxcWlSmOqcMIePXq05ffXXnuNp59+ulQSB1i3bh0xMTG2i06SJOkuyy82YDCZKS4upl7eWYSdiV5Rg6o8WYOVFx0PHjxI7969y5T36tWL33777baDkiRJqgrX1Dq2Xcjkl7MpxB/biz1mOvYZQKMQ/6oODbAyYatUqlILC1x38OBBPDw8bjsoSZKkqlCoMyIEKOwdUfn489CDw2hbP6iqw7KwapTIpEmTePvttzl58iQtW7YEICYmhgMHDvDqq6/aNEBJkqS7xcfBSJBDMU4qb9q36YOLo31Vh1SKVQn7wQcfpH79+mzYsMGyMku9evX49NNPadOmjU0DlCRJupO0BhNn0wuxNxRz9tCvuLm58eCDD1Z4VNzdZPU47DZt2sjkLElSjffFsatcSblGo8LzqNxcuP/++6tlsobbuNMxKyuLtWvXsnDhQvLy8oCShXCTk5NtFZskSdIdU1Bs4MtjV9kefYWQnBg0wp77BgzCze3Ozrh3O6xK2DExMTz00EP8+uuvbN68GbVaDcCJEydYvny5TQOUJEmytbNpBTy14TRv7YwlIU9LtnDFqVFHAr2r96AJqxL2kiVLmDBhAmvWrMHJyclS3qVLF/744w+bBSdJkmRL19Q6fjiTxvLf4snLycbOZMTs4I4+tA1PdmtY1eHdklV92LGxscybN69Mube3N7m5ubcdlCRJ0p3wy4VMlv+WgH1xAWPqpHHV1RPXevV4tW/jajcipDxWJWx3d3dycnIICQkpVR4bG4ufn59NApMkSbKlPbHX+Pi3BAyFOYz3zSQfV54Y1p/mwXWqOrQKs6pLZMCAAXz00Ufk5uZarqbGxMSwdOlSBg0aZNMAJUmSbtf3f6TxyPpT6AuymOCbSbZwxaVRxxqVrMHKhD116lTCwsIYNGgQGo2G0aNH8+STT9KsWTOefvppW8coSZJkte/PpLBgVywFOgOuChPJBlfadO3Nv3o1qurQKs2qLhEnJydmz57N008/zaVLlyguLqZJkyaEh4fbOj5JkiSrpeQXc/BKLj4OBgJVThQ6u9Oma31GtA2t6tCsYlXCXrVqFY899hiBgYEEBgZayrVaLWvXrmXSpEk2C1CSJKkyNHojv8XnUqQ3Yq8AReE1hrvEkxfcmH8N7YHSueau22JVl8iqVavQaDRlyrVaLatWrapUW8uWLSMyMhJnZ2fGjh1707oKhQJ3d3eUSiVKpbJMf/m3335Lw4YNcXNzo1+/fiQmJlYqFkmSar74nGKS8or5NS6LvSfP0VhzCTz8mTqkW41O1mBlwhZClHvr5uXLlys9W19wcDCzZ8+u8Fn5iRMnUKvVqNVqtm3bZik/f/48EydO5JNPPiE7O5vWrVuXmsNbkqR7g5+7I3FZReiyU2mivUyxmx/hrTuhdHas6tBuW6X+3fTv3x+FQoFCoWDUqFGlkrbZbEatVjNy5MhKBXC9fnR0NFlZWZV67t+tW7eOgQMHMmDAAADeeust/Pz8OHv2LC1atLC6XUmSao70fC1bz2VgFoLm7jpM9kGMHdgPf5UrdnbVc36QyqhUwn7hhRcQQvDWW2/x9NNPo1Qq/2rIwYHg4GBat25t8yD/rm/fvphMJiIjI1m0aJElGcfExNCxY0dLPZVKRcOGDYmJiblpwl68eDGLFy8uVda4ceM7E7wkSXeE2Sw4cCWb1Uev4mTSU4QjYfVb0qOBD4Ge1XdukMqqVMJ+4IEHgJJujDZt2uDgcHf7g/bu3UvXrl3R6XS88847DBgwgPPnz+Ph4YFarcbLy6tUfS8vLwoLC2/a5syZM5k5c2apsqioKFuHLknSHVCoNVKkN6I1mPj0cCLkJNPFOZMUv7aMaRuCr9K5qkO0Kav6sHNycspdCmzfvn3s2rXrtoO6kd69e+Pk5IRKpWL+/Pk4ODjw+++/A6BUKsnPzy9VPz8/H5VKdcfikSSpauiNZo4n5fH1qRR2xWbx1akUPIpS6e2SQZqjHxN6NK11yRqsTNgrV67E2bnswXBzc+PTTz+97aAqys7ODiEEAC1btiQ6OtqyTa1Wc/nyZcuKOJIk1R5Hr+Zy/Gou608mMe+Xi1y5cJZ2ijQyXIO5v08PGvvXzhM1qxJ2WloaYWFhZcpDQkJISUmpVFtGoxGtVovRaMRsNqPVajEYDGXqnT17lpMnT2I0GtFoNMydO5fi4mK6du0KwKOPPsq2bdvYtWsXWq2WOXPm0Lp1a3nBUZJqIQFcyiriQqaaPE0xDe3yyHYPZWT/XnRv4FvV4d0xViVsLy8vrly5Uqb88uXLle6CmD9/Pq6urvz3v/9l48aNuLq6Wob4KZVKy2K/mZmZjB8/Hk9PT8LCwjh8+DC//PKLpd+6WbNmrFmzhsmTJ+Pt7c2pU6fYsGGDNS9PkqRq7Fx6IfkaA1qDkVCVA67OzmzV1yOqZxdaBFXv+axvl0Jc71OohCVLlrBnzx7mzZtH27ZtgZJheXPnzqV3797MmDHD1nHeVVFRUZa1KiVJqj5yNHo2n0ln58VMmpjT8DHmoq/fiSndGqBydbp1AzWcVcM8nnvuOfLy8pg8ebJlpIjJZGLw4MFMnTrVpgFKkiQBqHVG1h2/ykcH4+nllkOIWz6K4KZMuy+iVoyxrgirzrCvS0lJsXSNNGrUiKCgIJsFVpXkGbYkVS/x2Ro2RCez6nAiPV2u0VWl5px9XV4Y0Yd63rVnnPWt3NZA6pCQkDKLGEiSJNlSscHE1nPpxGVp8FLoiXQvYpcmgEf7t7+nkjVUImEvXbqUyZMn4+rqytKlS29ad/r06bcdmCRJkt5o5oczafx8Lh2TEHRuEoqdb0PebBJMy+DafYGxPBVO2OfOncNoNFp+v5HyJoWSJEmyRoHWwM/n02ltTsIsoHV4Nx7vWDPnsraFCifslStXlvu7JEmSLWUU6ricVURYHVdOJeUSoU3Axy6fS24NiQzzrOrwqlTNnhxWkqRaxWwW/GfHRX69lI1CmBnrk4W/KCDbrwUj2jXDx80Zs1ncM6NC/qnCCXvKlCkV7u5YsWKF1QFJknTvSsgp4oezGeQXG4hUFuErCtDVbcMjXVpyKiWfq7nFhNVxpXt976oOtUpUOGE3b97c8rvRaGTLli0EBQXRqlUroGR609TUVIYNG2b7KCVJqtWEEFzO1nAsMRdvd3t0RhOxJhXXghowpUcLVC4OFBvMAOQVl5264l5R4YT995Ef77zzDiNHjiwzGmTp0qUUFRXZLjpJku4JFzLVHEvMZWN0MsPd0sj28uG5wV1pHuhh+Wbfvq4naQVamgfUzomdKsKquUS2b9/OiBEjypSPGDGCHTt23HZQkiTdW9IKdFzIyKeziCfQvpgwf29aBHmW6oaN8FfSp5Ev/qraN21qRVl10dHR0ZEzZ86UmbHvzJkzODrW/HXTJEm6s1Lzi/khJh29SeDn7sSJxCzC8s/jqNCS6deSAS0bVnWI1ZJVCXvMmDEsWLCAixcvllqia/PmzUycONGW8UmSVMuYzIKPDiZwIUNNbrGecC83gtSXcbTTENS2B891iqjqEKstqxL2U089RUhICBs2bODHH38EoF69erz++usMHDjQpgFKklS7nErJJ7dYz+UsNfk6IypnB7zqNEDrYUf3ZvWqOrxqzepx2AMHDpTJWZKkCjOZBeczCjmZlEex1ojJqGd8nWuk2bvwRPcI6nq61splvWzJ6oSt1Wo5ePAgycnJPPTQQ6hUKhITE/H09CyzGK4kSdLa40kcS8ojOiUPnVbLOM9U3O3NJAtBI18lSmd5H9+tWHWEEhISmDp1KmazmZycHPr3749KpWLz5s3k5eUxZ84cW8cpSVINVqQzEp1aQKZaS0ZOAU8HXMNRITjr3ow+Tepif4/euVhZVg3re++99+jduzc//fRTqcV4e/fuzcmTJ20WnCRJtYOzgx3uTnacz1DzqG8WdgiueDZn6n3NGdEqEFdH+6oOsUawKmGfOXOGcePGYWdX+un+/v5kZWXZJDBJkmouk1lwKauI9AItULJgboHWgI+7M+ft65IX1JZp97WgXV0vvN1q/9JetmJVl4iDgwMajaZMeVJSEp6e9/ZsWpIkweGEHM5lqNGbzbQIUJGTk0cT/VUMfnWp6+XH4Gb+RPgrqzrMGseqhN2rVy8+//xzFixYAJTMgZ2ZmcmyZcvo06ePLeOTJKmGKdQa2Xclh8vZRWQW6jiXkE6L4ou4OjkxKtKf+5oG37Oz7d0uq7pEZsyYgVqt5v7770en0/Hss88yfPhwnJ2dmTZtmq1jlCSphsgs1HEmrQA/dycCVM542+torrkA9o70HziIfs1DZLK+DVadYatUKj7++GOio6OJi4tDo9EQERFB586d5YozknQPKtIZuZxdxLen0ziZnIef0pme4SoaZsRjdHahSadehPl6VXWYNV6lE7bRaKRfv36sXr2atm3b0rZt2zsQliRJNYXJLNgRe43YTDUbolMwmAUao4nx7esS1q0L9erVKzWaTLJepRO2g4MDnp6eCCHuRDySJNUwRrMZndGMo50ChUJBuLOepvZamgcqCfH0q+rwahWr+rAnTpzI8uXLUavVto5HkqQaxtnBnoY+bpy/pqaJu4GHVCl09dRRx8WBY1fzOJ9RWNUh1hpW9WFv3ryZhIQE7r//fkJCQnBxcSm1/csvv7RJcJIkVU9Zah1xWUV4ujhSqDVwMiUfY2EO9zsnoXFQMmLAAM6kq7mSXTL818vVkSAPl1u0Kt2KVQm7R48e9OjRw9axSJJUQ/yekEuR3sSplHQi/NxJTE2ns+kKeXZKfJt1ws/DnWvFJWfWCgW4OFj1ZV76B6sS9uTJk20WwLJly/jiiy84c+YMI0aM4Jtvvim33uHDh5kzZw7Hjx8HoEuXLixZsoTGjRsDsHfvXvr27Yubm5vlOa+99hqvvfaazWKVpHudWmfkZHI+l7KLEEJgpwAXR3sahQQQ7GtmXOf2eLi6oFAoaB3kQR1XR9ydHKgj72a0iduaHuvIkSPEx8cD0KBBAzp16lTpNoKDg5k9eza7du266W3tubm5PPnkk2zYsAFXV1feeOMNhg0bxvnz5y11/P39SU9Pr/wLkSSpQs6kFZCSryUxR4OPmxPN3IpponSnUbO61HFrUKquQqEgrI7bDVqSrGFVwk5KSuLll18mISGBoKAgANLS0ggPD2fRokVllg67mZEjRwIQHR1904Q9aNCgUo9feuklFi1aRHZ2Nj4+Pla8CkmSKsNoMnM8KZ9f465hFIIWrsXoLp+m2CGCOhHhVR3ePcGqjqW33noLb29vtm7dyqZNm9i0aRNbt27F19eX+fPn2zrGcu3bt4/AwMBSyTo7O5vAwEDCw8OZMmUKOTk5t2xn8eLF+Pv7l/opb54USbrXZRXpyS024OPuRBs3DVw9TVh4fXk96y6yKmGfO3eOGTNm4Ovraynz9fVl+vTpnD171mbB3ciVK1eYNm0aS5YssZQ1bdqU6OhoUlNTLQsrTJgw4ZZtzZw5k8zMzFI/f+8HlySphI+7EyGezgSacwktjKN+w4YM6N+3zKyd0p1jVZdIaGgo+fn5ZcoLCgoICQm57aBuJikpif79+/PKK68wZswYS3lgYCCBgYGW+JYtW0ajRo3QaDQyAUvSbbiaq+Gb6FTyNAYGN/MnKrQlSQkudOvaRU5FcZdZ9a/x+eef57333mPfvn1kZWWRlZXF3r17ee+995g+fTpardbyY0vJycn07duXyZMn8+KLL960rp2dHUIIeUemJN2GM6kFrDqcyIWMQoqzU1l3LIE4NXTtIpN1VbDqDHv69OkA/Pvf/7a8adcT44wZM0rVPXr06E3bMhqNlh+z2YxWq8Xe3h5HR8dS9VJTU7nvvvt49NFHmTVrVpl29uzZQ/369QkPDycjI4Pnn3+eAQMG4O7ubs1LlKR7WrHBxPGkPDZGp5KjMUB2Is1cr5Hp1IBMdSB5WoNceKAKWJWwV6xYYbMA5s+fz7x58yyPN27cyIQJE/jiiy9QKpVs27aNnj17smrVKi5dusTixYtZvHixpf65c+cICwvj1KlTPP7442RnZ1OnTh0GDhzIwoULbRanJN1LDifmcjW3mKR8LY1EJo1dr1HsFYbCK5Q6bo54ujjeuhHJ5hRC9hmUERUVxc6dO6s6DEm6q3I1eqJT8skqMnDxmhpXR3vc8hKxz7wE/g3xDG9Kj/rehHi5VnWo9yy5rrwkSZjMghW/J3I2oxCN3kSYV8ndih2V7jQNi6Rt23YIkKubVzGZsCVJwiwEhToj7k72FBbrcdLmUj80FBcPH9q3Dqrq8KQ/yYQtSRI5GgORoV6k5GoY6JnLtYRY6roE0jossKpDk/5GJmxJuodp9EZSCrT8GJOBytkBX3U811Ku0L17d1q0aFLV4Un/IBO2JN2jrmQV8eXxZI4n5ZJSUMwgVQ7hZBPeMpKmzZpXdXhSOWx+T+kzzzzDZ599ZvObZiRJsq19l7M4lZJP7DU1wgx2ZiPagGbkugRwObuoqsOTymHzhB0UFMSxY8d4+OGHbd20JEk2oDeaKdYbuZqnRW8y0VglaOjnjmuDtviE1ANA5Sy/fFdHNn9X5syZAyBnvJOkaiSzUMe5jEListQk5WkJr+NKoMqRYR7XsFNnETWoEw0CPMlS67CzU8i7GKupO/ZvVE64JEnVx5GrufyRVsCPZ0suLhoMRprp49EWZtCpRx8aBHgC4Kt0ruJIpZuxKmGnp6cTHR1NWloaWq2WOnXqEBERQcuWLcvMASJJUtXKKNRxIbOQLLUeDxcHnOwgKO8ien0eA++/v1ILjkhVq1IJe/v27fzvf/8jJiYGHx8ffH19cXZ2pqCggJSUFJydnRk4cCCPP/64ZSUaSZKqjkZv5P19l7hyTYO30olxbUNwUxgQ8Ul0v+9+6tatW9UhSpVQ4YQ9fvx4XFxcGDp0KIsWLcLPz6/UdoPBQExMDDt27GDChAm8/PLL9O/f3+YBS5JUMUIIkvKKiUkrJClPi1+Rlhe71yUi2AdzZH258EANVOHJn44ePVrhRXYLCgpITU2ladOmtxVcVZGTP0m1wf7L2ZxIyuOHs+mYjEZGqFIJ8XRh9MgH5VzWNVSFz7ArsyK6h4cHHh4eVgUkSdLtydHo+S0+h+/+SKNQa8RRYWKoMgWlWUv3rvfJZF2DWT1KxGQykZycTG5uLmazudS29u3b33ZgkiRVTkp+MadTC8hS63G0t8NOAU4KI/c7JaFCT9/7B1I3WM4NUpNZlbCjo6N54403yMjIKLMEl0KhuOUqM5Ik2d6hhFwuZKpRKCBY5UKXcG9Uhjw0V4z0ixpEvRCZrGs6qxL222+/TZs2bZg8eTK+vr7yK5YkVRGtwYTeZMbDxZHkvGIyCnWo9Uaa+7gQWEdJYq4DIZ2iCA0KqOpQJRuwKmGnpKTw3nvvySFBklSFCrVGfrmYicEkMJnNJOUVk6814IqRrDMHSXTxIaBRS3K1gny5BmOtYNW4no4dOxIXF2frWCRJqoS8YgNFehMXr6n5PSEXIcDZrKdF8UUwGQmt3xB7OwUBKme85BqMtYJVZ9h9+vThgw8+4MqVKzRs2BAHh9LN9OjRwybBSZJ0YzqTiYPxOWQUamnipwR9Me0NcTjYQ98Bg2gQJLsraxurFuHt2LHjjRusBRcd5ThsqTozGM18ejiRX+OukafVk1dspImfikn1DaQkXiHq/sEE+npVdZjSHWDVGfaxY8dsHYckSRX0W0I2nx1OJFOtR2s00cTHFcxmGjVvSbfItri4uFR1iNIdIu9NlaQa5mqOhnytERTQ1FPBw67xeJoLOJiQh07YV3V40h1k9Y0z+/btY+3atcTHxwPQoEEDHnvsMXr16mWz4CRJ+svJ5DyuZmv44Wwmddwc8bM3MMw9BaPCEZOTij9SC2gd5IGnq7zAWFtZlbC//fZb3n//fYYPH25ZWSY6OppXX32VF198kYceesimQUrSvUwIwft7L7Mj9hoZhVq0JkFzFUQ5JaFSumMI60DxlTw8XRw4l6GmRaBKXmyspaxK2OvWrWPmzJmMGDHCUjZw4ECaNGnCl19+KRO2JNmIySz47Egi30SnciGzEKPZjLerE119c/BUqRg57AHURgWeKjd0BjPODnYyWddiVvVhZ2RkEBkZWaa8Y8eOZGZm3nZQkiSVOJNWwKGEXK7maijWmzGbBUYBLTp246HhQ3FxccFX6cygpv50DvciqolvVYcs3UFWJezQ0FB2795dpnz37t3y7kdJspEjCTkcT8ojr9iAQqGgkZueWWFZ1HEwkaoRODv/tZyXq6M9DXzccXOSi+fWZlYl7MmTJ/PJJ5/wwgsvsGbNGtasWcMLL7zAJ598wpQpUyrd3rJly4iMjMTZ2ZmxY8fetO6+ffto2bIlbm5udOzYkdOnT5dpKyQkBKVSyUMPPURubm6l45GkqnQho5D39lzi3X2X+d+pFM6m59PJ28y0ujk4ODrRqb4fnq7yNvN7kVUJu3///qxevRqlUsmuXbvYtWsXSqWSNWvW0K9fv0q3FxwczOzZs5k0adJN62VnZzN8+HBefvllcnNzGTduHMOGDUOn0wGwc+dO5syZw9atW0lLS8Pe3p5nnnnGmpcoSVWiWG/ip3MZ/HQug98TcrmUraG+s46Brsng5sWA++/nme4NGdZCTuZ0L7LqTsc7Ze7cuVy4cIFvvvmm3O2rVq1i5cqVHD9+HCi5eh4WFsaKFSsYMmQIjzzyCEFBQbz77rsAxMXF0bx5c7KysvD09KxwHPJOR6kqmM2Cr08ms+y3eM6mFaI3m/F0UvBa3XRMLh7c168/7UK9qzpMqQpV+Axbq9WW+v1mP3dKTEwMbdu2tTxWKBS0bt2amJiYcrc3btwYJycnLly4cMM2Fy9ejL+/f6kfjUZzp16CJN1QTHoBG06ncTW3GL3ZDGYw2zmw0xCGLqQ1RYZqc24lVZEKX6Ho1asX27dvx9vbm549e5Y7dEgIcUfnElGr1dSpU6dUmZeXF4WFhZbtXl5eN9xenpkzZzJz5sxSZVFRUbYJWJJuoUBrIFdjwGgys/lMGql5GkxC0M5DT3v3Yo6IupicnfFyc0FrNN+6QalWq3DC/uSTTyzrNK5YseKOBXQzSqWS/Pz8UmX5+fmoVKoKbZek6kRrMPHLxWv8kVLA2hNJqPUmTCYznTx1jKiTjc7Nn4c7NcdP5UKxwUxTf2VVhyxVsQon7A4dOpT7+93UsmVLPv30U8tjIQR//PEHzz77rGV7dHQ0jzzyCACXLl1Cp9PV2NXbpdorS61DrTNSbDBxKCGba0V6zGZBe6WGEV55ZNr70KlDF7rW98HFUc4PIpWwapTIqVOnOHv2rOXxzz//zHPPPceiRYsoLi6udHtGoxGtVovRaMRsNqPVajEYDGXqjRw5kri4ONatW4der2fp0qVAyagVgIkTJ7JmzRpOnTqFWq1m9uzZjBw5slIXHCXpTkvKLWbbhWssPxjPtnPpXMrW4GKvIMTJwOMBeVw0epLgVp/cYhPHk/KqOlypGrEqYS9evJisrCwAEhISmD9/PmFhYZw6dYr333+/0u3Nnz8fV1dX/vvf/7Jx40ZcXV0tQ/yUSiUHDhwAwMfHh82bN7Nw4UI8PT1Zv349W7ZssdxAEBUVxdy5cxkyZAiBgYHo9foq676RpBtJytPw+ZEEvjyRzI7YLHRGEypXRxrU9eOSeyP0AU3pEOoFgKO9nFBT+otVw/p69uzJN998Q0hICKtWreLcuXN88MEHxMTE8O9//5vt27ffiVjvGjmsT7pTcjV6Xtl6jq+jUzCaSi4i9vcpxkflivCqS++GPjzSIYQCrZF8rZH63m7Y28m5QaQSVt3H6uDgYLlZ5ciRIwwcOBAAb2/vm47IkKR7WWahjr2XrvHzhUx0BjNmYLCvhgfq5HPZzgmT0gl7OwU5GgNBHi74KZ1v2aZ0b7EqYbdt25YPPviANm3aEBMTw/z58wG4evUq/v7+Ng1QkmqDCxmFfBOdwvbzmeRo9JiBAd5qHvAu5Ip9IG4hjWkT4oWnqyN+7vK2c6l8VnWQvfLKK9jb27N7925eeeUVAgMDAfjtt9/o2rWrTQOUpJoqV6NHazCRmqdh2cF4fjqbQWqBFhcHe7p7anjQt5D96jrE2gfSKtiTB1sFcV8jXxxkv7V0A9Xq1vTqQvZhS7crJq2Ag/E5nEzO41x6IRevFWGvAKNZ4OXqSKjSngbOGrKd/BjRKpC2IZ60CZajmaSbs6pLJCEhATs7O8LCwgA4fvw427Zto379+owfPx47O3mGIN3bYq+pOZteyNGkPIq0Bgp1JhSYGehThNbNj24RQYxuE4LOZMbBzo4IeVOMVAFWZda5c+dy/vx5ANLT03nhhRe4du0aX331FcuXL7dpgJJUEzX0cUfl4kCAyolCnREhTIzwLWBQnQLqu+gYEBFARICK1sGeNA9UyZEgUoVYlbDj4+Np3rw5UDKlacuWLfnwww/5z3/+wy+//GLTACWpphFCUNfLmTyNnrR8HXXcnHg8uIjenkUcF8E83KcjncPr3LohSfqH216e4vDhw/Ts2ROAkJAQuWCAdE8ymMwcTsxFqzeSVqhj7+Us9lzKRqM38UCdPCLdCol1rsf97VrSt7FfVYcr1VBWJeymTZvy2Wef0alTJ44fP87LL78MQFpaGt7ecr5e6d6TkKMhPlvDykMJxGYV4ahQUKAzYBKCczolvVvVZ3Kr5jTyk33VkvWs6hL597//zYULF1i0aBFPPPEE4eHhAOzatYvWrVvbNEBJqgkyC3V8cewqJ1MKyC82kF2ko7eqEHcHOzLNLlwxeeHl5iT7qqXbYtUZduPGjfnf//5Xpvz555/H3l7OLCbdWzR6I3suZZGt1mEwmRFCMCk4jzZKLcX5Krz9PahXx41rah0BKnn3omQ9q/uwtVotBw8eJDk5mYceegiVSkV6ejqenp5lFhGQpNpszdEkfjiXQVJ+MY4IngzJpbm7jlMODXi8dxNC67gC0NDXvYojlWo6q8dhT506FbPZTE5ODv3790elUrF582by8vKYM2eOreOUpGpFozfya1wW0akFfPdHKleyi9AbTEwKySXCVccBcxgTurYiKsJP3rko2YxVn6T33nuP3r1789NPP1mmNgXo3bs3J0+etFlwklRdJedrOZ6Ux+HEbBKy1BTpzZiEgkS9K7sNYQzt1AITyGQt2ZRVn6YzZ84wbty4Mnc0+vv7W+bJlqTaxmAy8/P5dL49nUqeRs+ZtEIuZhaBMNPNsxgXR3tSHAJo0iCMYoOZFgFyaTrJtqyeXrW8lcWTkpLk6i5SrfVrbBaf/JYIQPMAdwp0BoxGA1Pr5hLgYMBJ70eonwfBKhcGNwsg2NOliiOWahurEnavXr34/PPPWbBgAQAKhYLMzEyWLVtGnz59bBmfJFWpU8n5JOZpUGuN/BCTxh+pBahcHAhUOpNTqGGcVzp+DkZiXCMY0yaMIr2JTmFeMllLd4RVs/UVFhbyyiuvcPHiRdRqNQEBAVy7do0WLVrw4Ycf4ubmdidivWvkbH0SQKHWyI/nMkjOLyYuU83O2GtkFelwdrCnma8z/Ryu4mlvZJM6mL4t6/NYxzDsFQrq+9Tsz79UfVl1hq1Sqfj444+Jjo4mLi4OjUZDREQEnTt3RqGQNwZItYOrox0qFwdcNXaczyggOU+LEdAYjFzNtyPP25mf8/xpXi+Q7g28aSSH7Ul3WKUTttFopF+/fqxevZq2bdvStm3bOxCWJFU9o1nQo14dCrV6cjQGzIDK3oSvg4lCgz0x9uE8GOnLAy0CaC4vMEp3QaUTtoODA56ensh1D6Ta7GRyHmuPJ5NfbOBYUh5Xc4vxdjLxfEgOKBQccwqhWaAHD7cNpomcH0S6S6wa1jdx4kSWL1+OWq22dTySVOXMZsGBy9mcTS9g/5VsEnKLcLczMj0kG5WjgoKAlnRv4Euvht4yWUt3lVV92Js3byYhIYH777+fkJAQXFxKXxH/8ssvbRKcJN1tWWod8365aLnAWKgz4WFn5PnQHJwc7DA36MjwhkEYzYLWwR5VHa50j7EqYffo0YMePXrYOhZJqlLpBVo+/i2er0+lkFtsxEzJV1BXJwXFOOJUvx3jOjciyEMO2ZOqhlUJe/LkybaOQ5KqVEahjo9/S2D10atkFxsB8HU0YlbYU4QzurD2/KtfM5wd5GyUUtW5rRVnjhw5Qnx8PAANGjSgU6dONglKku4GIQS/J+RyNVfD2YxCVh+5SlqhHoBAJwMvhuZwDSVO9drQrb43jnJxaamKWZWwk5KSePnll0lISCAoKAgoWW0mPDycRYsWWVZTl6TqrEhvYu+lbPZcyuRceiHpagMAwU4GZoTmYLBzpGGL9ni4u5NfbERrNOHmdNur6kmS1aw6ZXjrrbfw9vZm69atbNq0iU2bNrF161Z8fX2ZP3++rWOUpDsiPb+YL44nsudSDmlqAwKo62zgpdBscg327DHVo2/TYLxcHWgeqJTJWqpyViXsc+fOMWPGDHx9fS1lvr6+TJ8+nbNnz1a6vby8PEaPHo1KpSI4OJglS5aUW2/9+vUolUrLj7u7OwqFgk2bNgGwd+9e7OzsStW5Pt+JJP3TqiOJXM0pxvy3siAnI+l6Bz7L9KNLA39CvVwZ1CyANsFyUjOp6ll1yhAaGkp+fn6Z8oKCAkJCQird3rRp09DpdKSkpJCYmEi/fv2IiIhg0KBBpeo98sgjPPLII5bH27ZtY+zYsQwcONBS5u/vT3p6eqVjkO4tJrNgR2wW+j+ztY+jkVyDPbEGJc52gUzp4ce07vVxcZQXGaXqw6qE/fzzz/Pee+8xZcoUWrRoAUBMTAyffvop06dPR6vVWur+c4z2PxUVFbFx40ZOnDiBh4cHrVq1YtKkSaxevbpMwv6n1atXM2bMmBo/2ZR0d5xNL+Rippr63m7EpBeQnq/FXgENXfRMq5vDYb0fysAQXuvfiLA6bnJeHKnasSphT58+HShZPf36h/r6reozZswoVffo0aM3bSs2Nhaz2UzLli0tZW3btrV0c9xIdnY2W7ZsYe/evWXKAwMDcXZ2ZuDAgbz99tt4e3vfsJ3FixezePHiUmWNGze+6b6lmulcRiEJORoW77nEpWsF5BabaeKm47mQXPLt3KlXvyFtQ70J8nCVyVqqlqxK2CtWrLBZAGq1usyiB15eXhQWFt70eevXr6dhw4Z07drVUta0aVOio6Np1qwZKSkpPPPMM0yYMIGtW7fesJ2ZM2cyc+bMUmVRUVFWvBKputPpjSzYGUu+zoQAWrjpeCYkhwzhjjGoFaPbh9HQxw0nBzl8T6qerErYHTp0sFkASqWSgoKCUmX5+fmoVDef/WzNmjU88cQTpcoCAwMJDAwESvrZly1bRqNGjdBoNLLb5B5WpDOSkFPE+/uvkKczWco7ehSTanLnsntjPLUmsov0tAuRFxel6svqRXj1en2Z8pycnDJdIrfSpEkTFApFqdEl0dHRpbpI/unUqVPExMTw2GOP3bRtOzs7hBByZsF7WHx2EauPXeU/O2OJvVYyWZm7XcmVxh2Fvng37UC4jxI/d2eK9MaqDFWSbsmqhP3777/z6KOPEhsbaynbu3cvY8aMKTeR34y7uzsPP/wwr7/+OoWFhcTExPDZZ5/x5JNP3vA5a9asYdCgQZaz6ev27NlDQkICQgjS09N5/vnnGTBgAO7ucmL5e4nBZCYmrYAtMWk8u/EP3tkdx4/n0ikyCNori/lvg0zqORvQmhXEZWsZ3Myf1kEeNJNzWkvVnFUJe/369bRr144nnniCL774gnnz5vH666/z5JNPsnz58kq3t3z5chwdHQkKCiIqKopZs2ZZRogolUoOHDhgqavX6/nqq6/KTeinTp2iZ8+euLu706FDB3x9fVm7dq01L1GqwaJTCjiTVsi8Xy6y/0o2KQV6igzQUVXMpOA8TqtdyDA44OFqT6CHC30a+TKmXYicKlWq9qxa0/G6FStW8Pnnn2Nvb8+KFStqzeozck3HmqtYb+SNbRc4GJ/L6dQ8tH92WXfx0DAhMJ+TGnd+MwTSxNedbvV9aRvsSVSEX9UGLUkVZNVFR5PJxIoVK1i/fj3jx4/njz/+4LXXXuPNN9+kS5cuto5RkipsR2wW+y5nE5tVhOHPZO2gEAzyUXNCoyQooi2LwusQ7u2OnUJBQ195MVqqOaxK2I8//jhqtZrly5fTrl07zGYzn3/+OTNmzGDkyJFlhslJ0t1wIaOQTadTOJNewPXBII4KgbODgo2FIfRqHEjTAA9aBHlSz1smaqnmsaoPu1GjRnz11Ve0a9eupBE7OyZNmsRnn33GkSNHbBqgJFXEpWuFPLLuBBujUy3Jul+dIl4Lz8LTyY42dX3oUs8HVyd7/NydqjZYSbLSbfVhl0er1d7ydvTqTvZh1zz3r/iNHXE5lscDvNU85FfIz9lKEhyC6NnQl2e7hRPhr8LeTt7FKNVMVt/StX//fqZPn86oUaMsky199913REdH2yo2SbqlxBwNH+y7xKGEXEvZYO9CHvIrZGeeCp1PPXo28CZA5Ux9bzeZrKUazaqEvXXrVt58800aN25MamoqRmPJDQd2dnZyAV7prtDojOyOvcbnhxN4e9cl1IaSL4r+jkYG+6jZmq3CM7wZq8e0Y1jLICJDPdGb5A1UUs1mVcJeu3Yts2fPZtq0adjb/zX9ZKtWrYiLi7NZcJJUHiEE7++/zCtbYvjP7stc0xgQCBQIMg0OzE3wI8Hen6HNA9CbBFlFBnI1RmLSbz4/jSRVd1aNEklJSbFMq/p3Tk5OaDSa2w5Kkm7mZHI+P59N50Sa+s8SwSi/QlztzXyZ7onZwZl3h7aif4QfxQYTjvYKDCaBj5tjlcYtSbfLqoQdGBhIXFycZT3H6w4fPkx4eLhNApOk8hhNZjbHpJGYVwyAAsFY/wJ6eWlYn+GJAgXebk408nPH3k6B0tmBIc0D0BnNeLnKhC3VbFYl7EcffZR33nkHg8GAEIILFy6wa9cu1qxZw6xZs2wdoyQBcC6jgHXHk/ktPhut0YQCwSMB+XT3LGZjthfndO74utvTv7Evvn8buufqaI+rXDlGqgWsStgjRozAxcWFZcuWodVqefXVV/H19eX555+/5SoxkmStL44k8fOFTBJyNRiNZiJVWrp7FrOlwBfv4LqMVjrTMlDF8JZBeMqzaakWuu1x2FqtFo1Gc9NVXWoaOQ67+tAaTGy/kMG+yzlsPZtGaoEOk1lgFoAQNPUQBAX4sXhocyL8lbg4ypXNpdrrtj/dLi4uNf5GGal6yinS8c6vl/jk90QK9SW3L9oheCIojzNqF85r3dE6OtMprA4tgzzlGGup1pNrIUnVUqHWyILdcXx1MtmSrB0UginBubRTasHOniBPF7qGe9M8QIVM1dK9QH5/lKqdbLWO9/dfYe2xJDI1JTdlXU/WTd10fJziTZGTBw81D6BDqBd2CgUmIbCTaVuq5WTClqqVPXFZLNgVS9y1QkuyBhjhW0iEm55lKd5onDzp39iPh9sEoVDYEeLhgqO9/LIo1X4yYUvVxpWsIub8cpGjV3P421q5APycreRkoQsaR3da+LrRr7EvfkoXGvnK5d+ke4fVCfvw4cOcOHGCnJwczGZzqW1z5sy57cCke0t2kZ6vTiVzOjXfkqydFWYeDcznhywVejsnXD096BnqRdtgD7qE16GBj5zTWrq3WJWwP/nkE9asWUPLli3x9fVFoZB9h5L1jibmsuboVbZfyKTgz2ztYmfm+bo5BDoZ+a1ARXiIN4ObB+Bkb8fwloEEeciRSdK9x6qE/f333/PWW28xcOBAW8cj3WMuZqpZsv8yv8Zlkas1AOD2Z7L2czLyQZIPdQN86NfEj1yNgSZ+SrKK9DJhS/ckq7tEmjdvbss4pHtMfrGBuKwijibmcuBKNplFBkru4BI8G5KLj6OJ95N8yNA7MiBAicrZAXsF2CmQK8ZI9yyrEva4ceP49ttvefHFF20dj3SP+D0hh/QCHV+dTCalQM9ft9sq2JylQmNSkGd2ol2IinEdQmkd5IHSueTjKucFke5VViXsCxcucOTIEQ4ePEiDBg1wcCjdzMKFC20SnFQ7mcyC+GwNX59M4lBiPgLwsDcx0FvNd9c8uFzshKeTgs7hXjzVKZzu9bxxkUlakqxL2G5ubtx33322jkW6R/wWn8X3Z1I5nJiPGfByMPFiaDaOCsHuXCUKJyfaBHsxpWs9hrcMxMlBjrGWJLAyYcthe5I1ig0mcov0vLUjll8v5SAAbwcjL4bmoADeveqDm5sr9zcNoG9jX0a1Da7qkCWpWpE3zkh3xZHEXK5ka7icXcTJ5AIE4Gpn5qWwHEwCliT54O+lpKGfCncnB5r6K6s6ZEmqdqxO2Js3b2bnzp2kp6dbFuG97ocffrjtwKTa5UhiLmfTC9l+IYM8bcnnpdhsx84cd04VuuDo5ESPej7cF+FH5zB5U4wklceqzsHVq1ezbNky2rRpQ1paGlFRUTRv3pyCggJGjBhR6fby8vIYPXo0KpWK4OBglixZcsO6CoUCd3d3lEolSqWyzIIJ3377LQ0bNsTNzY1+/fqRmJhY6Xgk24jP1rD3UhYfH7zCu3sv8fmRq1zN1xHgZKCbR8nan3vz3NEIe1ydHHipb0PGt69LQ193eTOWJJXDqjPsLVu2MHv2bPr06cO6det48MEHqVu3LmvXruXSpUuVbm/atGnodDpSUlJITEykX79+RERE3HD1mhMnTtC0adMy5efPn2fixIls2rSJnj178tprrzF69GiOHDlS6Zik23NNrWPf5SyOXM3lm5PJ5OlKpi8IdjIwIzSHPKMdRwtcUSgUeDo7MKlLGE0DPKo4akmq3qw6w7527ZolYbq6uqJWl6xefd9997Fv375KtVVUVMTGjRv573//i4eHB61atWLSpEmsXr260nGtW7eOgQMHMmDAAFxdXXnrrbc4ffo0Z8+erXRb0u0xmgTphVr2xF2zJOu6zgZeCs0m12jPkiQfXJ3sqePmSGSYF091los3S9KtWJWwAwICyMrKAiAkJISjR48CJeOzHR0rt5ZebGwsZrOZli1bWsratm1LTEzMDZ/Tt29fAgICGDJkSKlkHBMTQ9u2bS2PVSoVDRs2vGlbkm2ZzIItMem8t+8yP55N53JWyermwU4GXgzNJtPgwAdJ3vh6uNE62JOOYXXoULeOvBlGkirAqi6R3r17c+TIEVq2bMnYsWN544032LJlC6mpqYwbN65SbanVajw9PUuVeXl5UVhYWG79vXv30rVrV3Q6He+88w4DBgzg/PnzeHh4oFar8fLyqnBbAIsXL2bx4sWlyho3blyp1yCVEEKweE8cX59M5eK1wlJTpGYZ7DmU78bWbCWuTg6se6Qd9bzduJJdTF0vF+q4ydvNJelWrErY06dPt/w+YMAAAgMD+eOPPwgLC6NXr16VakupVFJQUFCqLD8/H5VKVW793r17A+Dk5MT8+fNZu3Ytv//+OwMHDkSpVJKfn1/htgBmzpzJzJkzS5VFRUVV6jVIJU6nFrDnUjZnMwox/XmveUNXPTqzgmSdIxuveeBkB53CvfFTulDXy426XnI0iCRVlE3GYbdu3ZrWrVtb9dwmTZqgUCg4e/YsLVq0ACA6OrpUF8nN2NnZcX3h95YtWxIdHW3ZplaruXz5coXbkqyXkF3E2mNX2X85y5Ksm7jqmFY3l9NqZz5Pq4Md4Kd0pkd9H3zcKtd1JknSbSzCu3//fqZPn86oUaNIT08H4LvvvuPw4cOVasfd3Z2HH36Y119/ncLCQmJiYvjss8948skny9Q9e/YsJ0+exGg0otFomDt3LsXFxXTt2hWARx99lG3btrFr1y60Wi1z5syhdevWln8Eku0VaA3svHiNjw5c4f+OJaL9sxukmZuOf9XNIVbjxP+le2EPdK/vxZwBjXmxdwN8lc5VGrck1URWJeytW7fy5ptv0rhxY1JTUy03ztjZ2fHll19Wur3ly5fj6OhIUFAQUVFRzJo1yzKkT6lUcuDAAQAyMzMZP348np6ehIWFcfjwYX755RdLv3WzZs1Ys2YNkydPxtvbm1OnTrFhwwZrXqJUQRcy1STkFvHVqVSytSWn1i3ctEwNyeFckTMrUutgj4IX+zRg5cNtGdc+FDcneYOtJFnDqr+ctWvXMnv2bPr3718qIbZq1YqPP/640u15eXmxcePGcrddHzIIJcMGL1y4cNO2Ro0axahRoyodg1R5ap2Ri5mFvPfrJdLVekt5rtGewwWufJ3hiQkFbk721PVwoVngja8lSJJ0a1adYaekpJTbzeDk5IRGo7ntoKTqL7/YwKeH4pm97SIXs0uG7jV10+GsMJOqd2RdhhdCocDHzZF2oZ6oXBzJ1ehv0aokSTdj1Rl2YGAgcXFxBAUFlSo/fPgw4eHyBoja7nBCDkv3X+bHc5moDSU3xXRUFfNEUB6bs1TsyFHiZg9j24XQIbQOSid7PFwdcZddIZJ0W6z6C3r00Ud55513MBgMCCG4cOECu3btYs2aNcyaNcvWMUrViMFkZsbmsxxPzsP452iQLh4aJgTm83u+Kztz3HF3UPDusOZM6VYfhUJBrkaPu5ODnNdakm6TVQl7xIgRuLi4sGzZMrRaLa+++iq+vr48//zzN5z/Q6r51Doj7++9zIm/JevunhoeDchnf54b32R6AAqWDm/OU90aWJ4nb4qRJNuw+jvqoEGDGDRoEFqtFo1Gg7e3ty3jkqoZk1kwfdMffHk8hb9PpmuPYE+uGxuulSTrYJUTLUK8qihKSardbrtT0cXFBRcXF1vEIlVTBpOZVb8nsD76r2Rd30VPvNaJ/fnulnoN6zgxqFkgQSr5eZCkO8GqhJ2Xl8fy5cs5ceIEOTk5ljsNr6vsjH1S9bb9fDqv/Xwe3Z/ZekAdNQ/5F/LuVW/iiktugPF1d+SpLvUY3CyQcG95u7kk3QlWJezZs2eTnp7OyJEj8fb2lpPN12KpeRoeW3+KfH3JaJBB3oU86KfmhyylJVnXcbGnU6gXz3ZrgJe85VyS7hirEvbp06dZs2YNjRo1snU8UjWRX2zg8yNXWbLv0p/JWjDUR80Dvmo2XVPxS44SO6CRrytj24TwWKcwmawl6Q6zKmE3atSo1B2IUu3zyo9nWXM4ieu3uiiAEGcDGzNV7MotSdZNfN14tEMIM/s2kUP2JOkusCphv/7667z//vuMHTuWBg0a4OBQupnAwECbBCdVjS+PJvG/6NQ/k7XAz9HENYMDK1PrICjp/hrawpcHmgfTs4GPTNaSdJdYlbBNJhNZWVm89NJLpfqvhRAoFArLCjRSzXMlq5DXf44hT2tCgWCMfwFdPIqZHe+H2lSyKkyIyomH29SlS7g3DeXq5pJ011iVsOfMmUNgYCD//ve/5UXHWkJnNPFrbBbPbz5DcqERBYLxAQX08NSwNt0Ttckee6BNiIp/9ajPmLYhONrLM2tJupusStjJycksWrSIsLAwW8cjVYHMQh3f/ZHG0v1xXMrWokDweGA+XTyK+SLdkyMFJWfREf5uzB3QlKEtZZeXJFUFq06R2rdvz+XLl20di1QFCrQGtl3I4KODV7iYpQVAZW+msauez9O8LMk61MOJPg39cHeWEzhJUlWx6q+vb9++vP/++1y+fJlGjRqVuejYo0cPmwQn3VnpBVouZRWx+shVzmcWYYfAxU5QYLJnboIfRlHS1dWznif3NfanXYgn7UI8qjhqSbp3KcQ/b1OsgI4dO964wVpw0TEqKoqdO3dWdRh31IWMQnbGXmPpvstcztVij2BScC6eDmYWXfWxjAaJ8HXl5fuaMLRlAH5yWS9JqlJWnWEfO3bM1nFId5HOaOL7M2l8sO8S1zQmHBSCKcG5NHXTsSLlr6F7oZ5OfDiyFb0a+OLiaF/FUUuSJDsk7zE6g4lPfitZKcYMOCoEz4bk0thVx/IUby5oSs6iA5WOfPRgawZEBFRtwJIkWVT4ouP69evR6yu2xNOFCxc4dOiQ1UFJd4ZGb2Tejgu88tN5zH+W1XfRU99Fz4fJfyVrT2d7HmoTwn2NfasuWEmSyqjwGfb58+cZOnQoUVFR9OjRg+bNm+PhUXIBSghBfHw80dHRbN++ndTUVObNm3fHgpasc/xqLlti0tGbS86sDQJii515/Yo/GrMdjsCgZv6Mj6zLmLYhVR2uJEn/UOGEPX/+fC5evMjGjRt59dVXKSoqwtHR0bLwrhCCJk2a8OCDDzJ06FCcneUFqupk8x+pjP3yBDoBLnZmnq+bQ6zGic1ZHmjMJV+0mga60y/Cj/sj/Ks4WkmSymPVKBGz2UxcXBxpaWnodDq8vLyIiIjAy8vrDoR499WmUSJ6o5lTKfmMW3uM+Fwdbn8maz8nI0uSfEjSlcywF6B0YMnwloxuWxc7O3nnqiRVR1ZddLSzsyMiIoKIiAhbxyPZUKHWyJL9l1h+MIGMIgPudmZeCM2mjoOZ95N8SPkzWY9p7c/4DmEMbREopxmQpGrM6lEimZmZfPvtt8THxwNQv359HnroIQIC5KiC6uJMWj6rDiWSUWQAoHedIjwdzLyX5E2aviRZ16vjwuqxHXCTdzBKUrVn1a3pv/32GyNGjGD//v14eXnh5eXFgQMHGDlyJAcPHrR1jJIVdl7MYMyXR0kq0KOgpNdrW7aSBYm+lmQd6ulMzwY+FBlMVRmqJEkVZNVp1Xvvvcfo0aOZPn16qfKlS5fy/vvvy1vTq1hyroYZm2NILjDi5WDiX3Vz+C7Tg3MaZ/KMJTfANPV14/GOYXQM9ZJ3MEpSDWFVws7IyGDEiBFlykeMGMGGDRtuOyjJeuuOX2Xu9vNcztXj7WDkxdAcFECGviRRK4Bu4Z7MjoqgfxM/HOQUqZJUY1j119quXTvOnDlTpvzMmTO0adOm0u3l5eUxevRoVCoVwcHBLFmypNx6hw8f5v7778fHxwcfHx+GDBlCXFycZfvevXuxs7NDqVRafhYsWFDpeGoig8nMD2dSmbzhNJdz9fg6GnkpLAcBvJvkQ7bRASegW5gn303sxMBmATJZS1INY/VsfR9++CEXL16kRYsWAMTExLBjxw6mTJlSqh+7It0j06ZNQ6fTkZKSQmJiIv369SMiIoJBgwaVqpebm8uTTz7Jhg0bcHV15Y033mDYsGGcP3/eUsff35/09HRrXlaNJITgTFoh648n8dHBKxT/2R09PqAAoxneT/Ih/8+VYp7oXJfHIkMJ8HCpwoglSbKWzWfrK9V4BWbuKyoqwtvbmxMnTtCyZUugZM3I2NhYNm7ceNPnZmZmEhAQQFZWFj4+Puzdu5exY8fedsKuSeOwTyTlsfVsOgt3x6EzAwhAgYd9SeYu+DNZt/Rz5+sJHWge4CHHWUtSDVXls/XFxsZiNpstyRqgbdu2bNq06ZbP3bdvH4GBgfj4+FjKsrOzCQwMxNnZmYEDB/L222/j7e19wzYWL17M4sWLS5U1btzYildy96l1RnbEZvLh/svozBDsZOCRgHxWpdWxXFwEaB+o5JX+jWkZ5FmF0UqSdLuqvBNTrVbj6Vk6kXh5eVFYWHjT5125coVp06aV6u9u2rQp0dHRpKamcvDgQZKTk5kwYcJN25k5cyaZmZmlftzcqv/Csmaz4N09cbz580VydWbqOht4MTQHRzuBwfzXGXRzP1feGtKMXg3lRE6SVNPZ9G6J7OxsNm3axKRJkyr8HKVSSUFBQamy/Px8VCrVDZ+TlJRE//79eeWVVxgzZoylPDAwkMDAkvUGQ0NDWbZsGY0aNUKj0dSIJFwZm/5IYd7OSwCEORt4ITSbTL0DS5O9KTbb4QB0Cvfki7HtaeyvrNpgJUmyCZueYWdnZ7Nq1apKPadJkyYoFArOnj1rKYuOji7VRfJ3ycnJ9O3bl8mTJ/Piiy/etG07OzuEEFjRTV9t5RTp+PxwAhO/OgWUzLo3tW4OaXoHlvyZrO2ArZM6sW9qD5msJakWqdQZ9q3uYkxJSal0AO7u7jz88MO8/vrrrF27lsTERD777DPWrFlTpm5qair33Xcfjz76KLNmzSqzfc+ePdSvX5/w8HAyMjJ4/vnnGTBgAO7u7pWOqzpS64zM+vEcXxxNxvBnmUEoWJlShxSdAzpR8v936bBmDGwqpwiQpNqmUgl7xowZKBSKm56xWjN50PLly5k0aRJBQUGoVCpmzZplGdKnVCrZtm0bPXv2ZNWqVVy6dKnMhcJz584RFhbGqVOnePzxx8nOzqZOnToMHDiQhQsXVjqe6shoMjNzSwyrjiYD0NhVR2ePYtZneHJF62SpN7tvQ8Z3DKuqMCVJuoMqNaxv0KBBvPLKK/Tp06fc7RcvXuSxxx6Ti/DeAcv2X+JfP5SMN2/qpmNqSA6xGmc+Sa1jWd38i9GtmNC5XhVGKUnSnVSpPuymTZty4cKFG26/1dm3VHnFeiNv7bhgSdYt3LRMC8nh/D+S9chWATwSKc+sJak2q1SXyOOPP05xcfENt4eGhrJixYrbDkoqYTYLXvzhDCsOl3SDBDoZeDYklzNFLnyW6oXpz9XNn+0SytsPtJC3mktSLVephN2uXbubbnd1daVDhw63FZD0l49/u2JJ1gDpege+TPfieKEL5j+T9WPtgvhwZGuZrCXpHiD/yqupr08m88qP5wCIVBXT2UMDKDha6GpJ1uGeTrzYp5FM1pJ0j7DqxpkpU6aUOxpEoVDg5OREaGgoQ4YMoVmzZrcd4L0oW61j8oZTaIzQ2UPDxMB8fst35UjBXzf/TOwQxH8Gt6Cul2sVRipJ0t1k1alZkyZNuHDhAgUFBYSGhhIaGkphYSHnz5/H39+fc+fOMXHiRA4fPmzreGs9vd5E04U7URug25/J+kC+G+sz/rp9f9fTnVgzPlIma0m6x1h1hm00Ghk1ahRTp04tVb58+XLUajWrV69m6dKlfPLJJ3Tp0sUmgd4Ltp/P4OlvTpBVLGinLGZCUD67c93YkOkBKPBygqQ5g1C6yPUXJeleZNUZ9vbt2xk2bFiZ8qFDh7Jt2zbL7wkJCbcV3L3kwOVsHvr8KCnqkmlRz2uc+SbDw5Ks7YD/G9teJmtJuodZlbAdHByIiYkpU3727FkcHUsWeFUoFDg4yORSEVlqHfd9/DsaAb29ighwNKI127Enzx3+vMC4YHAThrUJqdpAJUmqUlZl1NGjR7NgwQIuXLhA8+bNgZLbwzdt2mSZzvTw4cNERETYLtJa6lRSLh2WHEQAg7zVPOhXyMZMFRm5f03aNCjCh+m9GlVdkJIkVQtWrTgDsG3bNjZs2EBiYiIA4eHhjBkzhoEDBwKg0Wiws7PDxaXmLUd1t25Nz9HoCXnjF7QIHvBRM9RXzeZrKrbl/JWs+zWsw6dj2tLAR866J0n3OqsTdm12NxJ2cl4xvZbuJb7AyGDvQob7qdmYqWLX386sv328PSNaBcslvSRJAm5zAYOjR48SHx8PQMOGDYmMjLRJULWdwWgiYv4uNH/+qzyncaYow459eSXTwNoD/zeuLQ/JPmtJkv7GqoSdnp7Ov//9by5dukRwcDBQMld1kyZNWLx4MQEBci7mGzEYTQS/8TMaIejlqeH3AjcStE4k/DlFqqsdfP9UR+5vGljFkUqSVN1YlbAXLVqEu7s7W7duxc/PDyhZwfzNN99k0aJFvPfeezYNsrY4nZJPz4/2ozYIxgcU0NNTQ7bBnrOakn5+dwdIm3s/KlenW7QkSdK9yKphfceOHeOll16yJGsAf39/XnjhBZuuqF6bHLyUSdv3S5L1Y4H59PTU8EW6pyVZezop2PJkZ5msJUm6IavOsB0cHNDr9WXK9Xo99vb2tx1UbaPVG+nzyRFAMCEwn04exXye5sXxwpJbyxt6OnDh9fvlJE6SJN2UVRmie/fu/Pe//+X8+fOWsnPnzvH222/To0cPmwVXG5xLy0f16jZK7l9UkKJzYFXqX8m6Z7iK2NkDZbKWJOmWrBrWV1BQwJw5czh48CBOTiVf4Q0GA927d2fevHl4eHjYPNC7yVbD+owmM16v/IRWCFq46/ijqPSY9D7hbvw4pTfuzvKOUEmSbs2qTOHh4cEHH3xAYmKiZb6Q+vXrExYml6j6uxbzt6NDMCUkl2ZuOt6M9yfXWNJl1CVMxXdPd5PJWpKkCrutbBEeHk54eLitYqk1hBB0WLyL+EIjzwbn0sRNx/IUb0uy3ji+LcPbhuAou0EkSaqECifspUuXVrjR6dOnWxVMbfHgZ78Rk1nM1JAc6rsa+DDZm7hiZwB2T+lE3yZynLokSZVX4YR97ty5CtUrbyWae0lqXjFbLuTiqIBisx0fJntzubikn3/HpI4yWUuSZLUKJ+yVK1feyThqBSEEabmFhDgbSNE5sjK1jmVb8duDcXGSQx4lSbKe7ES1EbNZsP1sCof27eaZ4Fzs+GvwzbHp3WWyliTptskhCjaSry7i6skDKAzFHBahBKoEkaEefDuhE44OMllLknT7ZMK2geLiYn7++WfMei3RTk3o0cCfacFeDIjwk1OjSpJkMzJh24BOp0NnEng070YXOxd6N/KhoY/7PX8BVpIk26oWfdh5eXmMHj0alUpFcHAwS5YsuWHdffv20bJlS9zc3OjYsSOnT58utX3ZsmWEhISgVCp56KGHyM3NvcPRg5eXFx36DMTORUk9bzeZrCVJuiOqRcKeNm0aOp2OlJQUfvnlFxYsWGBZff3vsrOzGT58OC+//DK5ubmMGzeOYcOGodPpANi5cydz5sxh69atpKWlYW9vzzPPPHNXXkObEE8ebhNEvyZ+MllLknRHVPkSYUVFRXh7e3PixAlatmwJwOuvv05sbCwbN24sVXfVqlWsXLmS48ePAyXD6MLCwlixYgVDhgzhkUceISgoiHfffReAuLg4mjdvTlZWFp6enhWO6W6t6ShJklQZVX6GHRsbi9lstiRrgLZt2xITE1OmbkxMDG3btrU8VigUtG7d2lL3n9sbN26Mk5MTFy5cuOH+Fy9ejL+/f6kfjUZz+y9MkiTJxqo8YavV6jJnv15eXhQWFpZb18vL64Z1b7W9PDNnziQzM7PUj5ubm3UvRpIk6Q6q8oStVCopKCgoVZafn49KpSq3bn5+/g3r3mq7JElSTVblCbtJkyYoFArOnj1rKYuOji7VRXJdy5YtiY6OtjwWQvDHH39Y6v5z+6VLl9DpdDRt2vSOxS9JknS3VHnCdnd35+GHH+b111+nsLCQmJgYPvvsM5588skydUeOHElcXBzr1q1Dr9dbZhDs378/ABMnTmTNmjWcOnUKtVrN7NmzGTlyZKUuOEqSJFVXVZ6wAZYvX46joyNBQUFERUUxa9YsBg0aBJR0cxw4cAAAHx8fNm/ezMKFC/H09GT9+vVs2bIFZ+eSqUujoqKYO3cuQ4YMITAwEL1ez4oVK6rsdUmSJNlSlQ/rq47ksD5JkqqjanGGLUmSJN2aTNiSJEk1hOwSKUfnzp1vufK7RqOR47WtII9b5cljZp2adNxcXFzYunXrLevJhG0lf39/MjMzqzqMGkcet8qTx8w6tfG4yS4RSZKkGkImbEmSpBpCJmxJkqQaQiZsK82cObOqQ6iR5HGrPHnMrFMbj5u86ChJklRDyDNsSZKkGkImbEmSpBpCJmxJkqQaQiZsSZKkGkImbEmSpBpCJmxJkqQaQibsG8jLy2P06NGoVCqCg4NZsmTJDevu27ePli1b4ubmRseOHTl9+vTdC7SaqcxxUygUuLu7o1QqUSqVlkUr7jXLli0jMjISZ2dnxo4de9O68rP2l8oct9ryWXOo6gCqq2nTpqHT6UhJSSExMZF+/foRERFR5o3Ozs5m+PDhfPjhh4wZM4bly5czbNgwYmNjLSvh3EsqetyuO3HixD2/5mZwcDCzZ89m165dZGVl3bCe/KyVVtHjdl2t+KwJqQy1Wi2cnJzEmTNnLGWvvfaaePjhh8vU/fTTT0WHDh0sj81ms6hbt6748ccf70qs1UlljpsQQgDi/Pnzdyu8am/OnDlizJgxN9wuP2vlu9VxE6L2fNZkl0g5YmNjMZvNpVZub9u2LTExMWXqxsTE0LZtW8tjhUJB69aty61b21XmuF3Xt29fAgICGDJkCGfPnr0bYdZY8rN2e2rDZ00m7HKo1eoyK617eXlRWFhYbl0vL68K1a3tKnPcAPbu3UtCQgKXLl2iXbt2DBgwgIKCgrsRao0kP2vWqy2fNZmwy6FUKsu8mfn5+ahUqnLr5ufnV6hubVeZ4wbQu3dvnJycUKlUzJ8/HwcHB37//fe7EWqNJD9r1qstnzWZsMvRpEkTFApFqa9N0dHRpb7qX9eyZUuio6Mtj4UQ/PHHH+XWre0qc9zKY2dnh5Bzkd2Q/KzZTk39rMmEXQ53d3cefvhhXn/9dQoLC4mJieGzzz7jySefLFN35MiRxMXFsW7dOvR6PUuXLgWgf//+dzvsKleZ43b27FlOnjyJ0WhEo9Ewd+5ciouL6dq1axVEXrWMRiNarRaj0YjZbEar1WIwGMrUk5+10ip63GrVZ61qr3lWX7m5ueLhhx8W7u7uIjAwUHzwwQeWbe7u7mL//v2Wx3v27BEtWrQQLi4uIjIyUpw6deruB1xNVPS4/frrryIiIkK4ubkJHx8fcf/994vo6OgqirpqzZkzRwClfiZMmCCEkJ+1m6nocatNnzU5H7YkSVINIbtEJEmSagiZsCVJkmoImbAlSZJqCJmwJUmSagiZsCVJkmoImbAlSZJqCJmwJUmSagiZsCVJkmoImbCtMHny5JuupFLVXn/9ddatW1fVYUh/On78OJGRkWg0Gpu1eeTIEcaPH4/ZbL7rsaSmphIZGcmlS5duWOfAgQNERkZavY+aqCLH5XbJhF1NnDx5khkzZjBw4EAiIyM5cOCAVe1cvHiRo0eP8vDDD9s4Qslabdq0Yfv27bi6utqszc6dO+Pk5MT27dtt1mZNtHXrVvr163fLeitXruSxxx6z2X7nzp3Lyy+/bLP2Kkom7GqiuLiYxo0b88orr9xWOxs2bCAqKgoXFxcbRVb7lTdhkC05Ojri6+uLQqGwabtDhw5lw4YNNm3zXmc0Gqs6hJuSCdsG8vPzmT17Nn369KFHjx7MmDGD9PT0UnVWrVpF//796d27NwsXLuTdd99l8uTJlu3du3fnueee47777rvhfiIjI/nuu++YOnUq3bt3Z8SIERw8eNCy3WQysXv3bnr27Fnmed9//z3/+te/6N69O2PGjOHChQvExcUxceJEevTowdSpU8nJySn1vO+//56RI0fSrVs3Ro0axZYtW0ptX7JkCSNGjKB79+4MHz6czz//vNRX9OtnIatXryYqKoqoqCg+/vjjWx7Pm+33zTffZPz48ZY/LK1Wy0MPPcTixYuBv77y//bbb4wZM4Zu3brx9NNPc/XqVUsb18+2vvvuO4YOHUrfvn0BKCgo4K233qJfv3707t2badOmkZCQYHlebGwskydPplevXvTu3ZvHH3+c+Ph4ANLS0pgxY4blMzB27FhOnDhRKqa/d0Ps2rWLUaNG0aVLF4YPH863335b5j3bvHkzL7zwAt27d2f06NGlplYF6NmzJzExMaSlpd3ymP5TTEwM48ePp0ePHjz33HNkZmZW+D0oz4EDByyfhalTp5Zp70ZudhzK62K4dOkSkZGRpKamcvz4cebNm0d+fj6RkZFERkaycuXKMvvYunUrq1at4vz585Z6W7duBUqO86ZNmyzH+auvvir3rP1///sfQ4cOBUo+Pz/++CO//vqrpb3jx49b6iYlJfH000/TvXt3JkyYwJUrVyp0LCqkqmefqokmTZpUaha6F154QYwePVpER0eL8+fPi0mTJonx48cLs9kshBDi559/Ft26dRNbt24VCQkJ4qOPPhK9evUSkyZNKrf9Dh06lJqh7e/l/fv3F1u2bBHx8fHigw8+EF27dhVpaWlCCCHOnz8vOnToIHJzc8s8b/DgwWLnzp0iPj5eTJ06VYwaNUpMmTJFnDhxQly8eFGMHDlS/Oc//7E85+effxaDBg0Se/bsEcnJyWLHjh2iV69eYs+ePZY6q1atEqdPnxYpKSli9+7dom/fvuK7776zbJ8zZ47o1auXePfdd0V8fLzYtm2biIyMFL///vsNj+2t9ltYWCgeeOAB8dFHHwkhhFiwYIF4+OGHRXFxsRBCiGPHjokOHTqIMWPGiKNHj4rY2Fjx7LPPitGjRwuTySSEEGLFihWiR48eYvr06eLixYsiLi5OCCHEs88+K1566SVx7tw5kZCQIBYuXCgGDx4sNBqNEEKIUaNGidmzZ4v4+HiRkJAgtm3bJuLj44UQQkyfPl0899xzIi4uTiQlJYk9e/ZY1ra8HlNRUZEQQoiYmBjRsWNH8fnnn4uEhATx3XffiS5dupQ6th06dBBDhgwRO3fuFImJieK1114TgwcPFgaDodTxGjBggPjpp59ueDz/6XosTz31lIiOjhYXL14Uo0aNEq+++mqF34OUlBTRoUMHy3FLTU0VXbp0EUuWLBHx8fHihx9+EP369Su1/mR5bnUc/rkfIYSIi4sTHTp0ECkpKUKv14uvvvpK9O3bV1y7dk1cu3bNcoz/rri4WHzwwQdi3LhxlnrXPy8dOnQQAwYMEFu3bhXJyckiLS1NbNmyRfTt27dUG99884144IEHhBBCFBUViVmzZokXXnjB0p5er7fEO2rUKHHo0CERHx8vJk+eLCZOnFjh9+dW5KrptykxMZEDBw7wxRdfWCaSnz9/PkOHDuXIkSN06dKF//3vf4wcOZIHHngAKFlZ3NrVLu6//37Lf/oXXniBQ4cOWc6609PTcXJyKrOMFMDw4cMt8yY/9thjTJ06leeee4727dsD8OCDD/Ldd99Z6q9cuZIXX3yRPn36ABASEsKFCxfYtGmTpezpp5+21A8ODiYuLo5du3YxcuRIS3mdOnV48cUXUSgU1KtXjw0bNnD8+PEbzkV8q/0qlUrmzZvHc889h729PVu2bGHNmjVluoCmTJlCx44dAXjrrbd44IEHOHr0KF26dAFKvo3MmzfPsqRZdHQ0Fy5c4JdffsHR0RGAmTNnsmfPHg4ePEhUVBQZGRk8/vjj1KtXD4Dw8HDL/tLT0+nbty+NGjUCoG7duuW+PoCvvvqKLl26WOYJDw8PJzY2li+//NLyuqH0ezZ58mQeeughkpOTLfsH8PX1teoM+7nnnqNNmzYAjB07ttSZaUXe+7/77rvvqFevHtOnTwegXr16nD9/no0bN940hooehxtxdHREqVQCJcfhRlxcXHB1dcXe3r7ceoMHD7b8bVaEm5sbzs7OmEymctt7/PHHLZ+zJ554gmnTpqHT6Wyysr1M2LcpPj4eR0dHWrRoYSnz9/cnODiYhIQEunTpQmJiImPHji31vBYtWpCYmFjp/f1zdZFWrVpZvpZrtVpLsvmnxo0bW3738fEBoEGDBpYyb29vcnNzgZL+9OTkZObMmcO8efMsdYxGI0FBQZbHO3bs4JtvviE5OZni4uIy26/v4+99tz4+PmW6Xq6r6H7bt2/P6NGj+fzzz3n22Wdp2rRpmbb+fpx8fX0JCgoiPj7e8ocUHBxcav3J2NhY1Gq1pXvkOp1OR3JyMlCS2P7zn//w888/06VLF6KioixxjR49mnfeeYcjR47QuXNnoqKiqF+/frmvMz4+vsxX7jZt2rBjx45SZdeT//XXAJCTk1MqYTs7O6PVasvdz838s+3r70lF34O/S0hIKPO5bN26tSVhp6enM2rUKMu2J554gieffLLCx+FOa9asmU3bK+99y83NJTAw8Lbblgm7FvHy8kKj0WAymbC3ty+1zcGh7Fv99zKFQmHpf77e1zpnzpwyH+brz/njjz944403ePbZZ+ncuTPu7u788MMPZf7Y/rlfhUJxw6WZKrJfKDk7PnPmDPb29iQlJZXb1q38c8SGRqPB39+fTz75pExdDw8PAJ599lkGDRrEwYMHOXDgACtXruSDDz6gU6dOjBw5kq5du3LgwAEOHTrE6tWrmT17dqXO3P7pn+8PUGYYX0FBAXXq1Lnttq+/JxV9DyrD19eXr776yvL4+vG8FTu7spfY7sRFwX9+Fsq7OFyZ/ZZ3nCo7/PJG5EXH21S/fn0MBkOpdQwzMzNJTU21nGGFh4eX2g6UeVxRMTExZR5f309ERARCCMsZt7V8fHzw9fUlJSWF0NDQUj/Xz7L++OMPQkJCmDhxIs2aNSMsLIzU1NQ7vl+A1atXk5aWxooVK9i5cyd79uwp09bfj1NWVhZpaWk3POMFaNq0KdeuXcPR0bHMvv9+Jl6vXj0effRRVq5cSYcOHdi2bZtlW1BQEKNHj+aDDz5g+PDh/Pjjj+Xuq379+pw+fbpU2enTp28aX3n0ej0pKSlERERU6nk3U9H34O/q1atX5vN85swZy+8ODg7lHs9bHYfrXXtZWVmW7XFxcaXqOzo6VigZOjo6YjKZblkPSrrxCgsL0el0t71fW5MJ+zaFh4fTs2dP/vOf/3D69GkuXLjAG2+8QaNGjejUqRMAY8aM4fvvv+fnn38mMTGR5cuXk5KSUuo/uUaj4eLFi1y8eBGAlJQULl68WOrDCiXdED/++COJiYksXbqUxMREHnroIaDkgxYREVFmNIE1nn76aVavXs3//vc/EhMTiYuLY9OmTZar+KGhoaSmprJz506Sk5NZu3Ythw4duuP7PXv2LJ9//jlvvvkm7dq1Y+rUqfz3v/8tc5w+/fRTjh8/TlxcHHPnziUsLMzyfpSnU6dOtGjRgpdeeonDhw+TkpJCdHQ0S5cuJSEhAa1Wy6JFizh58iRpaWmcOHGCuLg4S/fEe++9Z3leTEwMp0+fLtV18Xfjx4/n8OHDrFmzhqtXr7Jp0yZ++OGHSo8TPnv2LE5OTjZfhPdW78E/jRw5kitXrvDRRx+RmJjIjz/+yM6dO2+5n1sdBxcXF1q0aMGXX35JQkIChw8fZu3ataXaCAoKQq1Wc/z4cfLy8m7YPRQcHExKSgqxsbHk5eWh1+tvGFeLFi1wcnJixYoVJCUl8f3335e5LyIoKIi4uDgSExPJy8u7a8MBZZeIDcyZM4d3332X559/HpPJRMeOHZk3b54lIQ8aNIikpCTee+89jEYjgwYNYsCAAWRkZFjaOHfuHM8884zl8bvvvgvApEmTmDJliqV88uTJ/PzzzyxYsAB/f3/eeeedUn1jDz74IL/88stt3zjz8MMP4+rqyrp161iyZAlubm40adKExx9/HIDevXszbtw4Fi5ciNFopFevXkyYMIFNmzbdsf1qtVrefPNNRowYQbdu3QAYN24cBw8e5D//+Y9lUVqAqVOnsmjRIpKSkmjevDmLFi0q9yv2dXZ2dnz44YcsX76cuXPnkp+fj6+vLx06dMDDwwN7e3vy8vJ44403yMnJwdvbm4EDB/LII48AJV+Z3377ba5du4ZKpaJHjx7861//KndfLVq0YMGCBaxcuZKVK1cSEBBQ6iJfRe3YsYOBAweWupg1d+5cUlNT+fTTTyvV1t/d6r3/p+DgYBYuXMiSJUv4+uuvadu2LVOmTGHhwoU33U9FjsMbb7zBW2+9xSOPPEJERARTpkwpda9CmzZtGDlyJK+88gr5+fll/l6u69u3L7/++itTpkyhsLCQOXPmWC7e/5OXlxdz587lo48+YuPGjfTu3dsyDPS6ESNGcOLECR577DE0Gg0rVqwgODj4pq/XJmw23kSqlIkTJ4p33nmnUs+50XC/vysuLhaDBw8W586du53waqx/DqGrrfLy8kS/fv1EcnJyqfJJkyaJFStWVFFU0p0mu0TugqKiIr766iuuXLnClStXWLp0KTExMQwePNjm+3JxcWHevHk3HIkh1Q5paWm8/PLLhISEWMqKiopISUmx6S3YUvUiu0TuAjs7Ow4cOMBnn32GwWCgfv36LFmyxOZ9j9fda5Pu3IuaNm1aZjiju7s7P/30UxVFJN0NCiFuMMZKkiRJqlZkl4gkSVINIRO2JElSDSETtiRJUg0hE7YkSVINIRO2JElSDSETtiRJUg0hE7YkSVINIRO2JElSDSETtiRJUg3x/8I6iAQg8rYFAAAAAElFTkSuQmCC", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAvgAAAFiCAYAAACDN6xbAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAQ6wAAEOsBUJTofAAAjHBJREFUeJzs3Xd8U+X+B/BP2rRpk7ZJdxqgLVB2EWQKMgpliuxVkCugTEEv4mXc60IERfFeRVERUVCWWGQjVFYRGbJkFGV3QNu0TelOkzbt+f3RXyOhg+6Mft6vV14kJ+c8+Z6Q5nzznO95HpEgCAKIiIiIiMgm2Jk7ACIiIiIiqjlM8ImIiIiIbAgTfCIiIiIiG8IEn4iIiIjIhjDBJyIiIiKyIUzwiYiIiIhsCBN8IiIiIiIbwgSfiIiIiMiGMMEnIiIiIrIhTPCJiIiIiGwIE3wiskoikQhLliwxdxgWZ8CAAZg7d26ttB0ZGQmRSIRdu3bVSvvmtmHDBohEIsTExNRYmyEhIQgJCamx9ioiMDAQU6ZMMT6uyn7duHEDDg4OuHbtWs0HSES1jgk+kRUqPmCXdYuKijJ3iDZrypQpJu+1QqFAx44d8fXXX6OgoKDE+kePHsWIESPg6+sLR0dHKJVKjBw5Evv37y+1/XPnzkEkEsHV1RW5ubmViu3XX3/FsWPH8K9//atK+0ZUrEWLFnj22Wf5I5rISonNHQARVd3y5cvh7+9fYnmjRo3MEE3dys3NhVhsnq8wqVSKr776CgCQmpqKjRs3YsaMGbh16xY+/PBD43qvv/463nvvPbRo0QIvvfQSGjVqhJSUFOzduxfPPvss9u/fj2eeecak7c2bN6Nhw4ZITEzEnj17MH78+ArH9dFHH2HAgAEIDAyskf2k+m3mzJl45plnEBMTw88UkZVhgk9kxZ555hm0b9/erDEIggCdTgdnZ+c6fV0nJ6c6fb2HOTg4YNKkScbHM2bMQIsWLfD5559j+fLlcHBwwLZt2/Dee+8hLCwMGzduNPkxsmjRIuzcubPEPhQUFGDbtm2YPn06Tp8+jS1btlQ4wU9KSsKBAwewbt26mtlJqvdCQ0Ph7u6O77//Hm+99Za5wyGiSmCJDpENe+utt2Bvb4/ffvvNZPnEiRMhk8lw+/ZtAH+X/Jw8eRLTp0+Hu7s7FAoFpk2bhqysLJNtAwMDMWLECBw4cAAdOnSAk5MTtm3bBgB48OABXn75ZTRs2BASiQQtWrTAp59+WiKuzz77DG3atIFUKoW7uzs6deqELVu2GJ9Xq9WYOnWqsR0/Pz8MHz7cpIa4tBr8ixcvYuDAgXB1dYWrqysGDRqEy5cvm6yzZMkSiEQiREdH4/nnn4dcLodcLsfUqVOh1Wor/R4DgLOzM7p27QqtVouUlBQAwNtvvw1PT098/fXXpZ5pGDlyJPr27Wuy7OjRo1Cr1Rg/fjzCwsJw8OBBPHjwoEIx/PzzzzAYDOjXr5/J8vz8fLzzzjto1qwZnJyc4OnpiR49euDQoUPGdR6t2S5WVv24wWDA4sWL4evrCxcXF4wePRpqtdpknVu3bmH06NFQKpVwcnJCw4YNERYWhoyMDOM6H330Ebp37w5PT084OzujY8eO2L59e4nXE4lEmDdvHrZs2YKWLVtCKpWiV69euHXrFgBg5cqV8Pf3h7OzM4YNG1biPXv4M9uuXTs4OTmhbdu2+Pnnnx/7vgLAvn370L17d8hkMsjlcowcOdL4t/OwtWvXomnTpnB2dkaXLl1w4sSJCrVf7LvvvkOnTp0glUrh4eGB0NDQEm2sX78eHTp0gLOzM7y8vDB58mQkJSVV6nUA4Pz58xg4cCC8vLzg7OyMxo0b44UXXjBZx8HBAb1798bu3bsr3T4RmRd78ImsWHp6OjQajckyOzs7eHh4AADefPNN7NmzB1OnTsXly5chlUqxY8cObN26FZ9++imCgoJMtp09ezY8PT3x7rvv4tq1a/jqq6+QmJhYol78zz//xKRJkzB79mzMmDEDLVu2RE5ODnr37o2kpCTMmjULDRo0wLFjx/DPf/4TaWlpePvttwEAX3/9NV555RW8+OKLmDdvHrRaLS5fvozff/8dEydOBACMHj0ad+/exUsvvQSVSgW1Wo1Dhw4hLi6uzFKBa9euoVevXnB3d8e///1vAMCXX36Jnj174vfff0erVq1M1h89ejSaNm2KFStW4OLFi1i3bh18fHzwwQcfVOn/Ijo6Gvb29lAoFLh16xZu3LiBF198ES4uLhVuY8uWLWjTpg3atGkDlUqF2bNnY/v27ZgxY8Zjtz116hR8fHzQoEEDk+VLlizBypUrMWfOHAQHByM9PR3nzp3DxYsX0b9//0rvJwAsXboUYrEY//nPfxAfH49Vq1bhzp07OHfuHBwcHJCXl4eBAwfC3t4e8+fPh6enJ+7du4d9+/YhPT0dcrkcALBq1SoMGzYMzz33HPLy8vDDDz9g7Nix2LdvH4YMGWLymseOHcOuXbvw0ksvIT8/H++99x5GjhyJiRMnYufOnfjXv/6F6OhorFq1CgsWLMA333xjsv3169fx3HPPYfbs2Zg8eTLWrVuH4cOH4/jx4+jevXuZ+7phwwa88MILePbZZ/Hhhx8iKysLn332GXr06IHLly/D19cXAPDNN99g5syZ6NGjB+bNm4e7d+9i2LBh8PDwqFDJ3Jtvvolly5ahZ8+eWLZsGUQiEU6fPo0TJ06gZ8+eAIB33nkHS5cuxYQJEzBjxgwkJiZi1apVOHfuHC5cuFDhs2jJyckYMGAAGjdujDfeeAMymQzR0dHYuXNniXU7duyIPXv2IDs7u1KfZSIyM4GIrM769esFAKXePD09Tda9dOmS4ODgIMydO1dISUkRfHx8hJCQEKGwsLBEe126dBHy8/ONy9955x0BgBAZGWlcFhAQIAAQjhw5YvI6S5cuFVxdXYW7d++aLJ81a5bg5OQkPHjwQBAEQRg+fLjwzDPPlLlvaWlpAgDhxx9/LPc9ACC8/fbbxscjRowQnJychJiYGOOy6OhoQSKRCKNGjTIue/vttwUAwowZM0zaGzlyZIn3rjSTJ08W3NzchJSUFCElJUW4fv26MH/+fAGAMGTIEEEQBGH37t0CAOHjjz9+bHvFcnNzBTc3N2Hp0qXGZUOGDBF69epVoe179OghdO3atcTydu3aCS+99FK52wYEBAiTJ08usbx3795C7969jY+PHTsmABD8/f2FrKws4/LvvvtOACBs2LBBEARB+OOPPwQAwtmzZ8t9Xa1Wa/I4Ly9PCA4OFvr27WuyHIDg5OQkxMXFGZd98MEHAgChWbNmgk6nMy6fMGGCIJFIhLy8PJP9AyDs2bPHuOzBgweCp6enEBISYlxW/HcQHR0tCIIgZGVlCXK5XJg7d65JPHfv3hWcnZ2FRYsWGeP28fEROnbsaPK6a9euFQCYvIeluXnzpmBnZyeMHTtWKCgoMHmu+O80OjpasLe3Fz766COT50+fPi2IRCLhyy+/NNnfh/8/H92vnTt3CgCE5OTkcuMSBEHYsmWLAEC4cOHCY9clIsvBEh0iK7ZmzRocOnTI5LZjxw6Tddq1a4c33ngDn3/+OZ555hlotVp8++23EIlEJdqbOXOmSTnJnDlzAAAHDhwwWa9Zs2Ylyku2b9+O3r17w9XVFRqNxngbMGAAdDodfv/9dwCAQqHAtWvXcP369VL3ydnZGY6Ojjh48CBycnIq9D4UFBTgl19+wahRoxAQEGBcHhgYiJEjRyIiIqLECDezZs0yedyzZ0+kpqYiMzPzsa+XmZkJb29veHt7o2XLlvjf//6HgQMH4ttvvzU+DwCurq4Vih8oKgPJzMw0qbkPCwvDiRMncO/evcdun5qaCnd39xLLFQoFfv/9d9y/f7/CsTzO5MmTTXpzJ06cCHd3d+PnpLiHfu/evcjLyyuznYd7nNPS0pCRkYGePXvi4sWLJdYdMGCASU94165dja8tkUhMluv1eiQmJpps7+/vj6FDhxofu7u7Y8KECfj111/L/JwdOnQIGRkZGDdunMln2tXVFe3atUNkZCSAonKX5ORkzJ49Gw4ODsbtp0yZYnwvyrNr1y4UFhbirbfegp2d6WG5+O90586dEAQBo0aNMoklKCgIfn5+xlgqQqFQGNssLCwsd93iz9SjZwqJyLIxwSeyYl27dkW/fv1Mbr169Sqx3n/+8x+0bNkS586dw/vvv4/GjRuX2l6zZs1MHnt6esLd3b3E+NmlbX/r1i3s27fPmPgW30aNGgUAxtr0RYsWwWAwoFWrVmjdujVeffVVnD171tiORCLBBx98gA0bNsDLywuhoaH4+OOPjduXJiUlBVqtFi1atCjxXHH50KMJyqOjDxUnMmlpaWW+TjGZTGb8QXXixAkkJCTg4MGD8PHxAQC4ubkBQInrF8qzefNmBAUFwc7ODrdv38bt27fRpk0b2NvbY+vWrRVqQxCEEsuWLl2KO3fuwN/fHx07dsTrr7+OP//8s8JxlebRz4lYLEZgYKDxc9K4cWPMnz8f7777Ljw9PfHss8/iq6++KvF+7Nu3D0899RScnJzg4eEBb29vfPnllyZ1+sUe/f8qTpwfLX8pXv7o/+Oj5WjF+1FYWFjmD6jiGv9evXqV+FyfOXPG+JmMjY0t9X1xcHBAkyZNSm37YXfv3oW9vT1atmxZ5jq3bt1CYWEhmjRpUiKWhISEcv8+HtW7d2+MHj0aM2fOhLe3N8aOHYuNGzeW+mOs+DNVWocAEVku1uAT1QM3b97E3bt3AaBGxsgvrda3sLAQgwYNwmuvvVbqNm3atAEAtGrVCjdu3MD+/ftx8OBB/PDDD/jkk0+wdOlSvPnmmwCAefPmYdiwYdi1axd++eUXLF68GMuWLcPRo0fRrl27ascPAPb29qUuLy1JfpRYLC5xMevDihO1ir7X6enpOHDgAPR6fYkkESiqzV+4cGG5bXh6epb646RXr164c+cO9uzZg19++QVffvklPvzwQ6xduxZTp04FUHbyVlBQUOb79Dj//e9/MXXqVOzevRsRERF46aWXsHz5cpw+fRoNGjTAiRMnMGzYMPTq1QtffPEF/Pz84ODggPXr15tccF2srDiq8//4OMW921u2bIG3t3eJ5+ty5KjCwkLY29vjwIEDpf5/lXb2piwikQjbt2/H77//jr179yIiIgLPP/88PvroI5w8edLk7EzxZ8rLy6v6O0FEdYYJPpGNKygowOTJk+Hr64uxY8fiv//9L8aOHYvQ0NAS6966dct4QR9QVPaRlpZmUvZSlqZNmyI3N7fcxLeYTCbDuHHjMG7cOOTn52PMmDFYunQpFi5caCy3aNKkCebPn4/58+fj7t27aNeuHT766CNs3LixRHve3t6QSqW4ceNGiedu3LgBmUxWpwlK8+bN0bx5c+zevRurVq2CTCYrd/3t27dDr9fjm2++Mfb+F7t48SLef/99/Pnnn2jdunWZbbRs2bLM0U48PDwwZcoUTJkyBTk5OQgJCcFbb71lTPDd3d2Rnp5eYrvY2NhSe6CLe7aLGQwGxMTElLhoNzg4GMHBwXj99dfx+++/46mnnsKaNWvw7rvv4qeffoKTkxMiIiJMSmzWr19f5j5WR2mj3ty6dQt2dnZlXgTbtGlTAICfn1+5s9EW/33cunXL5Axafn4+oqOjH/ujtGnTpigoKMD169cRHBxc7jrNmjWrsTHpu3btiq5du2LZsmX48ccfMX78eGzbtg0vvviicZ3ii8dL++FJRJaLJTpENm7FihU4f/481q1bh/fffx9PPvkkXnzxxVLLR7766isYDAbj488//xwAMHjw4Me+zpgxY/Drr7+WWgus0WiMPaqpqakmzzk4OKBNmzYoLCxEXl4etFptiRlcAwMDIZfLodPpSn1te3t7DBgwADt27EBcXJxxeVxcHHbu3Gkc0aUuvf3229BoNJg5c6bJe1ps9+7dOHbsGICiHuJWrVrhhRdewJgxY0xuixYtglgsxubNm8t9vW7duiElJcVk/4GS77dMJkOzZs1M3sumTZvizJkzJiUa+/btK7N05bvvvkN2drbx8ZYtW5CWlmb8nGRmZpbY5+DgYIjFYuPr2tvbQyQSmVwbERMTg127dpW7n1UVFxeHvXv3Gh+npaVh69at6NmzZ5k/wAYMGAA3Nze89957pf4fFpd9derUCd7e3lizZg3y8/ONz2/YsKHUH06PGj58OOzs7PDOO++UqIkv/rsZOXIk7OzssHTp0hLbFxYWVng4VaBo3x89w1E8n8ajf2MXLlzAE088YXUj6BQPh1t88/b2Rt++fUsduvTq1auYOHEiVCoVHB0d4evri1GjRuHIkSMl1s3Pz4eXlxfs7OxK/K3VB/PmzeOkZ1aCPfhEVuznn38utQwkJCQEDRs2xNWrV7F06VLMnDnT2LtaPNb2ggULsGbNGpPtcnNz0b9/f4wePdo4TObAgQPL7b0stnDhQuzevRsDBw7ECy+8gPbt2yMzMxOXL1/GTz/9hKysLIjFYgwYMABKpRLdu3eHUqnE9evXsXr1agwZMgSurq64dOkSQkNDMXbsWLRu3RqOjo7YtWsX4uPjERYWVubrL1u2DIcOHUKPHj0we/ZsAEXDZIrFYixbtqwS72rNmDhxIi5fvowPP/wQFy5cwIQJE9CoUSNoNBrs378fx48fx88//4z4+HgcP368zNImuVyOHj16YOvWrVi+fHmZrzdkyBCIxWIcPnzYZDzz1q1bo3fv3ujUqRM8PDxw4cIFbNu2zXgBNQBMmzYN27dvx6BBgzBu3DjcuXMHmzZtMvZglxZTr169MHnyZOMwmcHBwXjuuecAFI3nP3fuXIwZMwYtWrRAQUEBNm3aBJFIhNGjRxvj/d///odBgwZh4sSJSE5Oxueff46goCBcuXKl0u/347Ro0QKTJ0/GSy+9ZJyfICMjA++++26Z28jlcqxevRqTJ09Gp06dMH78eHh6eiImJgZ79uzBiBEjsGzZMjg4OGDZsmWYOXMm+vbti3HjxiE6Ohrr16+vUA1+s2bNsGjRIrz//vsICQnBiBEjIBaLcfr0abRt2xb/+c9/EBQUhKVLl+KNN97AnTt3MHToUMhkMty5cwc//fQTXn/9dUybNq1C78V3332HL774AiNGjEDTpk2Rk5ODdevWwc3NzWRm5fz8fBw/fhyvvPJKhdq1NM7Ozjh69CgA4P79+3j33XcRGhqKixcvGs+U7N69G+PHj0dwcDCWL1+Opk2bIiUlBTt27MCAAQPw4MEDkwulIyIijD+at27dikWLFtX9jhFVhPkG8CGiqipvmEwAws6dO4X8/HyhQ4cOQkBAgJCZmWmy/fLlywWRSCQcPnzYpL0TJ04IL774oqBQKAQ3Nzdh6tSpQkZGhsm2AQEBwvDhw0uNKyMjQ1i4cKHQpEkTwcHBQfDx8RF69eolrFq1yjjc31dffSX06tVL8PT0FCQSidC0aVNhwYIFxtfRaDTCnDlzhJYtWwoymUyQy+VC165dSwybiUeGyRQEQbhw4YLQv39/QSaTCTKZTBgwYIDwxx9/mKxTPExmWlpaqe9p8VCCZZk8ebIgl8vLXedhv/zyizBs2DDB29tbEIvFgo+PjzBixAjhwIEDgiAIwkcffSQAEI4fP15mG//9738FAMLJkyfLfa2hQ4cKgwYNMlm2bNkyoUuXLoJCoRCcnZ2Fli1bCsuXLzcZzrH4NRo0aCBIJBLh6aefFs6fP1/mMJnh4eHCwoULBW9vb0EqlQojR44U4uPjjevdvXtXeOGFF4SmTZsKTk5OgoeHh9CnTx/j563YN998IzRr1kyQSCRCy5YthfXr1xv/fx4GQPjnP/9psqx4KM7169ebLC/+f3z4/734M/vzzz8Lbdu2FSQSidCmTRth7969pW776Gfg8OHDQr9+/QQ3NzfB2dlZaNasmTB9+nTh6tWrJut98cUXQuPGjQWJRCJ06tRJ+PXXX0u8h+VZu3at0K5dO0EikQgeHh5CaGio8Ntvv5mss23bNqFbt26CVCoVXFxchNatWwvz5s0zGR72ccNkXrx4UZgwYYLg7+8vSCQSwcfHR3j22WeF8+fPm7zWgQMHBJFIVGLoW2vw9ttvCzKZzGRZbGysIBKJhDlz5giCIAiJiYmCm5ubEBoaKuj1+hJtHD16VMjJyTFZNmHCBEGhUAgdO3YU2rZtW3s7UAsKCwtNhpStin/+859CQEBAzQREtYoJPhGVmhSR9YmMjBTs7e2tMiGrTeX9KKWyjRw50mQOCWtSWoIvCILg4+NjnIdj6dKlAgDhypUrFWozOztbkMlkwrRp04TPPvusUtvWpOLv69OnTwt9+vQRnJ2dhYCAAOGbb74xWW/y5MlCmzZthP379wtPPPGE4ODgIISHhwuCIAinTp0S+vTpI0ilUsHNzU2YMGGCkJSUZLJ9fHy8MHToUMHZ2VlQqVTCBx98wATfirAGn4jIRvTu3Rt9+/bFf//7X3OHQlbuxo0b2Lt3b6k1/9YqMzMTqampUKlUAIDjx49DpVKhbdu2Fdp+165dyMnJwcSJEzFu3DiIxeJSR3yqK2FhYejfvz927tyJPn364MUXX8TBgwdN1klISMArr7yCV199FQcPHkT79u1x+vRphISEQC6XY9u2bVi7di3OnTuH4cOHm2w7fPhwnDt3Dl9++SW++OIL7Ny5E9u3b6/LXaRqYA0+EZEN+eWXX8wdAtmAFi1amFwwbK2KL46+f/8+XnvtNRQUFGDMmDEAgPj4+BLzK5Rny5YtaNCgAXr37g07OzuEhoZiy5YteO+998wyT8Dzzz+Pf//73wCAgQMH4u7du3jnnXcwaNAg4zppaWk4cOCAcWI4AHjxxRfRqVMn7Nixwxh327ZtERwcjJ9//hnPPPMMDh48iPPnz+PIkSPGSQ1DQkLQqFEjeHh41OFeUlWxB5+IiIhsTk5ODhwcHODg4IDGjRvj2LFjWL16NQYOHGhcp6KJuUajwS+//ILx48cbZxueOHEi4uLi8Ntvv5W7rcFgqPTtcTMMA0UjKz1s9OjRuHDhgsnIVJ6enibJvVarxcmTJzF27FgUFBQYX6958+Zo1KgRzp07BwD4/fffIZfLTWYsl8vlFRoGmSwDE3wiwpQpUyAIgnGoPCJbUpvDb5LlcnZ2xrlz53D+/HnExMRAo9GYjB7VoEGDCg91+eOPP8JgMGDIkCFIT09Heno6+vTpA4lEUm6ZTkxMjPFHRmVuD4+EVZbimbOL+fr6Ij8/32TWbl9fX5N10tLSUFBQgFdffbXEa8bFxRmHxk1MTCx1crdH2yPLxRIdIiIisjl2dnbo1KlTmc+HhITgyJEjuHbtmnGm7bIUJ/GlTRAYHh6OTz/9FA4ODiWeU6lUxl7xyqjIxHzJyclo0KCB8XFSUhIcHBxMtn30DIVCoYBIJMJ//vMfjBgxoszX9fPzQ0pKSonnk5KSKroLZGZM8ImIiKjemTZtGlauXIlXX30V+/fvL5GgR0ZGokuXLkhJScGpU6cwa9YsjB8/3mSdS5cuGS9gHTp0aInXcHR0LPdHRnXs3LkTTz75pPHxTz/9hI4dO5Y7qZ9MJkO3bt3w119/lTs/SJcuXZCRkYGjR48ay3QyMjJw+PBh1uBbCSb4REREVO8olUp8//33GDduHJ5++mnMmTMHTZo0gUajwa5du7B582akpqYae+8XLFhQYuKyHj164P3338eWLVtKTfBr0/fffw9nZ2d06NABP/zwA3799Vfs37//sdutXLkSffv2xfjx4xEWFgZ3d3fcv38fhw4dwtSpUxESEoJBgwahQ4cOeO655/DBBx9AoVDg/fffh5ubWx3sGdUE1uATERFRvVQ8FGRQUBAWL16Mvn37YubMmcjKysKhQ4cgl8uxZcsWPP3006XOSiwWizFx4kTs2bMH2dnZdRr71q1bERERgREjRuDo0aNYu3atyUzEZenevTt+++03ZGdnY+rUqXjmmWewdOlSSKVSBAUFASgq7dm9ezc6duyImTNnYtasWRg2bJhxBCKyfCJBEARzB0FEREREj7dhwwZMnToVKSkpFarVp/qJPfhERERERDaECT4RERERkQ1hgk9mtWTJEohEIuPN29sbffv2xYkTJ0qse/XqVUycOBEqlQqOjo7w9fXFqFGjcOTIkRLr5ufnw8vLC3Z2dhUe57imXbp0CUuWLIFWq62Tdjds2ACRSGQyBjIREdmW4nlLWJ5D5WGCT2bn7OyM06dP4/Tp0/jyyy+RmpqK0NBQREVFGdfZvXs3OnfujJs3b2L58uU4fPgwvvjiCzg7O2PAgAHIyMgwaTMiIgKpqakQBAFbt26t610CUJSIv/POO7WS4NdGu0RERGQbOEwmmZ2dnR2eeuop4+MuXbogMDAQa9aswerVq6FWq/H888+jR48e+Pnnn+Ho6Ghcd/To0Zg2bVqJ8Yu3bNkChUKBpk2bYvPmzVi0aFGd7U9V5ObmwtnZ2dxhEBERkQ1gDz5ZHH9/f3h7eyM6OhoA8PXXXyMzMxMff/yxSXJfrE+fPpBKpcbHOTk52LNnD8aMGYMpU6bg6tWruHr1ap3FD/w9ygEAeHt7QyQSITAw0PicSCTC6dOn0b9/f8hkMixYsAAxMTEQiUTYvn27SVvz5s0z2basdovdu3cPgwcPhkwmQ7NmzfD999/X6r4SERGRZWGCTxYnMzMTqampUKlUAIDjx49DpVKhbdu2Fdp+165dyMnJwcSJEzFu3DiIxWLjRCV1ZciQIXjjjTcAAAcPHsTp06exc+dOk3UmTpyIvn37Yt++ffjHP/5RY+0+99xzGDBgAHbt2oUnn3wSU6ZMwV9//VUDe0VERETWgCU6ZBEMBgMA4P79+3jttddQUFBgnFAjPj4e/v7+FW5ry5YtaNCgAXr37g07OzuEhoZiy5YteO+99yASiWol/kd5e3ujadOmAICOHTuWejHUrFmzTEqHYmJiaqTduXPn4qWXXgJQNKHJ/v378dNPPxl/GBAREZFtYw8+mV1OTg4cHBzg4OCAxo0b49ixY1i9ejUGDhxoXKeiiblGo8Evv/yC8ePHw86u6OM9ceJExMXF4bfffit3W4PBUOlbYWFhlfd7yJAhVd62PAMGDDDel8lkCAgIwP3792vltYiIiMjyMMEns3N2dsa5c+dw/vx5xMTEQKPRYM6cOcbnGzRoUOGhLn/88UcYDAYMGTIE6enpSE9PR58+fSCRSMot04mJiTH+yKjM7YUXXqjyfvv6+lZ52/IoFAqTx46OjtDpdLXyWkRUMatXr0anTp0gkUgQFhZW4+0XX8Pj4uJivM2aNavS7SQmJmLYsGFQqVQQiUS4fv16uev/+OOP6N69O6RSqclgCcUuXbqE3r17Q6FQwM/PD/PmzTOesb158yZGjBgBX19fKBQK9O7dG+fPn690zI9z7Ngx9OnTB3K5HEqlssbbJ7JELNEhs7Ozs0OnTp3KfD4kJARHjhzBtWvX0KZNm3LbKk7iQ0NDSzwXHh6OTz/9tMSIOwCgUqlw7ty5SkaOao1D/OhZCScnJwBAXl6eyfK0tLQqvwYRWQaVSoU33ngDhw8frtW5KjQajfG7pDwikQiCIJRYbmdnh0GDBuGNN95A165dH9uOh4cH5s2bh1u3bmHv3r0lnh83bpxxvpLk5GQMGDAAq1evxrx585Ceno5nnnkG69atg7u7O7744gsMHjwYMTExkMlkFdvhCpDJZHjhhRcwadIkvP766zXWLpElY4JPFm/atGlYuXIlXn31Vezfv79Egh4ZGYkuXbogJSUFp06dwqxZszB+/HiTdS5duoRXX30VBw8exNChQ0u8hqOjY7k/MqqieMSfivae+/j4wMHBweSC2Ly8PBw/frxa7RKR+Y0aNQpA0XfRown+uXPn8Nprr+Hq1atQKpVYvny5cf265uvra7yGpyL69esHoGiEr0cJgoDY2Fg8//zzEIvFUKlUGDx4sHFUsy5duqBLly7G9V9++WUsWrQI169fR8eOHQEAmzZtwooVK3D//n20bdsWa9aseWxHz6OKXycyMrJS2xFZM5bokMVTKpX4/vvvcfz4cTz99NP47rvvcOLECezcuROTJ09Gv379kJ+fb+y9X7BgAUJCQkxuc+fOhY+PT52OptOqVSsAwOeff47ff//9sUN12tnZYdSoUVi9ejW+++477N+/H8OHDy/Ry1bZdonIciUmJmLQoEGYP38+NBoNNmzYgGnTplV55KugoCCoVCqEhYXh3r17NRxt5YhEIsybNw/r169HXl4e4uLisH//fgwaNKjU9c+dO4fCwkIEBQUBAPbu3Ys33ngDW7duRWpqKiZNmoShQ4eWOMtJRCUxwSerMHz4cJw7dw5BQUFYvHgx+vbti5kzZyIrKwuHDh2CXC7Hli1b8PTTT6NJkyYltheLxZg4cSL27NmD7OzsOon5ySefxJIlS7Bp0yZ079691DMHj/rss88QEhKCV155BTNnzsSgQYMwcuTIardLRJZp48aN6NevH0aMGAF7e3t07doVI0eORHh4eKXa8fLywrlz5xATE4MrV65AJpNh6NChKCgoqKXIK2bo0KHYtWsXpFIpAgIC8NRTTxlHSHtYcQK/dOlSyOVyAMCXX36JRYsWoW3btrC3t8fMmTMhEolw5syZut4NIusjEBERUZ14++23hfHjxxsfz549W5BIJIJcLjfeZDKZMGvWLEEQBOGzzz4TAJR5O3bsWKmvo9PpBAcHB+Gvv/4SBEEQNm/ebPIaAEwenzhxokQbAIzbP8769euFrl27mixLTU0VXFxchLVr1wp5eXlCUlKSMHjwYGHBggUm66WnpwsdO3YUXnnlFZPlrVq1EmQymUmczs7OwpYtWwRBEITRo0eX+9486tixY4Kvr2+F9ofI2rEHn4iIyEz8/f0RFhZmHPUrPT0d2dnZ+PLLLwEUzWshCEKZt5CQkFLbFYlEJhfSTpw40eQ1AJg87tGjR43v2927dyEIAqZPnw4HBwf4+Phg6tSp+Pnnn43rZGRkYODAgejcuTNWrVpV4r1ZvXq1SZxarRYTJkwAAGzfvr3c94aoPmOCT0REVMsMBgN0Op1x/gydTof8/HxMmjQJBw4cwN69e2EwGJCXl4fff/+90jX4xdsUFhYiPT0dc+fORVBQEJo3b17pWHU6nfEi/ry8POh0ujIT5oKCAuO+CIIAnU5nrJFv0aIFxGIx1q9fj4KCAjx48ADfffcd2rVrB6Bo1vJBgwahdevW+OKLL0q0PXv2bKxYsQKXL1+GIAjIzs7G3r17kZWVVan9KX6/i+PS6XTQ6/WVaoPI6tTxGQMiIqJ65+233y5RQjJ58mRBEATh/PnzQt++fQUPDw/B09NT6NOnj/DHH39Uqv0tW7YIjRs3FqRSqeDr6yuMHj1auHv3bpnrl3f4fzROAEJ0dLQgCIKwfPlyYdCgQcZ1169fX2Ld3r17G5+PjIwUunbtKri5uQne3t5CWFiYkJycLAiCIGzYsEEAIEilUkEmkxlvv/76q8l+tWvXTnBzcxOUSqUwduxYITMzs1LvzbFjx0rEGBAQUKk2iKyNSBB4HouIiIiIyFawRIeIiIiIyIYwwSciIiIisiFM8ImIiIiIbAgTfCIiIiIiG8IE/xGcFZSIiGoKjylEZA5M8B9RPPYvERFRdfGYQkTmwASfiIiIiMiGMMEnIiIiIrIhTPCJiIiIiGwIE3wiIiIiIhvCBJ+IiIiIyIYwwSciqgciIyPRpEkThIaGon///jh58iQ2bNiA1q1bIzQ0FEOGDMGff/4JAFiyZAl2794NAPjoo4/www8/4M6dO2jXrh2Cg4ONbf7555/o1q0bunbtivDwcLPsFxERlcQEn4ionpgxYwaOHDmCLVu2YNGiRcjOzsZbb72FI0eOYPXq1Zg1axYKCwvx2muvYdWqVUhLS8PBgwcxfvx4+Pn54bfffkPDhg2N7a1fvx4ff/wxTpw4gTVr1phxz4iI6GFM8C1QtFaP39KyEK3VmzsUIrIACem52Hc5AetO3MW+ywlISM+tVnve3t4YOnQoZDKZcVnjxo3Rpk0bxMbGwtXVFUOHDkW/fv3wr3/9CyKRCFKpFK6uribtNGvWDBkZGcjNzS3xnKWaMWMGGjRoADc3NwQGBuK9994rc93jx48jODgYUqkUnTt3xuXLl+swUqKKMWhyobuTDoOmet8LVHlp6gTERV1BmjrB3KGUwATfwkRr9diUqMHmhFRsStQwySeq5xLSc7H1bBx++SsJN5Oy8MtfSdh6Nq7aSb5SqURSUlKZy0JCQnD79m307du3zDZCQkIwZ84ctG3bFtOnT69WPHVl3rx5uH37NjIzM3HixAls2rQJP/74Y4n1UlNTMXz4cCxcuBBpaWmYMGEChg0bBr2e38lkOQyaXGSfUyPnrBrZ59RM8utQmjoBV49E4OrRCFw9EmFxST4TfAsTr89Dgi4fLWROSNDlI16fZ+6QiMiMLsamIfaBFq18XdHMxxWtfF0R+0CLi7Fp1Wo3MTERSqWyzGXvv/8+Xn75Zaxdu7bMNt58800cOHAAf/31Fz7++ONqxVNXWrduDWdnZ+NjOzs73L59u8R6O3bsQFBQEJ5//nlIJBK8+uqrKCwsxOHDh+syXKJyGTL0KEjXw8FXioJ0PQwZ/AFaV7I0GmSlauDZ0B9ZqRpkaTTmDskEE3wL00DiCJWTA27k6KByckADiaO5QyIiM1Jn6uAktoPYvujrWmxvByexHdSZuiq3mZqaiv379yMnJ8e4LDY2FtevX0dAQADOnDkDHx8fLFmyBOHh4dBqtaW2IwgCFAoFnJ2dkZdnPZ0R//73vyGTyeDv74+cnBxMmjSpxDpRUVFo37698bFIJMITTzyBqKioOoyUqHxiuQT2Cgnyk7SwV0gglkvMHVK94erlBVdPL6Tej4OrpxdcvbzMHZIJsbkDIFONpRJM8vNCvD4PDSSOaCzlHytRfaZ0c8KV+AwYCgohtreDoaAQOkMhlG5OlW5r7dq1OHToEOzs7LBixQrcunULS5cuxddffw1nZ2d88cUXEIlEePfdd7Fu3TqIxWK89NJL+PTTTzFt2jSEhYXh0qVL6NevH3bs2IHXXnsNw4YNQ2FhodWU6ABFZyfee+89nD9/Hrt27YK7u3uJdbKzs0ssVygUyMrKKrPdlStXYuXKlSbLmjVrVjNBW6lorZ7Hs1ok9nKGS2clDBl6iOUSiL2cH78R1Qh3pQptQwciS6OBq5cX3JUqc4dkQiQIgmDuICxJ//79cejQIXOHQVRpqampyMzMhJubGzw9Pc0dDtWQ4hr82AdaOIntoDMUIsBDigld/KFS8GBeXe+++y7S0tLwv//9z2T5P//5T2i1Wnz99dfGZc8++yx69uyJRYsWVbj9+nxMKb6mLEGXD5WTAyb5eTHJJ6ojLNEhsgGpqam4ePGi8ZaammrukKiGqBTOmNDFHwNa+aK5rysGtPJlcl+DDAYD7ty5U2J5cHAwLl26ZHwsCAKuXLliMg8AlY/XlBGZDxN8IhuQmZmJzMxMeHt7G++T7VApnPFsOxWm9WyCZ9upmNxXUVpaGjZu3IjMzEwUFhbi5MmT+PLLL9GvX78S644aNQq3bt3Cpk2bkJeXh1WrVgFAqetS6XhNGZH5MMEnsgFubm5wc3NDSkqK8T4RmRKJRFi/fj0CAgIgl8vx4osv4rXXXsPcuXMBAC4uLjhx4gQAwNPTE7t27cKKFSsgl8uxefNm7NmzBxIJS0xKo9XG4EHaaWi1McZlxdeUPafyZHkOUR1jDf4j6nO9JFk31uATWZ76cEzRamMQn7ANer0aEokSDVTjIZUGmjssonqNo+gQ2QhPT08m9kRU53T6ROj1ashkQcjJuQ2dPpEJPpGZsUSHiKgeiIyMRJMmTRAaGor+/fvj5MmT2LBhA1q3bo3Q0FAMGTIEf/75JwBgyZIl2L17NwDgo48+wg8//GBsZ/78+Zg1a5bx8TvvvIN+/fohLCysbneILIaTxA8SiRI5ObchkSjhJPEzd0hE9R578ImI6okZM2Zg8eLFSElJwciRIxEWFoa33noLYWFhiI6OxuTJkxEZGYnXXnsNw4cPR69evXDw4EFjiUlaWhpu374NlapovOeTJ09CJpNxdtd6TioNRAPVeOj0iXCS+LH3nsgCsAefiMjSZdwHonYAp1cX/Ztxv1rNeXt7Y+jQoZDJZMZljRs3Rps2bRAbGwtXV1cMHToU/fr1w7/+9S+IRCIAwJo1azBjxgzjNgcOHEBMTAz69OmDb7/9tloxkXWTSgPh4d6NyT2RhWCCT0RkyTLuAxc2ADf2A8nXi/69sKHaSb5SqURSUlKZy0JCQnD79m307dsXAKDX63Hjxg2TceCTk5OhUqlw+PBh/PDDD0hPT69WTEREVDOY4BMRWbJ7Z4G0aMCnDeDdoujftOii5dWQmJgIpVJZ5rL3338fL7/8MtauXQsA+P777/Hcc8+ZrC+Xy9G7d2/Y29ujY8eOiI2NrVZMRERUM5jgExFZsqwEQOwM2DsUPbZ3KHqclVDlJlNTU7F//37k5OQYl8XGxuL69esICAjAmTNn4OPjgyVLliA8PBxarRa3b9/Gf//7X0yePBk///wzIiIi0KVLF1y9ehUA8Ndff6Fhw4bV2lUiInPJi41FzpnfkWcjHRW8yJaIyJK5qoCEP4CC/KLkviAfMOQWLa+ktWvX4tChQ7Czs8OKFStw69YtLF26FF9//TWcnZ3xxRdfQCQS4d1338W6desgFovx0ksv4dNPP8UHH3wAAIiJicGKFSswcOBA5OXlYerUqdi8eTOeeeYZDtNKRFYpLzYWaeHbYVAnQqz0g/vYMXAMCKjV10xTJyBLo4GrlxfclZX/Pn8cTnT1iPowKQkRWZHiGvy06KKee0Mu4N4Y6DgFkLPH3NLxmEJk+XLO/I707eFwbBqEvDu3oRgzFrKnutba66WpE3D1SASyUjVw9fRC29CBNZ7kswefiMiSyRsWJfP3zhaV5biqgEZdmNwTEdUQBz8lxEo/5N25DbHSDw5+ysdvVA1ZGg2yUjXwbOiP1PtxyNJomOATEdU78oZM6ImI/l+0Vo94fR4aSBzRWCqpdnuOAQFwHzsG+YlqOPgpa708x9XLC66eXki9HwdXTy+4ennV+GtY1UW2q1evRqdOnSCRSErMmhgVFYWnnnoKUqkUrVu3xtGjR80UJREREdmM1DtA9K9F/5LZRWv12JSoweaEVGxK1CBaq6+Rdh0DAiB7qmutJ/cA4K5UoW3oQLTtO7BWynMAK0vwVSoV3njjDUyfPt1keX5+PoYOHYphw4YhLS0Nb7/9NkaOHInk5GQzRUpERERWL/UOcPH7v29M8s0uXp+HBF0+WsickKDLR7w+z9whVYm7UgX/4CdqJbkHrCzBHzVqFEaMGAGvR05lREZGQqvVYvHixZBIJBg/fjyCg4MRHh5upkiJiCxLZGQkmjRpgtDQUPTv3x8nT57Ehg0b0LJlSxSPtTBo0CAAwJQpU0zGvO/SpQtWrFiBhQsXIiQkBB4eHggJCcHgwYNx6tQptG7d2rgtkU3JjC+6ebf8+z6ZVQOJI1RODriRo4PKyQENJI7mDski2UQNflRUFNq2bQs7u79/r7Rv3x5RUVFmjIqIyLLMmDEDixcvRkpKCkaOHGksdTx06BAGDBhgsq5arUZBQQGSk5NRUFAAAPjwww8BFP0QOHjwIAAgMzMTFy5cwMiRI+twT4jqiFuDolvK9b/vk1k1lkowyc+rRmvwbZFNJPjZ2dlQKBQmyxQKxWNnVVy5ciVWrlxpsqxZs2Y1HR4RUbWoc9S4lHwJydpk+Eh90N6nPZSyqo/y4O3tjaFDh0Imk2HmzJn46quvSiT4PXr0wMmTJ3Hjxg08++yzZbbl5uZW5TiILJ5nU6DD80U9924Nih6T2TWWSpjYP4ZVleiUxcXFBRkZGSbLMjIy4OrqWu52CxYsQHJysslNKpXWZqhERJWizlEj/EY4jt07htvpt3Hs3jFsv7kd6hx1tdpVKpVISkqCn58fnJ2dER0dbfL80KFDsW/fPpw4cQI9e/as1msRWTXPpkDjXkzuyarYRIIfHByMq1evorCw0Ljs0qVLCA4ONmNURETVdyn5Eu5n30dz9+ZoqmiK5u7NcS/rHi4lX6pWu4mJiVAqi84CzJkzB59//rnJ8w0bNkRMTAzEYjHEYps42Utks7TaGDxIOw2tNsbcoVikvNhY5Jz5HXmPqeywJVaV4BsMBuh0OhgMBhQWFkKn0yE/Px8hISFwdnbGhx9+CL1ej/DwcFy9ehVjx441d8hERNWSrE2GxF4CsV1Rki22E0NiL0GytuqjhKWmpmL//v3IyckBAHTr1g1XrlxBbm6uyXrDhg3DxIkTqx48EdU6rTYG8QnbkJDwI+ITtjHJf0RebCzSwrcjfXs40sK315sk36oS/GXLlsHZ2RnLly9HeHg4nJ2dMX36dDg4OGDPnj3YuXMnFAoF3nrrLezYsQM+Pj7mDpmIqFp8pD7QF+hhKDQAAAyFBugL9PCRVv77be3atQgNDUVYWBhWrFgBmUxmfO65555DUlKSyfqTJk1Cv379ym0zKioK/fr1w/nz5zF06NBKx0RE1aPTJ0KvV0MmC4Jer4ZOn2jukCxKfqIaBnUiHJsGwaBORH5i9cobrYVIKB4fjQAA/fv3x6FDh8wdBhERgKIa/O03t+Ne1j1I7CXQF+jRyLURxjQfU60Lbalu8JhSJC4zDuocNZQyJfzd/M0djk0p7sHX69WQSJRooBoPqTTQ3GFZjOIefIM6EWKlH9zHjqmTyazMjYWVREQWTClTYkzzMTU6ig5RXYrLjMP2m9uRpE2Cr9QX3VXdYSeyY7JfQ6TSQDRQjYdOnwgniR+kuQVA0q8c9ef/OQYEwH3sGOQnquHgp6wXyT3ABJ+IyOIpZUoMasyJpMg6qXPUSNImIUgRhMspl5GYkwg7kR18pb4Y03wMk/waIJUGFvXaF8+8WzysZ4fnmeSjKMmvL4l9MauqwSciIiLropQp4Sv1xe3025CKpcg15CJIEYQkbVK1h3ulR3DmXfp/7MEnIiKiWuPv5o8xzcdAnaNGoVCIUwmncDv9Nnylviw1q2mceZf+HxN8IiIiqlX+bv7GUhyVi4oX3NYWzrxbu1LvWM17yxIdIqJ6IDIyEk2aNEFoaCj69++PkydPYsOGDWjdujVCQkIwdepU47r5+flQKBS4cuWKcZmHhwdCQkIQGhoKtVqNU6dOoXXr1hg0iNcGUOX4u/mji18XJve1hTPv1o7i6xuKb6l3zB1RuZjgExHVEzNmzMCRI0ewZcsWLFq0CNnZ2XjrrbcQGRkJV1dXREVFASj6MTB8+HDs2bPHuG2XLl0QGRmJV199FV9//TWCg4Nx4cIFc+0Kkc0xaHKhu5MOgyb38SvXtdQ7QPSvFp/U1ioru76BJTpERBYuPzER2j8uwZCcBLGPL6RPtoeDn1+V2/P29sbQoUNNJrrKyMhA8bQo+/btw7JlyzBv3rwS27Zt2xb79u2Dm5tblV+fiEwZNLnIPqdGQboe9goJXDorIfZyNndYRTgyTxEru76BPfhERBYsPzERaT/+iOwjR6C/dRvZR44g7cdw5CdWb7ZKpVKJpKQkLF26FK1atUJmZibatm0LAEhISECjRo3QoEEDqNWmo5ycPHkSzZo1q9Zrm4ter8e0adPQuHFjuLq6ok2bNtiyZUuZ64tEIshkMri4uMDFxQWDBw+uw2ipPjFk6FGQroeDrxQF6XoYMvTmDulvFem5rg89/MXXNxTfLPxHDnvwqyEvNrbeTZxARHVL+8cl5Mfdg6RFC4jEYggGA/Q3bkD7xyXIq9GLn5iYCKVSibfeegsjR47EsGHDkJubi1u3buHy5csYNGgQHjx4gL1792L69Ok4e/Ys+vTpAz8/P6xZs6YG97DuGAwGqFQqHDlyBI0bN8bJkycxZMgQNG7cGN26dSt1mwsXLqBly5Z1HCnVN2K5BPYKCfKTtLBXSCCWS8wd0t8e13NdzR5+q8qlPJtafGJfjAl+FdXXqY+JqG4ZkpMgcnKCSFz0dS0SiyFycoIhOanKbaampmL//v0ICwuDk5MTJBIJwsLCsGnTJiQlJWHz5s3o3Lkz8vLy8I9//APTp09Hly5dcPDgwZraLbOQyWRYunSp8XGPHj3w9NNP49SpU2Um+ETRWj3i9XloIHFEY2ntJN5iL2e4dFbCkKGHWC6xnPIc4PEj8zzcw59yveh+BZNg5lK1hyU6VZSfqIZBnQjHpkEwqBORn8jJOoio5ol9fCHodBAMBgCAYDBA0Okg9vGtdFtr165FaGgowsLCsGLFCpMa/LFjx2Lr1q04evQonnzySQCAo6MjsrOzodVqS7QVFRWFfv364fz58xg6dGgV9868cnJycP78eQQHB5e5Tt++feHr64shQ4bg2rVrdRgdWYJorR6bEjXYnJCKTYkaRGtrr3RG7OUMp6YK8yf3pZXblDcyTzVq05lL1R724FeRg58SYqUf8u7chljpBwc/TtZBRDVP+mR76G/dgv7GDYicnCDodHDw94f0yfaVaickJAR37941Wfb0008b77u4uODo0aMlttu/fz8AlOi9Dw4OxuHDhysVgyUpLCzElClT0LlzZwwYMKDUdSIjI9GtWzfo9Xp88MEHGDBgAP76668yLzBeuXIlVq5cabLMWq9XoCLx+jwk6PLRQuaEGzk6xOvzaq0X3yJUpdzmMT386claZKfp4eIugcJHavIcc6naIxKKh00gAED//v1x6NChCq1rVXVjRGS1anoUnfpOEATMnDkT165dQ0REBFxcXCq0XUBAAL766qtKjf1fmWMKWZ7iHvwEXT5UTg6Y5Odl2wl+9K9FCX5xuU2H54t67qsoPVmLP39LMCb4rXuoSiT5zKVqB3vwq8ExIIAfRiKqdQ5+ftW6oJb+JggC5syZg0uXLuHw4cMVTu4BwM7ODuwTq18aSyWY5OdV6zX4FqOGh4LMTtMjO00PDz8ZHiTmIDtNXyLBZy5VO5jgExFRvTF37lycOXMGR44cKXcs/2vXrkGv1+OJJ55AXl4ePvzwQ+Tm5tb6xbgxmhwkZORCJXdGoJfs8RtQrWssldh+Yl/scRfUVpKLuwQu7hI8SMwx3rdGqampyMzMhJubGzw9Pc0dToXwIlsiIqoXYmNj8cUXX+DPP/9Eo0aNjOPbv/feewCKrkM4ceIEACA5ORkTJ06EXC6Hv78/zpw5g4iICCgUilqLL0aTg63n4rDt7D1sPReHGE1Orb0WUZnKu6C2khQ+UrTuoTLeHu29twapqam4ePGi8ZaammrukCqEPfhERFQvBAQElFtik52dbbzfp08fXL9+vS7CMkrIyIU6XYdmvi64lZSNhIxc9uKT1VP4SK0ysS+WmZmJzMxMeHt7IyUlBZmZmVbRi88efCKieiAyMhJNmjRBaGgo+vfvj+PHj2PGjBkAgPz8fISEhOD69esQi8W4d+8eAGDKlCnGmWyXL1+O3r174+mnn8bq1auRlZWFTp06mfRob9iwAT/88APOnz+P7t27o3v37vjkk0/qeletlkruDKXCCbeSsqFUOEElt6Cx0MnmaLUxeJB2GlptTI23Ha3V47e0rFodVrSuuLm5wc3NDSkpKcb71oA9+ERE9cSMGTOwePFipKSkYOTIkfDx8cGNGzdw9OhRTJ48GU5OTggKCsKXX35pLFsBgD179iAjIwPHjx8HAJw4cQLOzs6IiIjAc889V+J1GjVqhKNHj8LJyQl9+/bF7NmzIZFYZ+1tXQr0kmFCZ3/W4FOt02pjEJ+wDXq9GhKJEg1U4yGVBtZI27Y28pCnpyc6dOhgdTX4TPCJiCxc1gMd1HczkJOuh0whgbKJHK4eTlVuz9vbG0OHDoVCocC///1vZGdn48CBA7h37x5CQkJw5swZ6PV/97zt2LEDH3zwgfFxz549AaDMA52v79+TcDk6OkIkElU51vom0EvGxJ5qnU6fCL1eDZksCDk5t6HTJ9ZYgm+Lcwd4enpaTWJfjAk+EZEFy3qgw5+/JSAjJRf2DnZIjs3Cg4QctO6hqlaSr1QqIZFIIJfLERYWBnt7e+NzYWFh+OGHH4yPk5OT4ePjU+nXOHr0KAIDA+Ho6FjlOImo5jlJ/CCRKJGTcxsSiRJOkpobhreBxBEqJwfcyNFB5eSABhLb+fs3aHJhyNBDLJeYf8bhx2CCT0RkwdR3M5CRkgvPBjLY2duhsKAQqfE5UN/NqFaCn5iYiKeeegoBAQHw9/c3eW7SpEkYMWIEVCoVAMDHxwdJSUlQKis+y6Rarcby5cuxe/fuKsdIRLVDKg1EA9V46PSJcJL41VjvPWC7cwcYNLnIPqdGQboe9goJXDorLTrJ50W2REQWLCddD3sHO9jZF31d29nbwd7BDjnpVb94LTU1Ffv370fnzp1LfV4qleKJJ57A77//DgAYNWoU/ve//xmfP3nyZLnt5+fnY+rUqVi9enWlJpIiorojlQbCw71blZP7vNhY5Jz5HXmxsSWeayyVoIe7q80k9wBgyNCjIF0PB18pCtL1MGRY9gXETPCJiCyYTCFBQX4hCgsKAQCFBYUoyC+ETFH5A+fatWsRGhqKsLAwrFixAjJZ2bXec+bMwc2bNwEAw4YNg5ubm3EUnT/++AMAMHDgQJw/fx79+vUzrgsA27dvxx9//IHZs2cjJCTEOBIPEdmGvNhYpIVvR/r2cKSFby81ybc1YrkE9goJ8pO0sFdIIJZb9o8XkcB5t030798fhw4dMncYZMPSk7XITtPDxV1i1WMDU914tAa/IL8Qcm/natfgU93gMYVsUc6Z35G+PRyOTYOQd+c2FGPGQvZUV3OHVetYg09EpUpP1uLP3xKMCb61zuxHdcfVwwmte6hqdBQdIqLqcPBTQqz0Q96d2xAr/eDgV/Hrc6yZ2MvZ4hP7YkzwiepQdpoe2Wl6ePjJ8CAxB9lp+uon+Kl3gMx4wK1BjUwtTpbH1cOJCT2RmfHs698cAwLgPnYM8hPVcPBTwjEgwNwh0SOY4BPVIRd3CVzcJXiQmGO8Xy2pd4CL3/+d4Hd4nkk+EdmEuMw4qHPUUMqU8Hfzf/wGtYhnX0tyDAiw2MQ+RpNT7yeMY4JPVIcUPlK07qGquV6gzPiim3dLIOV60X0m+ERk5eIy47D95nYkaZPgK/XFmOZjzJrk18rZV6oVMZocbD0XB3W6DkqFEyZ09q+XST5H0SGqYwofKRq2cK+Zg4Nbg6JbyvW/7xMRWTl1jhpJ2iQEKYKQpE2COse8IzHV+NlXqjUJGblQp+vQzNcF6nQdEjJyzR2SWTDBJ7Jmnk2LynKKb+y9pzJERkaiSZMmCAkJQUhICEaPHo0OHTqgV69eeP311wEAMTExmDVrlnGbKVOmcIhLMgulTAlfqS9up9+Gr9QXSpl5L+IsPvtafGPvveVSyZ2hVDjhVlI2lAonqOTWcVFsTbOZEp0pU6Zgy5YtJlOi//nnnyVmaCSyOZ5NmdhThcyYMQOLFy8GACxZsgQLFizAU089heHDhyMrK8vM0RH9zd/NH2Oaj7GYGnygKMlnYl8+rTamVmbHrYxALxkmdPZnDb65A6hJ8+fPx4oVK8wdBhFRjcrUpCDx5l/ITnsAF3cP+DVvBTcv7xppu7CwENnZ2SgoKKiR9ohqir+bv0Uk9jXFEpLf2qTVxiA+YRv0ejUkEiUaqMabNcmvr4l9MZtK8ImIbE2mJgVXjhxERpIaYgdHqO/cQsq9ODwROrDSSf7atWtx8OBBAEDLli3x0ksvIS0tDSNGjIBCoUB6ejr27NmD69evAwCuX7/OThOiGmBJyW9t0ekToderIZMFISfnNnT6RJvbR2tiUzX4a9euhYeHB9q1a4dvv/3W3OEQEVVb4s2/kJGkhrd/ADwbNoK3fwAykhKRePOvSrc1Y8YMREZGIjIyEkqlEl988QUuXLiAGzduoHhS82HDhhnXGTRoUE3vDlGNyYuNRc6Z35EXG2vuUB7r4eRXr1dDp080d0g1zkniB4lEiZyc25BIlHCS+Jk7pHrNZhL8V155BTdv3kRycjI++eQTLFy4ED/99FO526xcuRI+Pj4mN61WW0cRExE9XnbaA4gdHGFnX3TC1c5eDLGDI7LTHtRI+x4eHujcuTMOHz5cI+0R1YW82FikhW9H+vZwpIVvt/gkvz4kv1JpIBqoxkOlGmeTZyisjc0k+B06dICXlxfEYjH69OmDOXPmIDw8vNxtFixYgOTkZJObVMoLaIjIcri4e8CQn4fCAgMAoLDAAEN+HlzcPSrd1tq1a42j6Jw+fdq4fMqUKfjmm29qLGai2pafqIZBnQjHpkEwqBORn2jZoz3Vl+RXKg2Eh3s3m90/a2KzNfh2dnbGU85ERNbKr3krpNyLQ0pcLMQOjjDk50Hu6we/5q0q1U5ISAju3r1b6nONGzfGDz/8AABYs2aNcfmGDRuqHDdRbXLwU0Ks9EPendsQK/3g4GfeYTQrQioNZOJLdcZmEvwff/wRgwcPhkwmw6lTp7B69Wp89tln5g6LiKha3Ly88UTowFobRYfIGjkGBMB97BjkJ6rh4KeEY0CAuUOqW6l3imYud2vAYZKpVDaT4K9evRozZsxAQUEB/P39sWzZMoSFhZk7LCKianPz8mZCT/QIx4CA+pfYA0XJ/cXv/07wOckhlcJmEvxff/3V3CGQFcuLja2/PUFERGQ9MuOLbt4tgZTrRfeZ4NMjbCbBJ6qq4tEYDOpEiJV+cB87xqKTfFufLIWIiMrh1qDolnL97/tEj2CCT/Xew6Mx5N25jfxEtcUm+PVhshQiIiqHZ9OishzW4FM5bGaYTKKqsqbRGOrDZClUOyIjI9GkSRPjMJmjR4/G7t27AQAfffQRfvjhByQmJmLYsGHo3bs3Jk6ciKysLADArVu3MGjQIPTu3RtDhgxBfHw8AECv18PPzw9nzpwx235R/aTVxuBB2mlotTHmDsU8PJsCjXsxua8jaeoExEVdQZo6wdyhVBh78Knes6bRGOrDZClUe2bMmIHFixcDALKysjB8+HD06tULBw8exKFDhxAWFobly5ejbdu2OHDgAJYsWYL//ve/mDNnDtatWwd/f3/cv38f+fn5AIDvv/8erVpVbrhOourimUyqS2nqBFw9EoGsVA1cPb3QNnQg3JUqc4f1WEzwiWA9ozEUT5bCGvz6xZCuR15sJgoy9bB3k8AxwA1ihaRabbq6umLo0KHo168fli9fjoKCAmRnZ6Nt27YAgMGDB+OTTz5BTEwMmjZtCn9/fwBAw4YNAQCCIOD48ePo1atX9XaOqJIePpOZk3MbOn0ivwstVTWH80xP1iI7TQ8XdwkUPuaZiDRLo0FWqgaeDf2Rej8OWRqNVST4LNEhsjL1cqbA1DtA9K9F/9YzhnQ9cs4mIvevVOQnaZH7VypyzibCkK6vdFsPz2SrVqsREhKC27dvo2/fvtBoNPD19TVZ387ODmq1Gn5+Jc8U7d69G88++2yV98sc9Ho9pk2bhsaNG8PV1RVt2rTBli1bylz/+PHjCA4OhlQqRefOnXH58uU6jJbKwjOZVqJ4OM/iWyW/v9OTtfjztwTjLT1ZW0uBls/Vywuunl5IvR8HV08vuHp5mSWOymKCT0SWrZoHCWuXF5sJwwMdHHylcPCRwsFXCsMDHfJiMyvd1owZMxAZGYnIyEgolUq8//77ePnll7F27Vp4eXkhKSnJZP3CwkIolUokJJSsO/3xxx8xZsyYKu+XORgMBqhUKhw5cgSZmZn46quvMHv2bJw+fbrEuqmpqRg+fDgWLlyItLQ0TJgwAcOGDYNeX/kfVlSzis9kqlTj6nV5jsXXhT88nGfx/UrITtMjO00PDz+Z8b45uCtVaBs6EG37DrSa8hyACT4RWbpqHiSsXUGmHiKxHUT2RV/XIns7iMR2KMis3sHuzJkz8PHxwZIlSxAeHo68vDxIpVJcvXoVAHDgwAG0bt0agYGBuHv3Lu7duwcAiI+PR0xMDP766y88++yz2LRpE1599VUUFhZWb0frgEwmw9KlS9GkSROIRCL06NEDTz/9NE6dOlVi3R07diAoKAjPP/88JBKJcR8PHz5shsjpUVZ7JrOGzkYW14VfPRqBq0ciLDPJr+Zwni7uEri4S/AgMcd431zclSr4Bz9hNck9wBp8IqoD1aqjrOdjPtu7SZAXnw2hoBAiezsIBYUQDIWwd6v8wW7t2rU4ePBgUbv29ti0aRPEYjFeeuklfPrpp1i1ahVmzJiBrKws+Pn5Ye3atQCKZgqfPn06cnNzIZPJsG7dOvzxxx8AgCVLlmDQoEGws7O+/qKcnBycP38e//znP0s8FxUVhfbt2xsfi0QiPPHEE4iKisKQIUPqMEqyGTU4A61V1IVXczhPhY8UrXuozF6Db62Y4BNRrSquoyz+km7dQ1W5L+p6PuazY4Ab8pNykJ+khUhsB8FQCLGHExwD3CrVTkhICO7evVvqc+PHjzfe37dvX4nnmzdvbvxh8KglS5ZUKg5LUVhYiClTpqBz584YMGBAieezs7Ph7u5uskyhUBiHDi3NypUrsXLlSpNlzZo1q5mAyfrV4Ay0VlMX7tm0Wt/ZCh8pE/sqYoJPRLXq4TrKB4k5yE7TV/4Lu5oHCWsmVkgg6+JX46Po1GeCIGDWrFlISEhAREQERCJRiXVcXFyQkZFhsiwjIwOurq5ltrtgwQIsWLDAZFn//v1rJmhbVs2RVqxGDZ6NLK4Lz9Jo4OrlZXm992R2TPCJqFbVZR1ltFaPeH0eGkgc0VhqOwmwWCGBWOFt7jBsgiAImDNnDi5duoTDhw/DxcWl1PWCg4ONJUrF2125cgWzZ8+uq1DrhxosW7F4NXw20l2pKjWxT1MnWH3inxcbaxVz01gyJvhEVKvqoo5Sq43BjUw1dqXLkVzgDJWTAyb5edlUkk81Y+7cuThz5gyOHDkCN7eyy5xGjRqFBQsWYNOmTRg3bhy++OILAEC/fv3qKtT6oQbLVqxCLZ+NrMikTJbeEZIXG4u08O0wqBMhVvrBfewYJvlVYH1XRRGR1VH4SNGwhXulk/vU1FRER0cjNTW1zHWKZ7W8lngcd9Kj0cQxFwm6fMTr88potP6OqV/fxcbG4osvvsCff/6JRo0awcXFBS4uLnjvvfcAFJXlnDhxAgDg6emJXbt2YcWKFZDL5di8eTP27NkDicTyEiKrVs8voq9pD198m5WqQZZGY/J8tFaPTYkabE5IxaZEDaK1ljfsa36iGgZ1IhybBsGgTkR+otrcIVkl9uATkUVKTU3FxYsXkZmZCTc3N3To0AGenp4l1iue1dLfJRie+lRcz/ZBoJsbGkgcjesYT/c658Ex+XDVygHqS52wDQsICIAgCGU+n52dbfI4JCQEUVFRtR1W/VbPL6KvaY+7+DZen4cEXT5ayJxwI0eHeH2exfXiO/gpIVb6Ie/ObYiVfnDwU5o7JKvEBJ+ILEqMJgcJGbkozEpFZmYmvL29kZKSgszMzFIT/OJZLT30URghb4JChRxN5H+X55ic7nXOh3sjDRxbPlm5cgAbqBP+9ttv8d1330EQBERHR6Nhw4ZIS0uDnZ0dfHx88M4772DkyJF44oknAADr16/Htm3bcODAAej1eqxatQpdu3ZFUlISZs+ejQcPHsDe3h6rVq1CcHAwAKBDhw5YuHAhwsLCzLmrZG3q8UX0Ne1xF982kDhC5eSAGzk6qJwcTDpCLIVjQADcx45hDX41McEnIosRo8nB1nNxUKfroJAIaC52RUpKCtzc3Mqsly6e1VKnT0QTiV+JiW8ePt2b9+cl5Cvd4FjZcgArrxO+fPkyjh49imPHjsHOzg6XLl1C8+bN8eOPP8LJycmYkHfp0sVkOMzXXnsNixcvxr179/Daa6/hxx9/xCuvvIK3334b7dq1Q3p6Ou7fvw8AOHToEBQKhTl2j6hs9fDMW1kX3wJAY6kEk/y8LLoGHyhK8pnYVw8TfCKyGAkZuVCn69DM1wW3krLh2TIILRQiuLm5ldp7X0wqDSxzRkuT070BzeDQowvgJlTugG/mOuGMjAzcu3cPWVlZcHV1RaNGjSCXyyu8/c6dO/HKK68YJ6N6eAKn8jg4OAAoKl1p27YtDAYDtFot2rVrB6BoXPjipH7z5s2YMGFCxXeKqLbZwJm32tBYKrHYxJ5qDhN8IrIYKrkzlAon3ErKhlLhhJaNfBHoJatWmzVyuteMdcIZGRm4cOEC0tLSIBaLkZCQgOTkZHTs2LHCSb5arYZSWVTH+vLLL+PEiRNYv359ifXOnj2LkJAQyOVy7N69GwAwbdo0REREYOPGjdBoNPD2Ljlc56VLl9CqVSvjDwIii2DlZ94siVYbA50+EU6lnCUly8QEn4gsRqCXDBM6+yMhIxcquXP5yf3/n3qPlvgh3llZ7unmGjnda6Y64Xv37iEtLQ0+Pj6wt7dHQUEBkpOTce/evQon+EqlEomJifD398dnn32GJUuWQK8vOXrGoyU6ALBu3Tqo1WqEhYXh8OHDSE5OLrHdl19+iRUrVhh/FBBZBI7QUyOKRyrT69WQSJRooBrPJN8KMMEnIosS6CV7fK/9/596j87OxibXrkjwaA2V3N0mx77PysqCWCyGvb09AMDe3h5isRhZWVkVbmPkyJFYuXIlOnfuDDs7OxgMhgptl5eXB0dHR7i4uEAmk0EsFkMmk+Hy5cto166dsXQoNjYWEyZMQHx8POzt7dG7d2/4+flVaX+JagxH6KkRxSOVyWRByMm5DZ0+kQm+FWCCT0RmYdDkwpChh1gugdjLudR1UlNTjcNkmtTg//+p93ivbkjIFdBClI0bOheLHPKtulxdXZGQkICCggJjD77BYICrq2uF22jXrh369u2L3r17w8nJCX5+fmjZsiWuX79e7navvvoqrl27hoKCAixfvhwAsGrVKrz00kt48OAB7Ozs8Nlnnxl7/Tds2GBsn8gicISeaiseqSwn5zYkEiWcJPz7tgYiobxBgeuh/v3749ChQ+YOg6jO1eXshgZNLrLPqVGQroe9QgKXzsoSSX654+DXox78R2vwDQYD3N3dK1WDT+bDYwrVtDR1QpnDYNYW1uBbH/bgE5FxdsMEXT5UTg61nijrbsZBfycFjio3FKSjqCf/kQQ/MzOz7HHw///Ue+PMeEyqQA2+NZPL5ejYsWO1RtEhosqry06PikpTJ+DqkQhkpWrg6umFtqED6yTJL2+kMrJMTPCJqE5nN8yLjUX2iUMwaKQwJLvDqXVjiOUlX6t47Psyx8H//1PvjQE0rpVILYdcLmdCT1SH6rrTo6KyNBpkpWrg2dAfqffjkKXR1FkvPlkXJvhEVKezG+YnqlGQEgNH/9bIiI6CNr8AdgYfuMO0B9/T0xMdOnQovQafiKgW1WWnR2W4ennB1dMLqffj4OrpBVcvL3OHRBaKCT5RfVHOjI51Obth8cRTD+5cQjQMMNwvgPxIVqmnmj09PZnYE1Gdq8tOj8pwV6rQNnRgndfgk/WxM3cARFQHimd0LL6l3imxSmOpBD3cXWsuuU+9A0T/WuK1iieesuvxNAxKH3i3aIWMuFiknDiBvNjYmnnt/xejycGpOxrEaHJqtF1rFBkZiRUrVsDDwwMhISEIDQ2FWq0GALz99tvo3bs3evfujZMnTwIAYmJi0KdPH/Ts2RNHjx41Z+hEda640+M5lafFlOcUc1eq4B/8xGOTe602Bg/STkOrjambwMiisAefqD74/2El8wQV8i9dg4P0Dzg+XYtDxz1minjHgAB4SxygztMi5cZfEKuTUah+gLTY+3AfO6b6k1KhKLnfei4O6nQdlAonTOjsX+1ZcW1B8WRW+/btw9dff4127dpBp9Ph+PHjyMjIwPDhwxEREYEPPvgAn332GZo3b45Ro0ahb9++5g6dqE41lkosKrGvDE5ORVXuwddoNNDpdCWWGwwGXLx4sVpBEdW1NHUC4qKuIE2dYO5QaodbA+Tp3ZAWcRLpl7KQdvRSjfeWm3h4ivji+48oPtXcIqgVGkMMz5atYVAnIj9RXSMhJGTkQp2uQzNfF6jTdUjIyK2Rds1Bp0tAUtJ+xMV9i6Sk/dDpqv85bdu2LeLj47Fjxw68+uqrAIou5h04cCBOnz6NuLg4tGnTBo6OjtDr9dBqtdV+zari8Yaoch6enEqvV0OnTzR3SFTHKp3gp6am4oUXXsAzzzyDkJAQvPPOO8jN/fvAmZGRgVmzZtVokES1qXjYsatHI3D1SIRVJflxmXE4m3gWcZlx5a/o2RT5Xj1hsG8Ax469YcjQVTyRLqPUplxlTBH/aMmMu1KFwC5Pwb1RAPLu3IZY6QcHP2XFX6ccKrkzlAon3ErKhlLhBJW89Mm0LJ1Ol4D4hB+QojmE7JybSNEcQnzCD9VO8k+ePIlmzZohOTkZSuXf77lKpYJarUazZs1w4sQJpKen4/Lly8jIyKjurlQajzdEVcPJqajSJTqrV69GYWEhNmzYgKysLHz++eeYNWsWVq9ebZxZkXNnkTWx1mHH4jLjsP3mdiRpk+Ar9cWY5mPg7+Zf5voOLTpAHHQXeQmJFU+kH1NqU6ZSpogvq2TG4C1A9GxTOD0IgmuD9jVSngMAgV4yTOjsj4SMXKjkzlZbnpOR8Qdyc+Mgk7WEnZ0YhYUG5ORcR0bGH3Byqvzn9OzZs+jTpw/8/PywZs0aXLlyBWq12pjkJyQkoFu3bvj3v/+NGTNmQBAEtG3bFl5mGK2DxxuiqpFKA9FANZ6TU9Vjle7B//333/Gvf/0LrVu3RteuXbFu3Tp4eXlh5syZxh4ekUhU44FWRHp6OsaNGwdXV1eoVCp88sknZomDrIu1DjumzlEjSZuEIEUQkrRJUOeU3yNffHGrYszYite5V6DUpkyeTYHGvYw/CIpLZjq5pcEl4TQe3PvLWCeaXHAcD3yuw+Bds8laoJcM3Zt6WW1yDwB6fRLs7JxgZ1fUH2NnJ4adnRP0+qQqtdelSxccO3YMW7ZsgZubG0aOHImPP/4YQFGP+MGDB/HUU0/B19cXu3fvxsaNGxEQEAAHB4ca26eKsuTjDVFFmetif6k0EB7u3Zjc11OVTvCzsrJMJpxxdHTEBx98AJVKhVmzZiEtLa1GA6yMuXPnQq/XIz4+HhEREXjvvfdw4MABs8VD1qG4Frxt34F1NitgTVDKlPCV+uJ2+m34Sn2hlD2+R94xIACyp7qWSO6jtXr8lpaFaK3edIMySm2qQiV3RiunFPjd+RF9tQfQ5N5PyEu6wDrRx5BIfFFYqENhoQEAUFhoQGGhDhKJb420P3z4cDg4OKBXr14YOnQoli1bBicnJ+zZswd9+vRBWFgY/vOf/9TIa1WWJR9viCqi+MzltrP3sPVcHEf0ojojEip5fjMsLAzTp09HaGioyXKDwYDFixfj+vXrSE5OxtmzZ2s00MfJycmBh4cHLly4gODgYADA66+/jps3byI8PLzC7fTv3x+HDh2qrTCJalRcZhzUOWooZcpyy3PK89gZG8sZP7+y1Jd+gf2ljbD3bQUP7V3oWg/APYdojvRQjuIa/NzcONjZOaGwUAdnZ380UIVVqUTHmljq8aYyeEyp307d0WDb2Xto5uuCW0nZGN+lEbo3rfxZ4hhNjtWXG1LdqnQPfvfu3bFz584Sy8ViMVasWIHmzZvXSGCVdfPmTRQWFhqTewBo3749oqKizBIPUV3wd/NHF78uVU7uAdMZGxN0+YjX55mu8EipTZkqcDGuslFTeDdoAg/tXcCtAZx8OqKBajxUqnFM7svg5KRCA1UYvL36w0XWHN5e/etFcg9Y7vGGqKJq4mJ/ngUwlRcbi5wzv9fuSHA2oNIX2b700kulDlcGFH3pfvjhh0hJSal2YJWVnZ0NuVxuskyhUCArK6vMbVauXImVK1eaLGvWrFmtxEdkqWpkxsaKXoxbysW3UoCJ/WM4OanqRUL/KEs93hBVVE1c7P/wkL+3krKRkJFbb3vx82JjkRa+HQZ10WARNTVvii2qdA++SCRCYmJiqV+6Op0O0dHR8PWtmdrQynBxcUFmZqbJsoyMDONIC6VZsGABkpOTTW5SqbS2QyWyKDUyY2NlLsat6BkBqvcs9XhDVBnVvdjfVob8rQn5iWoY1IlwbBpUo/Om2KJKJ/j79u3D0qVLSx1RQSwWY+nSpWa5sLV58+YQiUS4du2acdmlS5dMSnaIqHSNpRL0cHetUHJf6ogQNXgxLlExSz3eEFVZFeYVKT4LMLllAab63UOgqP4mtQ5+SoiVfjU+b4otqnSJzu7duzFp0iTY29uXbEwsxvPPP48ffvgBQ4YMqZEAK0omk2HMmDF4/fXXsXHjRsTGxmLdunVYv359ncZBZMvKGsu+tNKbmrw4l+onSz3eEFVJVecVARAoUiNQs7do25TKbVsb0pO1yE7Tw8VdAoVP3VU+FA/3nJ+ohoOfkuU55ah0D35sbCzatm1b5vOtW7dGTExMdWKqss8//xwODg7w8/ND//79sXjxYgwePNgssRDZoodrQdXpOiRk/D2rqEnpTfGBrPhWmVlwqVZERkZixYoVAICPPvoIb775JtRqNSQSCR48eAAAiImJgVgsxr179wAAU6ZMgVqtRnR0NPr06YPevXvjmWeeMXmu2C+//IKePXuiS5cuWLt2bY3EbMnHGzKP9GQt7t9IQ3qy1tyhVF515hWpzrY1LD1Ziz9/SzDe6vr/oqzhnslUpRP8vLw85OSUfQW3VqtFXl5emc/XJoVCgfDwcGRnZyMxMRHz5s0zSxxEtqrCtaAWdDAiU7t378aFCxewdOlS7N27F2FhYfj555+NzwcFBeHLL7802Wb16tVYuXIljh8/jq1bt5babp8+fXDixAmcOXMG3333XY3EasnHG6p75k4sq606pYwWVAaZnaZHdpoeHn4y432yPJUu0WncuDEuX75c5mgzf/zxBwIDA6sbFxHVEIMmF4YMPcRyCcRe1bs4q8IjQljQwcgWxOvycD4jB+q8fCgdHdBJLkMDp8qPdvTHH3/g0KFD2L9/P0QiEY4dO4ZPP/0U8+fPx6RJkwAAISEhOHPmDPT6vw/azs7OOHbsGNq0aVNitLJixXXyeXl5aNGiRRX2siQeb+hhDyeWDxJzkJ2mr9PykGorrZSxLratYS7uEri4S/AgMcd4nyxPpXvwhwwZgi+//BLXr18v8dz169fx1Vdf4dlnn62R4IioegyaXGSfUyPnrBrZ59QwaHIfv9FjVGhEiOKDUfGNNfhVFq/Lw8aEVBzUZOB6tg4HNRnYlJCKeF3le64PHjyI0aNHw8nJCbm5uXBwcICXlxcKCgqQn59vXC8sLAw//PCD8fG//vUvxMTEIDg4GAsXLiyz/dWrV6NFixZo3bp1pWMrDY839DCbSCyrM4qYhYxApvCRonUPlfFmVT+y6pFK9+CPGzcOFy9exOTJk9GlSxdj70lMTAzOnj2LXr16Ydy4cTUdJxFVgSFDj4J0PRx8pchP0hb15FezF7/CPJua/UBkC85n5CA2V4/WMmeI7UQwFAr4MycX5zNyKt2LP2/ePGzZsgXdu3dHXFwcLl26hEGDBiE+Ph6RkZHGnvJJkyZhxIgRUKmKxt5XKBT4/PPPUVhYiPHjx+Ovv/4qtf25c+di5syZCA0NxbRp06BQKKq17zV9vFm9ejU2bNiAq1evYuTIkSY/Yh4lEokglUohEokAAD179uSIPWZWnFia4+JOMqXwkfL9t3CVTvBFIhE++OADREREICIiAmfOnAEANGrUCO+88w4GDRpU40ESUdWI5RLYKyTIT9LCXiGBWG6FPV71nDovH052dhDbFSWaYjsRnOzsoM7Lf8yWJTk7O2PTpk0YO3Ys2rdvj8OHD8Pb2xv37t3DypUrMX/+fACAVCrFE088gf379wMAoqOj0bhxY9jZ2cHT0xOFhYUl2s7Ly4OjoyMcHBwgk8kgkVT/s1bTxxuVSoU33ngDhw8fhkajeez6Fy5cQMuWLasUO9UOJpZEFVPpBL+goAAbN27Er7/+CoPBgJ49e2LGjBlwcnKqjfiIqBrEXs5w6ayssRr8WsVhNUuldHTApUItDIWCsQdfV1gIpWPJseErIjAwEO+99x6GDRuGNWvWAChKmK9evWqy3pw5c/Dxxx8DACIiIrB+/Xo4OTmhXbt2aNOmDYCiHnZHR0cEBQWhY8eO2Lx5M/Lz8zFhwgQ4O1f/s1bTx5tRo0YBKJojpSIJPhGRtRIJgiBUZoPiseWfeeYZSCQSHDhwAL169cLbb79dWzHWqf79++PQoUPmDoPI6sRocowX39o5pkKdo4ZSpoS/m//jN67G+NC2Ll6Xh00JqYjJ1cPJzg66wkIEOkswSeVZpQttrUltHW+WLFmC69evP7ZEx8/PDwUFBejUqRM+/PBD4w+byuAxhSxFamoqMjMz4ebmBk9PT3OHQ7Ws0j34+/fvx5tvvmk8NTpo0CBMnz4db775JuzsKn3NLhFVlgX2dD88AZbMJR2OirPIFR7AV+qLMc3HPD7Jf3hYzZTrRfctZN/MrYGTIyapPGtkFB1rY87jTWRkJLp16wa9Xo8PPvgAAwYMwF9//QU3N7cyt1m5ciVWrlxpsqysEYCoFlngd6S5paam4uLFi8YEv0OHDkzybVylvyHVajU6dOhgfBwcHAw7OzukpKTUaGBEVAoLnUDq4Qmw7mcmIC4zEUGKICRpk6DOUUOrjcGDtNPQamNKb4DDapargZMjhvu6Y2YjHwz3da8XyT1g3uNN79694ejoCFdXVyxbtgxisRinTp0qd5sFCxYgOTnZ5CaVsl68Tlnod6S5ZWZmIjMzE97e3sb7ZNuqVINfPN5xMXt7exgMhhoLiojKYKE93Q9PgNXQTQVHNz/cTr8NX6kv5PYGxCdsg16vhkSiRAPVeEilgaYNWNAYz2Q5LOl4Y2dnh0pWtJI5WOh3pLm5ubnBzc0NKSkpxvtk2yqd4AuCgHfffReOjn/3IOn1enz44YcmF1UVT4lORDXIQnu6TSfAag47x6bGGnyXgngk6NWQyYKQk3MbOn1iyQQf4LCaVEJNH28MBoPxVlhYCJ1OB3t7+xI/Iq5duwa9Xo8nnngCeXl5+PDDD5Gbm4tu3brVzI5R7bHQ70hz8/T0RIcOHViDX49UOsEvbVKRZ555pkaCIaLHsOCe7kAv2UOTX8mMdfdabSEkEiVycm5DIlHCSeJnviDJqtT08WbZsmV45513jI/Dw8MxefJkbNiwAS4uLjhw4AB69uyJ5ORkzJ49G/fu3YOzszM6deqEiIiIao/rT3XAgr8jzc3T05OJfT1S6VF0bB1HPCCqeVptDHT6RDhJ/ErvvadaFxkZiTNnzuDBgwc4e/Ysrly5gieeeALOzs4YP348PvzwQ/j4+MDFxQU//fSTcRz7+fPnQ6vVYs2aNdiwYYNxvcaNG2P9+vWIjY3FpEmTAABdu3bFRx99ZM7dtDg8phCROXDYG6J6xKDJhe5OOgya3Np5gdQ7QPSvJS5sk0oD4eHejcm9Bfjwww8RGRmJLl26IDIy0jg761tvvYXIyEg8/fTTOHr0KAAgLS0Nt2/fNtm+eD1XV1dERUVBLpdj3759OHHiBJKSknD37t063yciIjJV6RIdIrJOBk0uss+pUZCuh71CApfOypqd+Ipj2deahPRcXIxNgzpTB6WbEzoEuEOlqJ1Jy7KysuDi4gIAWLNmDWbMmIF9+/aVWC8jIwOCIJiUrTg4OEAkEtVKXEREVHHswSeqJwwZehSk6+HgK0VBuh6GDH3NvsDDo1cU36dqS0jPxdazcfjlryTcTMrCL38lYevZOCSk1+xZmKVLl+Kpp57C/v370b17d+j1ety4cQPBwcEl1mvVqhUyMzPRtm1b4/Lr168jNTUVjRs3rtG46BFlnCUjInoYE3wiK/bY8eUfIpZLYK+QID9JC3uFBGK5pMx1o7V6/JaWhWhtJX4EcPSKWnExNg2xD7Ro5euKZj6uaOXritgHWlyMTavR13nrrbdw5swZzJo1C+Hh4fj+++/x3HPPlbrepUuXoNVqkZtb9CMjJycHc+fOxZdfflmjMdEjOMY7EVUQS3SIrJRWG/P48eUfIvZyhktnJQwZeojlEoi9nBFz7TRSY2/CM6A5AtsUDQEYrdVjU6IGCbp8qJwcMMnPC42lZf8YMOLoFbVCnamDk9gOYvui/hixvR2cxHZQZ+pq5fXkcjnS09MRHR2Nn376Cbm5uYiOjkZERIRxHYlEgrCwMGzatAnTp0/HzJkz8frrr0OlUtVKTPT/OMY7EVUQE3wiK6XTJ0JfkfHlHyL2cjbW3cdcO42obz8BUlKR6O0JvAAEtumGeH0eEnT5aCFzwo0cHeL1eRVL8AHAsyliBCUS0nOhEnIeGjaTqkrp5oQr8RkwFBRCbG8HQ0EhdIZCKN2cKt3WN998g8OHDwNAiVr5pUuX4quvvoKDgwO2bdsGd3d3AEBMTAxWrFiBgQMHYsOGDcb1x44di2HDhiE4OBj79u3D/fv38c477+CTTz5B+/btq7y/VA6eJaMKSk1N5Zj39RyHyXwEhzQja1HZHvxHXfj5OyT+sBH2TQJQcDcWfmH/QMdnJle9Bx9AjCYHW8/FQZ2ug1LhhAmd/ZnkV1NxDX7sAy2cxHbQGQoR4CHFhC7+tXahLdWcGj+mpN7hWTIqV2pqKi5evGhM8Dt06MAkvx5iDz6RlZJKA9FANb7K48t7BjRHorcnCu7GAt6e8AxoDgBoLJVgkp8X4vV5aCBxrHjvPYCEjFyo03Vo5uuCW0nZSMjIrXCCH5cZZ5z9tniSLAJUCmdM6OJfZ6PokIXjjM/0GJmZmcjMzIS3tzdSUlKQmZnJBL8eYoJPZMWk0sAqjy0f2KYb8AJK1OADRUl+ZRL7Yiq5M5QKJ9xKyoZS4QSVvGJJaFxmHLbf3I4kbRJ8pb4Y03wMk/yHqBTOTOiJqELc3Nzg5uaGlJQU432qf5jgE9VjgW26mST21W7PS4YJnf2RkJELldy5wr336hw1krRJaFKoxL2bdxDt+Cf8n2CCT0RUWZ6enujQoQNr8Os5JvhEVKMCvWSVrrtXypTw0bsi6cIluOvEyNLfQJrPE3BXclQWIqLK8vT0ZGJfz3EcfCIyO383f/SWd0WgyA9tWnSBKCsPWRqNucMiIiKySkzwicgiNPVvg8CGLVCQkglXTy+4enmZOySbEhkZCScnJ2RnZwMAPv/8c7Rq1QpTpkxB9+7d0b17d+NEVZGRkWjSpAlCQkLwj3/8A4IgYMqUKVCr1cb2Tp06hdatW2PQoEHGZbGxsQgJCUGPHj1w7dq1ut1BIiIyYoJPRBbBXalC29CBaNt3INqGDmR5Ti1o1aoVfvnlFwDA6dOn0ahRIwDAjh07cOrUKYSHhxvXnTFjBiIjI+Hr64vffvutRFvBwcG4cOGCybIPPvgAn3/+OX788Ue8++67tbgnRERUHtbgE5HFcFeqmNiXJuM+cO8skJUAuKqARl0AecNKNzN48GD88ssvGDBgAKRSKezs/u7j0ev10Gq1JbZp27Yt4uPjSywvbWSOuLg4tGnTBgCQlZVV6fiIiKhmsAefqA6kJ2tx/0Ya0pNLJlBE5cq4D1zYANzYDyRfL/r3woai5ZXk5uaG7OxsHDx4EP379zcuHzVqFAIDAzFhwoQS25w8eRLNmjWrUPuFhYWl3iciorrFBJ+olqUna/HnbwnGG5N8qpR7Z4G0aMCnDeDdoujftOii5VXQpUsXvP/++xg4cKBx2Y4dO3Dy5EmcOXPGuGzt2rXo06cPfHx80LFjxwq1LRKJjPcfPjtARER1iyU6NkqrjanyDKdUs7LT9MhO08PDT4YHiTnITtND4SM1d1hkLbISALEzYO9Q9NjeoehxVkKVmhs5cqRxApyHNWnSBBKJBDdv3gRQVIO/ePHiSrXt7++PP//8E+7u7nB1da1SfEREVH3sYrFBWm0M4hO2ISHhR8QnbINWG2PukOo1F3cJXNwleJCYY7xPVGGuKsCQCxTkFz0uyC967Fq1axUaNWpU5gWwU6ZMwbp168rcdty4cejXrx9mzZqFqKgo9OvXD+fPn8fQoUMBAIsWLcLs2bMxZswYvP7661WKj4iIqk8kCIJg7iAsSf/+/XHo0CFzh1EtD9JOIyHhR8hkQcjJuQ2Vahw83GtutlKqvPRkLbLT9HBxlxh779PUCcjSaODq5WV1F5byDFEdKq7BT4su6rk35ALujYGOU6p0oS3VLVs4phCR9bGJHvwNGzbA3t4eLi4uxtvmzZvNHZbZOEn8IJEokZNzGxKJEk4SP3OHVO8pfKRo2MLdJLm/eiQCV49G4OqRCKSpq1ZuYQ48Q1TH5A2LkvkWQwCflkX/MrknM4nLjMPZxLOIy4wzdyhEVA6bqcHv3LmzyQVi9ZlUGogGqvHsYbVgWRoNslI18Gzoj9T7ccjSaOqsF7+6ve86fSL0erXxDJFOn8jPWG2TN2RCT2YXlxmH7Te3I0mbBF+pL8Y0HwN/N39zh0VEpbCZBJ9MSaWBTLosmKuXF1w9vZB6P65as7aWVvpTnuLed71eDYlEiQaq8ZX+nPAMEVH9pM5RI0mbhCBFEG6n34Y6R80En8hC2UyCf+XKFXh7e0Mul2P06NFYsmQJnJ2dzR0WUamKZ22tTg1+8fCbxQl+6x6qxyb5NdH7zjNERPWTUqaEr9QXt9Nvw1fqC6VMae6QiKgMNlGD36tXL0RFRSEpKQn79+9HZGQkFi5c+NjtVq5cCR8fH5NbaTM5EtUGd6UK/sFPVLk05+HhN4vvP05Vet/T1AmIi7picp2AVBoID/duTO6J6hF/N3+MaT4Go5uNZnkOkYWz+FF0xowZg59++qnM50sL/9ixY5gwYQLUanWlX48jHpC1qEoPPlC5Gvzii4GzUjVw9fRC29CBVjfiD5E58ZhCROZg8SU627dvr/Q2dnZ2pSb+RLZE4SNF6x6qStXgA5W7PsOcFwMTERFR1dhEic6BAweQmJgIALh79y4WL16MkSNHmjkqotr36PCbNa2mLgYmIiKiumPxPfgVcfToUUydOhVZWVnw8vLCmDFjsHTpUnOHRWT1auJiYCIiIqpbNtGDv3LlSqjVauTk5CA2Nhb//e9/IZPJzB0WkU2o7sXARJZk9erV6NSpEyQSCcLCwspd9/jx4wgODoZUKkXnzp1x+fLlOoqSLJVBkwvdnXQYNLnmDoWoXDaR4BMREVWESqXCG2+8genTp5e7XmpqKoYPH46FCxciLS0NEyZMwLBhw6DXP360KrJNBk0uss+pkXNWjexzaib5ZNGY4BMRUb0xatQojBgxAl6PuZ5kx44dCAoKwvPPPw+JRIJXX30VhYWFOHz4cB1FSpbGkKFHQboeDr5SFKTrYcjgjz2yXEzwiYiIHhEVFYX27dsbH4tEIjzxxBOIiooyX1BkVmK5BPYKCfKTtLBXSCCWS8wdElGZmOATERE9Ijs7GwqFwmSZQqFAVlZWmdtw8kTbJvZyhktnJWRdlHDprITYy9ncIRGViQk+ERHRI1xcXJCRkWGyLCMjA66urmVus2DBAiQnJ5vcpNLaGcKWzEPs5Qynpgom92TxmOATERE9Ijg4GJcuXTI+FgQBV65cQXBwsPmCIiKqICb4ZLU4XBkRVZbBYIBOp4PBYEBhYSF0Oh3y8/NLrDdq1CjcunULmzZtQl5eHlatWgUA6NevX12HTERUaUzwySpxuDIiqoply5bB2dkZy5cvR3h4OJydnY1DZrq4uODEiRMAAE9PT+zatQsrVqyAXC7H5s2bsWfPHkgkvLCSiCyfSBAEwdxBWJL+/fvj0KFD5g6DHkN3Jx05Z9Vw8JUiP0kLWRclnJoqzB0WEZEJHlOIyBzE5g6AqCrqcriyNHUCsjQauHp5cTZXIiIisnhM8MkqFQ9XZsjQQyyX1NqIBmnqBFw9EoGsVA1cPb3QNnQgk3wiIiKyaKzBJ6tVF8OVZWk0yErVwLOhP7JSNcjSaGrttYiIiIhqAhN8onK4ennB1dMLqffj4OrpBdfHTG9PREREZG4s0SEqh7tShbahA1mDT0RERFaDCT7RY7grVXWW2Kcna5GdpoeLuwQKH86ASURERJXHBJ/IQqQna/HnbwnGBL91DxWTfCIiIqo01uATWYjsND2y0/Tw8JMZ7xMRERFVFhN8Igvh4i6Bi7sEDxJzjPeJiIiIKoslOkQWQuEjReseKtbgExERUbUwwSeyIAofKRN7IiIiqhaW6BARERER2RAm+ERERERENoQJPhERERGRDWGCT0RERERkQ5jgExERERHZEI6iQ0RERPVCjCYHCRm5UMmdEeglM3c4RLWGCT4RERHZvBhNDraei4M6XQelwgkTOvszySebxRIdIiIisnkJGblQp+vQzNcF6nQdEjJyzR0SUa1hgk9EREQ2TyV3hlLhhFtJ2VAqnKCSO5s7JKJawxIdIiIisnmBXjJM6OzPGnyqF5jgExERUb0Q6CVjYk/1Akt0iIiIiIhsiFUk+MeOHUOfPn0gl8uhVCpLPJ+eno5x48bB1dUVKpUKn3zySd0HSURERERkAawiwZfJZHjhhRfwv//9r9Tn586dC71ej/j4eEREROC9997DgQMH6jhKIiIiIiLzs4oa/C5duqBLly6IjIws8VxOTg7Cw8Nx4cIFuLm5oW3btpg+fTq+/fZbDB48uO6DJSIiIiIyI6vowS/PzZs3UVhYiODgYOOy9u3bIyoqyoxRERERERGZh9Un+NnZ2ZDL5SbLFAoFsrKyHrvtypUr4ePjY3LTarW1FSoRERERUa0ze4I/ZswYiESiMm+P4+LigszMTJNlGRkZcHV1fey2CxYsQHJysslNKpVWeV+IiIiIiMzN7An+9u3bIQhCmbfHad68OUQiEa5du2ZcdunSJZOSHSIiIqByo66JRCLIZDK4uLjAxcWF13URkdWwiotsCwsLkZeXh7y8PACATqeDSCSCRCKBTCbDmDFj8Prrr2Pjxo2IjY3FunXrsH79ejNHTURElubhUddiY2MRGhqKFi1alJm8X7hwAS1btqzjKImIqsfsPfgV8euvv8LZ2RkDBw5EUlISnJ2d0aJFC+Pzn3/+ORwcHODn54f+/ftj8eLF7GkhIiITxaOuLV++vMSoa0REtsQqevBDQkLKLddRKBQIDw+vw4iIiMjalDXq2o4dO8rcpm/fvigoKECnTp3w4Ycfok2bNnURKhFRtVhFDz4REVF1VXbUtcjISMTExOD27dt48sknMWDAgBKDOjyMI7MRkaVggk9ERPVCZUdd6927NxwdHeHq6oply5ZBLBbj1KlTZbbPkdmIyFIwwScionqhuqOu2dnZVWh0NyIic2OCT0RE9cLDo65lZWUhKioK69atwwsvvFBi3WvXruHixYswGAzQarVYsmQJcnNz0a1bNzNETkRUOUzwiYio3ihv1DUXFxecOHECAJCcnIyJEydCLpfD398fZ86cQUREBBQKhRmjJyKqGKsYRYeIiKgmlDfqWnZ2tvF+nz59cP369boKi4ioRrEHn4iIiIjIhjDBJyIiIiKyIUzwiYiIiIhsCBN8IiIiIiIbwgSfiIiIiMiGMMEnIiIiIrIhTPCJiIiIiGwIE3wiIiIiIhvCBJ+IiIiIyIYwwSciIiIisiFM8ImIiIiIbAgTfCIiIiIiG8IEn4iIiIjIhjDBJyIiIiKyIWJzB0BUa1LvAJnxgFsDwLOpuaMhIiKyKTGaHCRk5EIld0agl8zc4dBDmOCTbUq9A1z8/u8Ev8PzTPKJiIhqSIwmB1vPxUGdroNS4YQJnf2Z5FsQluiQbcqML7p5t/z7PhEREdWIhIxcqNN1aObrAnW6DgkZueYOiR7CBJ9sk1uDolvK9b/vExERUY1QyZ2hVDjhVlI2lAonqOTO5g6JHsISHbJNnk2LynJYg09ERFTjAr1kmNDZnzX4FooJPtkuz6ZM7ImIiGpJoJeMib2FYokOEREREZENYYJPRERERGRDmOATEREREdkQJvhERERERDaECT4RERERkQ1hgk9EREREZEOY4BMRERER2RCRIAiCuYOwJF27doWbm5u5w6gwrVYLqVRq7jCqzRb2g/tgGSxpH5ycnLB3715zh0FmZK5jiiX9HdQ07pv1sdX9Aup+3yp6XGGCb+V8fHyQnJxs7jCqzRb2g/tgGWxhH4iqy5b/Drhv1sdW9wuw3H1jiQ4RERERkQ1hgk9EREREZEOY4BMRERER2RAm+FZuwYIF5g6hRtjCfnAfLIMt7ANRddny3wH3zfrY6n4BlrtvvMiWiIiIiMiGsAefiIiIiMiGMMEnIiIiIrIhTPCJiIiIiGwIE3wiIiIiIhvCBJ+IiIiIyIYwwbdSx44dQ58+fSCXy6FUKks8P2XKFDg6OsLFxcV4i4uLM0OkZXvcPqSnp2PcuHFwdXWFSqXCJ598UvdBVtKGDRtgb29v8r5v3rzZ3GE9ljW+14+yhs88UU15+HPu4uICsViMYcOGlbn+8ePHERwcDKlUis6dO+Py5ct1GG3FVXa/RCIRZDKZcf3BgwfXYbSVFxMTgyFDhsDDwwM+Pj54/vnnkZWVVeq6UVFReOqppyCVStG6dWscPXq0jqOtnMrsW2BgIJydnY3/b23atKnjaCvu1q1bGDBgABQKBQICAvDNN9+Uua4l/Z0xwbdSMpkML7zwAv73v/+Vuc78+fORnZ1tvPn7+9dhhI/3uH2YO3cu9Ho94uPjERERgffeew8HDhyo4ygrr3Pnzibv+3PPPWfukB7LWt/rR1n6Z56opjz8Oc/IyICvry/GjRtX6rqpqakYPnw4Fi5ciLS0NEyYMAHDhg2DXq+v46gfrzL7VezChQvGbSz9e2vGjBnw8PBAfHw8bty4gfv37+PNN98ssV5+fj6GDh2KYcOGIS0tDW+//TZGjhyJ5ORkM0RdMRXdt2I7d+40/r9du3atDiOtOIPBgGHDhqFnz57QaDTYsWMHXnvtNRw/frzEupb2d8YE30p16dIF//jHP9C0aVNzh1Jl5e1DTk4OwsPDsXz5cri5uaFt27aYPn06vv32WzNEatv4XhNZt4MHDyI7OxujR48u9fkdO3YgKCgIzz//PCQSCV599VUUFhbi8OHDdRxp5Txuv6xRdHQ0JkyYAGdnZ7i7u2P06NG4evVqifUiIyOh1WqxePFiSCQSjB8/HsHBwQgPDzdD1BVT0X2zJjdu3EBMTAz+85//QCwWo2PHjhg5cmSpx0dL+ztjgm/D1q5dCw8PD7Rr187qkrWbN2+isLAQwcHBxmXt27dHVFSUGaOqmCtXrsDb2xtBQUFYtGgRcnNzzR1Suaz5vX6UNX/miapq/fr1CAsLg7Ozc6nPR0VFoX379sbHIpEITzzxhMX/jT9uv4r17dsXvr6+GDJkiMX2BBebN28etmzZguzsbGg0GoSHh5daVhQVFYW2bdvCzu7vNM3Sv5crum/FJk+eDG9vb4SEhODkyZN1GGnFCYKAR+eDFQQBV65cKbGupf2dMcG3Ua+88gpu3ryJ5ORkfPLJJ1i4cCF++uknc4dVYdnZ2ZDL5SbLFApFmfV8lqJXr16IiopCUlIS9u/fj8jISCxcuNDcYZXLWt/rR1n7Z56oKjQaDfbu3YsXXnihzHWys7OhUChMlln633hF9gso6umOiYnB7du38eSTT2LAgAHIzMysoygrLyQkBNevX4dcLoe3tzckEgleeeWVEutZ4/9ZRfcNADZt2oSYmBjExcVh/PjxGDx4MGJjY+s44sdr0aIFGjZsiKVLlyIvLw+///47du7cCa1WW2JdS/s/Y4JvgcaMGQORSFTmrSI6dOgALy8viMVi9OnTB3PmzKnTU3vV3QcXF5cSX9IZGRlwdXWtrZAfqyL71KRJEzRp0gR2dnZo0aIFVqxYYdGnVAHLfK+rwtyfeSJz2Lx5M4KCgtC1a9cy13FxcUFGRobJMkv/G6/IfgFA79694ejoCFdXVyxbtgxisRinTp2qoygrp6CgAIMGDcKQIUOQk5ODjIwMNGjQAJMmTSqxrrX9n1Vm3wCgR48ecHZ2hrOzM2bPno0nn3zSIq+fcHBwwO7du/Hbb79BpVJh/vz5mDJlCho2bFhiXUv7P2OCb4G2b99uPC1U2q0q7OzsqrxtVVR3H5o3bw6RSGRyuvXSpUsmZSR1rSr7VNfve1VY4ntdE6zhvSeqrvXr12Pq1KnlrhMcHIxLly4ZHxeXGFjy33hF9qs0lvx3n5aWhvv37+Pll1+Gk5MT3NzcMHv2bPz8888l1g0ODsbVq1dRWFhoXGbJ38uV2bfSWPL/W5s2bXDkyBFoNBqcPHkSSUlJeOqpp0qsZ3F/ZwJZpYKCAiE3N1eIiIgQfH19hdzcXEGn0xmf37Ztm5CZmSkUFBQIJ06cELy8vIStW7eaMeKSHrcPEydOFIYPHy5kZmYKV69eFXx9fYWff/7ZjBE/3s8//ywkJCQIgiAId+7cEZ566ilh5syZZo7q8azxvX6UNXzmiWrShQsXBLFYLKjV6nLX02g0glwuFzZu3Cjo9Xrh448/Fho1amTyfWtJKrpfUVFRwoULF4T8/HwhJydHePvttwVfX18hLS2tbgKtgiZNmgjvvPOOoNfrhezsbGHatGlC9+7dS6yXl5cnBAYGCu+//76g0+mEH3/8UXBzcxOSkpLMEHXFVHTfYmNjhV9//VXQ6/WCXq8X1q5dK8hkMuH27dtmiPrxLl++LOTk5Ag6nU5Yv3694OXlJWg0mhLrWdrfGRN8K3Xs2DEBgMktICDA+HzPnj0FuVwuuLi4CK1btxbWrFljvmDL8Lh9SEtLE8aMGSPIZDJBqVQKH3/8sdlirah//etfgq+vryCVSgV/f39h/vz5QnZ2trnDeixrfK8fZQ2feaKaNHfuXGH48OGlPieTyYRff/3V+PjYsWNCmzZtBCcnJ6FTp07CH3/8UTdBVkFF9+vo0aNCixYtBKlUKnh6egoDBw4ULl26VIeRVt7ly5eFvn37Cu7u7oKHh4fwzDPPGBPb1q1bC5s2bTKue+XKFaFLly6Ck5OT0LJlS+Hw4cPmCrtCKrpv165dE9q1ayfIZDLB3d1dePrpp4Vjx46ZMfLyLV68WHB3dxdkMpnQu3dvk78dS/47EwmChZ4TISIiIiKiSmMNPhERERGRDWGCT0RERERkQ5jgExERERHZECb4REREREQ2hAk+EREREZENYYJPRERERGRDmOATEREREdkQJvhEREREZLR69Wp06tQJEokEYWFhNd5+TEwMRCIRXFxcjLdZs2ZVup3ExEQMGzYMKpUKIpEI169fL3f9JUuWwMHBweR1T5w4YXx+ypQpcHR0NHk+Li7O+Py5c+fQvXt3uLq6okWLFti7d2+lY36cqKgoDBw4EF5eXhCJRNDpdFVqhwk+ERERERmpVCq88cYbmD59eq2+jkajQXZ2NrKzs7FmzZoy1xOJRKUut7Ozw6BBg7Br164Kv+bo0aONr5mdnY2ePXuaPD9//nyT5/39/QEA6enpGDJkCGbOnIn09HSsWrUK48ePx+3btyv82hXh4OCAcePGYcOGDdVqhwk+EYp+1Xfq1AmdOnXCU089hREjRuDrr7+GwWAAABQWFmLbtm34xz/+gR49eqBv376YOnUqdu3aZVzn6NGjmDNnDkJDQ9GpU6ca/6MnIiLrYc3HlVGjRmHEiBHw8vIq8dy5c+fQq1cvuLu7o1WrVtixY0edxFQaX19fvPTSS+jSpUutv9bJkyehUCgwefJk2NvbY9CgQejatSs2btxoXCciIgKdOnWCQqFAhw4dTM4OVFSLFi3w4osvIjg4uFrxMsEn+n89e/bEwYMHsWvXLsycORMbNmzAxo0bIQgC/v3vf2PNmjV49tln8c033+D777/H+PHjER4ejjt37gAAcnNz0b59e7z88stm3hMiIrIEtnZcSUxMxKBBgzB//nxoNBps2LAB06ZNw19//VWl9oKCgqBSqRAWFoZ79+7VcLSlO3DgADw9PdGqVSt88MEHKCwsNHl+7dq18PDwQLt27fDtt98alwuCAEEQTNYVBAFXrlwBAFy+fBkTJ07EJ598ggcPHmDp0qUYMWIENBpN7e9UKcRmeVUiC+Tg4GDsrRg8eDAuXLiAX3/9FX5+fjh69Ci+++47tG7d2rh+w4YN0a9fP+Tn5wMAhgwZAgBISEio++CJiMji2NpxZePGjejXrx9GjBgBAOjatStGjhyJ8PBwvPXWWxVux8vLC+fOnUP79u2Rnp6ORYsWYejQobhw4QLs7e1rKXpg7NixmD59Ovz8/PDHH39g/PjxEIvFeO211wAAr7zyCj766CMoFAqcOHECY8eOhVwux+jRo9GtWzdoNBp88803eP75/2vvDkKiaOM4jn9XJkUoCBQiMtlAuggmVhq0gnaJoAJPhYQayV702EogkYEVROQhUapLXSLbS7XlJh5SXMEuHUJEjXQjUUIQpcxxd519D/vu8O5b+qJvsu3w+8AchoeZ/T8DO89/nvnPTB19fX0MDw9z4sQJAO7fv09jYyMejweA06dPU1paSm9vL3V1ddvWp/VoBl9kHTk5OUSjUfr6+igvL085CScZhkFubm4aohMRkUyT6eNKOBzmxYsX7N692156enqYm5sDEg/nulyudZeBgQEAdu7cyZEjRzAMg/z8fLq6uhgbG+Pjx48APHnyJOU3gJT1UCi0pfiLi4vZt28fWVlZHD58mNbWVvx+v91eVlZGfn4+hmFQXV1NU1OT3Z6Xl0cgEODhw4fs2bOH7u5uzp07R0FBgX1s7t27lxLnyMiIfXF2+fLlDY9NOBzeUp/WowRf5F/i8Tijo6MEg0GOHj3Kly9fOHDgQLrDEhGRDOWUcaWwsJDz58+zuLhoL9+/f6e7uxuA5uZmu5TlV0tVVdUv95tMcpMlMLW1tSm/AaSsJ2fJ/6+srKyfym42avd4PIyMjLCwsMDr16/59OkTx44dAxLHpqWlJSXO5eVlrly5AsCdO3c2PDZut/u39MmO/bfuTSSDDQ4OUllZyfHjx7l06RKVlZV4vd50hyUiIhkqU8eVWCyGaZrEYjEsy8I0TaLRKBcuXCAYDBIIBIjFYkQiEd69e7fpGvzkNpZlsbi4SHNzM0VFRRw8eHDTsZqmab9KMhKJYJrmukn78+fPWVhYAODDhw/cuHGDmpoau/3Zs2d8+/YNy7IIhUJ0dnamtL9//55IJMLy8jK3bt1ifn6e+vp6ALxeLw8ePCAUCmFZFisrK7x9+5aZmZlN9Scej2OaJqurqwCsrq5u6VWZqsEX+VtFRQU+n8+umTSMxN9j//79TE9Ppzk6ERHJNJk6rrS3t3P9+nV73e/3U19fz6NHj+jt7aWlpYWGhgZcLhclJSXcvXt3U/ufmpqitbWVr1+/smvXLjweD69evdpS/f0/y5kOHToEwPT0NG63m5s3bzI0NEQwGAQSCXxjYyMrKyvs3buXixcv4vP57O07Ozvxer2sra1RWFhIe3t7yncAOjo6ePnyJZZlUV1dTX9/P9nZ2UCivOfx48f4fD7Gx8fZsWMH5eXldHV1bao/nz9/Trm7kyxR2uhOw6+44pvdQsSB2tra+PHjB7dv3/6p7c2bN1y9evWnh6EgMcsRjUZTTjCzs7OcPXuWp0+fUlRUtO2xi4jIn0fjiqSTSnRE/sPJkyepqqqiqamJnp4eJicnmZmZob+/n4aGBvsrd0tLS0xMTDA1NQUkrsInJiZYWlpKZ/giIvKH0bgi200z+CJsPNMCsLa2ht/vJxAIEA6HycnJwe12U1NTw6lTpzAMg0AgkHJLM+natWucOXNmu7sgIiJ/EI0rkk5K8EVEREREHEQlOiIiIiIiDqIEX0RERETEQZTgi4iIiIg4iBJ8EREREREHUYIvIiIiIuIgSvBFRERERBxECb6IiIiIiIMowRcRERERcRAl+CIiIiIiDqIEX0RERETEQZTgi4iIiIg4iBJ8EREREREHUYIvIiIiIuIgfwEL+53rXlVOwQAAAABJRU5ErkJggg==", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAqAAAAIrCAYAAAAwSyveAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjgsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvwVt1zgAAAAlwSFlzAAAQ6wAAEOsBUJTofAAA04NJREFUeJzs3Xd4VFX6wPHvnZLJTCbJpJKEGnoJAgLSUZEqxVUpigVEsSy4ru6iuOiirmBBV91FV10F/amsgoIdBRQVC1Va6DWBEAjpmZSp5/dHYGBIAklI5/08zzw655659713hpk3556iKaUUQgghhBBC1BBdbQcghBBCCCEuLZKACiGEEEKIGiUJqBBCCCGEqFGSgAohhBBCiBolCagQQgghhKhRkoAKIYQQQogaJQmoEEIIIYSoUZKACiGEEEKIGiUJqBBCCCGEqFGSgAohhBBCiBolCagQl6gffvgBTdP49NNPq/1Yhw8fRtM03nnnnWo/1rlOn+cPP/xQ48c+11VXXcVVV11V22FclHfeeQdN0zh8+HBth1IjLrXzFaKmSAIqRAWd/kE6/TCbzXTo0IGZM2eSk5NTZccpKiriiSeeqBOJU3l98803PPHEE7UdhiinDz/8kJdffrm2wxBCXIIkARWikubMmcN7773Hyy+/TEJCAs899xzDhg1DKVUl+y8qKuLJJ5+sdwnok08+WaK8efPmFBYWctttt9VCVKIslUlAb7vtNgoLC2nevHn1BCWEuCQYajsAIeqra6+9lq5duwJwzz33MG7cOD7++GN+++03+vbtW+n9ejwe3G53FUVZktPpRKer2b89NU0jMDCwRo8pqlZ+fj5BQUHo9Xr0en1thyOEqOekBVSIKnK6b9/pvmJer5cXXniBDh06YDKZiI2N5f7778dut/u9TtM0/vznP/N///d/tG/fHpPJxP/+9z/CwsIAePLJJ323+0/f3i6rL+HkyZNp0aKF7/npvpcvvfQSL774IvHx8ZjNZo4ePeqr43a7mTlzJo0aNcJqtXLjjTdy/Phxv/2uWbOGcePG0axZM0wmE02bNuXBBx+ksLDQ79ivvPKK75xOP86O49w+oN999x39+vXDYrEQFhbGuHHjOHToUIlzstlsHDlyhDFjxmC1WomKiuKvf/0rHo+n7DfkAn799VeGDBlCSEgIQUFBDB48mE2bNvm2z5s3D51O53etTrvvvvsIDg72O/8vv/ySvn37EhQURGhoKNdffz379++vVGynPxOLFi2iffv2WCwWBg4cyL59+3yxNWvWDLPZzJgxY8jMzCyxjwvFc9VVV/HZZ5+RlJTke69Of3ZO95tdsmQJjz76KHFxcQQHBwNl94n86quvGDhwIFarldDQUPr27XvB/sWn39v9+/czZMgQgoKCaNq0Kc8//3yJui+88AJ9+/YlIiICs9lM9+7d+fjjj8u8dp988gmdOnXCZDLRqVMnvvnmmxJ1jxw5wu233050dDQmk4nLLruM//3vf+eNGWDjxo0MGzaMyMhIzGYz8fHxTJky5YKvE0KcIS2gQlSRgwcPAhAREQHAnXfeyaJFi5gyZQp//vOf2bdvH/Pnz2fnzp2sWrXKl5wBrFixgo8++ohp06Zhs9no1q0bb7zxBvfccw/XX389N9xwAwCXXXZZpWJ76623cLlc/PGPf0Sn02G1Wn3bnnrqKQwGA3/7299ISUnhlVde4cCBA2zYsAGj0QjAkiVLKCgo4L777iMiIoL169fz73//m6NHj7JkyRKguBX4+PHjfPvtt7z33nsXjGnVqlWMGDGCtm3b8o9//IPc3FxeeeUV+vXrx9atW4mKivLVdblcDB06lH79+vHCCy+wcuVKXnzxRVq1asV9991X4euxatUqrr32Wnr37s1TTz2F1+vlzTffZODAgWzYsIGOHTsyfvx4Hn74YZYsWcKDDz7oe63H42Hp0qWMGTMGs9kMFCdlU6ZMYdSoUTz//PPk5eXx73//m/79+7N161YaNWpU4RhXr17Np59+yh//+EdcLhdz587l+uuvZ+LEiSxbtoy//vWvHDp0iFdeeYUZM2bw9ttv+15bnnhmzZqF3W4nKSmJl156CcDvcwHwxBNPYLFYeOSRR8jLyysz1rfffpu77rqLLl26MGvWLIKDg9m0aRMrV67kD3/4w3nP0+VyMXz4cAYMGMDzzz/Pl19+ySOPPALAww8/7Kv3yiuvMGbMGG655RacTicffvgh48aN48svv2TkyJF++/zxxx9ZsmQJf/zjH7FarfzrX//ixhtvJDk52ffvMzU1ld69exMQEMADDzxAeHg4n332GRMnTsThcDB58uRS401LS2Po0KHEx8fz2GOPERQUxKFDh1i2bNl5z1MIcQ4lhKiQhQsXKkCtXr1anTx5UiUlJam33npLBQYGqujoaJWfn69++uknBaiPP/7Y77UffvihAtTy5ct9ZYDS6/Vqz549fnWzsrIUoGbPnl0ihiuvvFJdeeWVJconTZqkmjdv7nt+6NAhBSibzabS09P96q5evVoBqlmzZiovL89X/u677ypAvfPOO76ygoKCEsd65plnlKZpKikpyVf2wAMPqNK+Vk7HsXDhQl9Z165dVUxMjMrKyvKV/fbbbwpQDz30kN85AWru3Ll+++zWrZvq3r17iWOd6/R5rl69WimllMfjUa1bt1ajRo3yq5edna1iYmLUhAkTfGW9evVSvXv39qu3atUqBajPPvtMKaVUXl6eCg0NVdOnT/erd/DgQWU2m9UjjzziKyvrfTsXoAIDA1VycrKv7LnnnlOAatOmjSoqKvKV33zzzcpkMimn01nheK677jq/z8tpp69ZmzZtVGFhod+205//Q4cOKaWKr5vValV9+/b1i0sppbxe73nP8/R7e/b77fF41DXXXKOsVqvf5/Lcz6DT6VQJCQlq0KBBfuWAMplM6uDBg76yrVu3KkD9+9//9pVNmTJFNWnSxO/zp5RSw4cPV7Gxscrj8ZR6vsuWLVOASktLO++5CSHOT27BC1FJV199NVFRUTRv3py77rqLFi1a8MUXX2CxWPj4448JDw/nyiuvJD093fcYOHAger2+xMCiQYMG0bZt22qLddy4cb6Wn3NNmjTJr+Vr4sSJhIWFsXz5cl/Z6ZY+KO4LmJ6eTt++fVFKsXnz5grHk5qaypYtW5gyZQo2m81X3rt3b3r37s1XX31V4jX33HOP3/MBAwb4Wp0rYuvWrezfv5+bb77Z771xuVwMGDDA770ZP34869atIzk52Ve2ePFiQkNDGTZsGAArV64kJyeH8ePH++0vODiYLl26VHoQ2dChQ2natKnvea9evYDi98dkMvmVOxwOUlNTqzyeyZMnX7Dv7ooVK7Db7Tz66KN+cQF+rfznM23aNN//63Q67rvvPux2Oz///LOv/OzPYFZWFjk5OQwYMIDff/+9xP6GDRtGfHy87/lll11GSEiI7/OilPK1Yrvdbr/rNHz4cFJTU9m7d2+psZ7+vC5btgyv11uu8xNClCS34IWopNdff51WrVphMBiIi4vzSyD37dtHZmam323ks508edLv+dk/ltXhfPtv06aN33ODwUCLFi38+vglJyfz97//nc8//5ysrCy/+pWZeiopKQmAdu3aldjWvn17PvzwQ78yq9VKeHi4X1lYWJhfLCdPnvTrE2q1WkvcUgZ8/ShvueWWUmM7e4DW+PHj+etf/8rixYt9fU6XLl3Kdddd50u2Tu9v4MCBpe6vZcuWpZZfSLNmzfyeh4aGAvglpWeXZ2Vl0axZsyqNpzyfy9NJXUJCQrn3e7bTn7eznf5Mnv0Z/PLLL3n66afZsmULDofDV15aknvutQP/z8vJkyfJzs7mtdde47XXXis1rpMnT9K+ffsS5VdeeSU33ngj99xzD48++iiDBg1izJgxTJgwgYCAgAuerxCimCSgQlRSr169fKPgz+X1eomNjeX//u//St0eFxfn9/zs1p3y0DSt1OmeyhqUU9H9n7vPIUOGkJmZySOPPEL79u0JCgoiJSWFyZMn10grUHlGXffs2dOX2ALMnj271DlJT8f70ksvXTBpatKkCX369PEloKtXryY9PZ0JEyaU2N+iRYtK/YOjste+rHMuq/z056Eq47mYz01VWrNmDWPGjGHgwIG89tprxMbGYjQaWbhwIYsWLSpRv7zXaPLkyWX+IVLWZ0PTND7++GPWrVvHF198wbfffsvtt9/OCy+8wC+//FLqHz1CiJIkARWiGrRq1YrVq1czYMCAErcly+t8ty/DwsJKvf18dgJWXqdbzE5zu90cPnyYIUOGALB9+3b27t3Lu+++y+233+6rt3LlygrFfLbTc0ju2bOnxLY9e/ZUao7JDz74wG9Uelktfa1atQKKb6UOHjz4gvsdP348f/7znzl06BCLFy8mLCzMd23O3l9sbGydWOWoIvGU9/0qz/ESExNLtGSWx+nP29nv1+nP5OnPwSeffEJgYCDffvut37+nhQsXVirmqKgogoODUUqV6zNQml69etGrVy+efvppFi9ezIQJE/joo4+48847K7U/IS410gdUiGowduxYnE4nzz77bIltDoeD3NzcC+7DbDajaRrZ2dkltrVq1Yrdu3f73crfunUrv/zyS4Vjfffdd/2mhlq0aBFZWVmMGDECONOadHaLq1LKN+XS2YKCggBKjflssbGxdO3alYULF/rdwl+/fj2//fZbiVHN5dGvXz8GDx7se5SVgF5++eW0bNmSF154gYKCghLbz+0eMW7cOHQ6HYsWLWLZsmXccMMNvtkBoLivZkhICHPnzi11/tb09PQKn8vFqEg8QUFBF71615AhQ7BarcydO9fv1jhQ7kUZXn31Vd//e71e/vOf/xAUFMSAAQOA4s+gpml+LfyHDx+u9DKyer2e66+/no8++qjUP4LO/QycLSsrq8R5nb4TUlRUVKl4hLgUSQuoENXg6quv5q677uKJJ55g06ZNXHPNNeh0Ovbu3cvixYv54IMPLtjyEhAQQOfOnfnoo49o27Yt4eHhJCQkkJCQwJQpU/jnP//JsGHDuPPOO0lLS+P111+nU6dO5UpuzxYaGsrAgQOZNGmSbxqmhIQE363J9u3b06pVK/7617+SkpJCSEgIn3zySYm+oADdu3cH4E9/+hPDhg1Dr9dz0003lXrcefPmMXz4cPr168eUKVN80zDFxsYyc+bMCp1DRej1et58801GjhxJ586dmTRpErGxsRw9epQVK1bQunVrv2mk4uLi6NevH8899xx5eXl+t9+h+PrNnz+fSZMm0aNHDyZMmEBERASHDx/m888/5w9/+ANPP/10tZ3PuSoST/fu3fnggw946KGH6NmzJ1arldGjR1f4eC+++CL33HMPvXr14qabbiIkJITff/8dk8nkl1yWxmKx8Nlnn5GVlUX37t358ssv+e6773jmmWd8t7NHjhzJP//5T4YPH87EiRNJS0vj1VdfpXXr1mzbtq1S1+nZZ59l9erV9OjRg7vvvpv27duTnp7Ohg0b+P3338tc+/3dd9/ltdde4w9/+AOtWrUiPz+ft956i5CQEK699tpKxSLEJam2ht8LUV+dnpZl8+bN563n9XrVf/7zH9WtWzcVGBioQkNDVdeuXdWsWbPUyZMnffUA9cADD5S6j7Vr16oePXook8lUYkqm999/X7Vs2VIFBASorl27qm+//bbMaZheeumlEvs+PdXOkiVL1MMPP6yioqKUxWJR119/vUpJSfGru3PnTjV48GBltVpVZGSkmjp1qm9qm7OnVvJ4POqBBx5QUVFRStM035RMpU3DpJRSK1euVH369PFdnxtvvNFv+hyliqfqCQ0NLRH/7NmzS53yqazzPD0N02mbNm1S1113nQoPD1cmk0nFx8erW265Rf38888l9jF//nwFqMjISOV2u0s9zqpVq9TgwYNVSEiIMpvNqk2bNmrq1Klq+/btvjoVmYbp3M/E5s2bS72GZX0eyxNPfn6+uvXWW1VYWJgCfJ+d09ds2bJlJWI7d1qi05YtW6Z69+6tzGazCg0NVX379vVNVVWW0+/tvn371ODBg5XZbFZxcXHqmWeeKVH37bffVm3atFEmk0m1b99eLVy4sNTPQFn/npo3b64mTZrkV5aamqruvfde1aRJE2U0GlVsbKwaNmyYeu+998o8399//13dfPPNqlmzZspkMqno6Gg1atQotXHjxvOeqxDCn6ZUFS1cLYQQQlTA5MmT+fTTTy/YZUMI0fBIH9BSVPQWlBBCXCrk+1EIURUkAS2FdCQXQtRHd999N40bNyYkJIQWLVowd+5c37YWLVpgNpt986N26tSpUseQ70chRFWQBFQIIRqIP//5z+zfv5/c3FzWrFnD+++/z+LFi33bly1bht1ux263s2PHjlqMVAhxqZNR8EII0UB07NjR77lOp2P//v21FM2FvfPOO7UdghCilkgLqBBCNCCPPvooQUFBNGvWjPz8fG699VbftkmTJhEVFcVVV11Vrjlj582bR3R0tN+jtLlThRCioiQBFUKIBuSZZ57Bbrezfv16Jk6cSFhYGADvv/8+hw8fJjk5mQkTJjBixIgLrpw1Y8YM0tLS/B4Wi6UmTkMI0cDViQR0/vz59OjRA5PJVOak1af9+OOPJCQkYLFY6NmzJ1u3bi2xr8aNG2O1WrnxxhtLnSy7qnm9MpOVEKLu0DSNnj17EhgYyOzZswHo378/ZrMZs9nMfffdR7du3Vi+fHm1xuE5tea6EEKcq04koHFxcTz22GNMnTr1vPUyMjK47rrrePjhh8nKyuLmm29mzJgxvuXfVq5cyezZs/niiy9ITU1Fr9dz7733Vlvcq/enM/7/NtJv/i/cs3grOYXOajuWEEJUlNvt5sCBA6Vu0+l05V4qs6IK3S4eWPspPb94hauX/4fPkhKr5ThCiPqrTiSgN9xwA3/4wx+IjIw8b72lS5fSunVrbr/9dkwmEw8++CBer5dVq1YBxR3a77jjDi6//HKCg4OZM2cOS5cuvei1jktzJKuQf6zYy8o9aaxLyuLNdclEPP4tm5IzqvxYQghxIVlZWbz33nvk5ubi9Xr55Zdf+M9//sPgwYNJTk5mzZo1OJ1OnE4n//3vf9mwYQNDhw6tllge//0bPkveQaajgEN5mfx1wxccyE2/8AuFEJeMOpGAlldiYiJdu3b1Pdc0jcsuu4zExMRSt7dp04aAgAB2795d5j4r28l+x4k8DmYWkF3k4XQbgkdBj1d+peVTX1Xm9IQQotI0TWPhwoU0b96c0NBQ7rzzTv7yl78wffp07HY7999/P+Hh4cTExPDuu+/y5Zdf0qpVq2qJ5acTB/2eu7xePk+WaZ+EEGfUq2mY7Ha7r0P9aTabjby8PN92m81W5vbSzJgxgxkzZviVDRky5IKxNAkNxO5wlbrtUI4X7S9foF6UFUOEEDXDZrPx/fffl7qtY8eObNmypcZiCQswk1Zo9ytrHBRaY8cXQtR99aoF1Gq1lridnpOTQ3BwcLm2V6VOMcH0bRF+3jraX77AIwOUhBCXmAc7DcSkP9W+oUFHWzTXN+9cu0EJIeqUepWAJiQk+P0Vr5Ri27ZtJCQklLp9//79OBwO2rdvX+WxaJrGZ1OuwGLQzlvPMONLlu84VuXHF0KIump4k/Z8OWQKD3Tsz7Pdr+WTQZMx6vS1HZYQog6pEwmo2+2mqKgIt9uN1+ulqKgIl6vk7e0bbriBffv28f777+N0OnnllVcAGDx4MACTJ09m4cKFbN68GbvdzmOPPcYNN9xAaGj13PrRNI3850ZdsN61Czbx9s9l90MVQoiGpk1IFA8mXMlNLbthNhhrOxwhRB1TJxLQp59+GrPZzJw5c1iyZAlms9k3JZPVamXNmjUARERE8Omnn/Lss88SGhrKBx98wOeff47JZAKK+24+8cQTjBw5kpiYGJxOJ6+//nq1x1+evp53LdtHx398Ue2xCCFEXXG8IBe7y1HbYQgh6iBNVddEcPXYkCFDWLlyZYVfp/2lfAmmDE4SQtRX5fl+zCjK54XEH0nKz8Kg6RjTrCNjW3SpoQiFEPVBnWgBbSjKm1hqf/mCtFxpFRBCNEz/O7iZpPziVejcysvSpEQO5skcyUKIMyQBrWLlTUIbPbmCgxn2C1cUQoh65pA9s0SZJKBCiLNJAloNypuEtpq7mpe+31PN0QghRM1qGxJVoqxdSHQtRCKEqKskAa0m5U1CH/pqLwP/9SNOt7eaIxJCiJpxc8tudLI1AsCsN3J7q+40tdpqNyghRJ1Sr1ZCqm/Ui6PLNTBpTVIuPed9y7q/DiXQKHPlCSHqt5CAQGZ1GUyeq4hAvVHmABVClCAtoNWsvC2h29LdmGd+zafbU5GJCYQQDUGwMVCSTyFEqSQBrQHqxdFElbOt+fp3NjLyv+s4mJ5fvUEJIYQQQtQSSUBrSNpzoxnVpnxr0i/fc5Ib39nAL4dKjiQVQgghhKjvJAGtQV/cexXvju1YrrpbUvO4d8lW0u0yX6gQQgghGhZJQGvY7X1asfmB3uWqm3jCTrcXfyQr31nNUQkhhBBC1BxJQGtB12ZRZD01tFx1j+Y6GPn2umqOSAghhBCi5kgCWktsQSZSZl1drrq/JWVz50ebqzkiIYQQQoiaIQloLYoLt5I/Z3i56i5Yf5Q7/7eJwxkF1RyVEEIIIUT1kgS0llkCjeQ+PaxcdRdsPEarud/x/saj1RyVEEIIIUT1kQS0Dgg2B+CeN6pcdb3Abf/bzBUv/8hdH23hhdUHSM6SVlEhhBBC1B+SgNYRep2G94VRBJXzHdlwJJd31x9h4YYkHlu+B4fbU70BCiHqvLvvvpvGjRsTEhJCixYtmDt3rm9bYmIivXv3xmKx0LFjR77//vtajFQIcamTBLQO0TQN+7zRdI0NKld9N7DzRD7Ld59g89Gc6g1OCFHn/fnPf2b//v3k5uayZs0a3n//fRYvXozL5WL06NGMGTOGrKwsZs+ezfXXX09aWlpthyyEuERJAloH/f6XqxnXuVG566fnu3h8+W68XllDXohLWceOHTGbzb7nOp2O/fv388MPP1BQUMDMmTMxmUxMmDCBhIQElixZUovRCiEuZZKA1kGaprF48hV8e2ePcr/mx4OZ/P3b3WTIpPVCXNIeffRRgoKCaNasGfn5+dx6660kJibSuXNndLozX/ldu3YlMTHxvPuaN28e0dHRfo+CAulzLoS4eJKA1mFDO8by2DWty1XX5VXMW32AYW+uJadQklAhLlXPPPMMdrud9evXM3HiRMLCwrDb7dhsNr96NpuNvLy88+5rxowZpKWl+T0sFks1Ri+EuFRIAlrH/ePaDiyd1L1cb5TTo9h0NIcuL/zIoXR7tccmhKibNE2jZ8+eBAYGMnv2bKxWKzk5/v3Ec3JyCA4OrqUIhRCXujqRgGZnZzN+/HiCg4OJi4vj5ZdfLrPue++9R7t27bBarQwcOJA9e/b4tv3www/odDqsVqvvcfYo0Prq+sviKHzuWgxa+eonZRfR9Z9rmPX1ruoNTAhRp7ndbg4cOEBCQgLbt2/H6/X6tm3ZsoWEhIRajE4IcSmrEwno9OnTcTgcpKSk8O233zJ37lyWL19eot7PP//MAw88wP/+9z+ys7MZNGgQY8aMwe12++pER0djt9t9j7/97W81eSrVJsCgJ+nxa9CXs36uw83LPx1k2zEZHS/EpSArK4v33nuP3NxcvF4vv/zyC//5z38YPHgwV111FWazmeeffx6Hw8GSJUvYvn0748aNq+2whRCXqFpPQPPz81myZAlz5swhJCSEzp07M3XqVBYsWFCi7meffcbYsWO5/PLLMRgMPPbYYxw6dIg1a9bUQuQ1Ly7UQsbTwwgoZ0togcvL7Ys2U+iSOUKFaOg0TWPhwoU0b96c0NBQ7rzzTv7yl78wffp0jEYjn3/+OcuWLcNms/H3v/+dpUuXEh0dXdthCyEuUbWegO7duxev1+t3K6is0ZlKKZRSfs8Btm3b5ivLyMggJiaG5s2bc88995CZmXne49e3UZ6h5gB+mNYHSznvx29NzaPTc99zLKewmiMTQtQmm83G999/T1ZWFnl5eezevZtHHnkETSv+rujcuTPr1q2jsLCQXbt2cc0111RrPHtzTvLe/k18kbyDfJejWo8lhKh/aj0BtdvthIaG+pWVNTrz2muvZcmSJaxfvx6n08mTTz6J2+32JYzt27dny5YtHDt2jJ9//pmjR48yadKk8x6/Po7y7BMfSebTI+gUXb44D2UV0eG51Qx7cy3zVu/jpwPp5BW5L/xCIYSohHUnk3hyywqWp+zmf4e28PfN3+Lyyp0YIcQZtZ6AWq1WcnNz/crKGp05aNAgnn/+eW6//Xbi4uIoLCykY8eONGnSBICYmBg6deqETqejadOmzJ8/n6+//rpOt2hWlsmoZ8W9/WgcElCu+rkOD6v2nmTOqv089PkO/vRpIompuRd+oRBCVNBXR3Zx9rIYqYV5bEo/WmvxCCHqnlpPQNu2bYumaezYscNXdr7RmXfffTe7d+8mPT2dxx57jMOHD3PFFVeUWlen05W4bd+QxIUG8vtfrqJdVPmW7vQqyClysy89n9wiF4s2p1RzhEKIS5FbeUuUSQuoEOJstZ6ABgUFMXbsWGbNmkVeXh6JiYm89dZbTJkypURdh8PB1q1b8Xq9nDhxgrvuuosbbriBdu3aAbB69WoOHz6MUorjx4/zpz/9iaFDhxIUVL4ErT6KtprY+fDV3NunGVZj+d7OvCIP65KzSc6SfqFCiKo3OLaN33NbQCA9IpvWUjRCiLqo1hNQgFdffRWj0UhsbCxDhgxh5syZjBgxAii+RX96lLvT6WTSpEmEhISQkJBAXFwcb7zxhm8/mzdvZsCAAQQFBdG9e3ciIyN57733auWcapJOp/GfsV1YdW8fOkZfONlWwIm8ItYnZ5FT6MLrbbitxEKImjcorg0PdOxPr8hmDG/cjqe6DcdsMNZ2WEKIOkRTknmUMGTIEFauXFnbYVRKvsNN/JxVnMx3XbCuBoSY9DS2BdIk1MLknk25+fLG1R+kEKLeqs/fj0KIuqNOtICKqhNkMvDqDZ0pzyRNCshxeNh1Ip9fD2Uwa/luvt51vLpDFEIIIcQlThLQBmhslzg6xwaX+81VQL7Ly/HcImZ9vYcCp0zRJIQQQojqIwloA6RpGivv6c2YTo2wmvSY9Bd+mxVQ6Pay96Sd2xZtrv4ghRAN2prjB5m3fTVv7V3HicKS8zoLIS5thtoOQFSP6OBAlk25gp3H8/h2Txqzv91DnuPC06C4PYrfkrLYdSKPDo1KzsUqhBAXsiJlD+/s3+h7vjH9CP+8YgwWQ/nmLRZCNHzSAtrAdYwJ5oEBLendPIwA/YV7hjq9inS7k4MZ+TUQnRCiIVqdesDvea7LIRPRCyH8SAJ6CdDpNIa0jaJfi3DMBg29BqenDNWVkpO6vIpXfz7MoVKS0AKnG5en5CTTQghxmkmvL1EWUEqZEOLSJbfgLxH39GlOoEGPzWzkWG4RkUEBpOQUkZRVQFZhyUFHq/adZPAba+nYKJg/dGqEXqdxMLOAA+kFmAw6rkuI4bqEmFo4EyFEXTe6aSf+ueNH33KcTS2hXB7RpFZjEkLULZKAXiJCAo3cPyCe+wfEo5TC7VV8ueME93y8FQ04dzJYlxdSsgs5llPEzuN5RFkDSLM7uSw2GIWBj7Yco0MjK22jrLVxOkKIOqx7ZBOevnw4608eIcxkZkCjlhh10gIqhDhDEtBLkKZpGPUawYEGjHodBr2Gy1NyPQKHRwGKpKxCjucVYTLoOZZbROvI4qRzT5pdElAhRKnigyOID46o7TCEEHWU9AG9hHWODSlu/VScd+J6j1I43F7yHG6SsgrxeIv7gLYIt9RInEIIIYRoWCQBvYQ1CjZxf/94NEC7wAB5jwKvUhQ4PexPL2Bgy3B0miaT1gshhBCiwiQBvcQ9ck0bJnSNpXFIIBbj+T8OwQF6WoZbKHC5eeGHA9z47gbGvbuJTUeyayZYIYQQQjQI0gdU8OKYBP798yF2pdnZdCSbpKzCEoOSAIx6HYFGHXtT8zGcmr8pI9/JSz8d5F9/SCA8SCaZFkJAkdvF2/vWs/ZkEmEBFm5u2ZU+0S1qOywhRB0iCaggOtjEP0a0x+5wk1fkYsR/17It1e6XhJoNOmJCAkmzO/AqhVcVJ6B2h4flu9K4172NFuEW/jQgniY2c+2ciBCiTvjw0BZ+STsMQLojn/m7fqFlcASNzLK6mhCimNyCFz5Wk4Ho4EB6t4jgimY2jDow6jSCTXp6Nw9jVMdoOseEYDKc+di4vAqFwqDXcTSniDd+S6rFMxDi0uVwOLjrrruIj48nODiYTp06sWjRIt/2Fi1aYDabsVqtWK1WOnXqVG2xbM9K9XuugMSs49V2PCFE/SMtoMKPXqdxe/cm5DvdhJkDOJpTSLtoK7d3b8LoTo14f+NRfj2cRY7LhffU6Pmos269bziSzbPf7SPAoOPa9tG0bxRMUmYB6flOOsYEYzbKXIBCVAe3201cXBzfffcd8fHx/PLLL4wcOZL4+Hj69OkDwLJlyxg+fHi1x9LYEkpqYV6JMiGEOE0SUFFC3/hwOsUEsz89n+ZhZiKtJt+2Q1mFhFmM5Ls8QPHI+EKXl5N5DtLsDo7bHTg9XhpZTWxOyaFTo2C2pRb/EFmMemZe05rWkUG1cl5CNGRBQUE89dRTvuf9+/enX79+/Prrr74EtKZMiO/KwbwMMp2FAFwd04r2tugajUEIUbfJLXhRqlCzke5NbX7JZ77Dze40O2GWAMItAYRZjISZjaDB9uO5pOYV4VWKncfzOJxZQHahmy92nPC9vsDlYdHvKbVxOkJccvLz89m4cSMJCQm+skmTJhEVFcVVV13FL7/8csF9zJs3j+joaL9HQUHBBV/XOCiUl3tdx+NdBvNiz9FMbdf7os5FCNHwSAIqys1k0GEx6gnQ6zDqNQL0OsxGPW2jgmgcGnhqXlAvOUVutqXmkpZXhF5/erCSi80p2fzfhmT+/s1udh7Pu8DRhBCV5fV6mTx5Mj179mTo0KEAvP/++xw+fJjk5GQmTJjAiBEjSEo6f5/tGTNmkJaW5vewWC68AIVSim9T9vBC4o/8df0XLNi7nmxHYZWcmxCiYZAEVJSbQa9jfNdYYoJNvoFIjUMDiQ0JxGzQYdDr8Hi9uDwKp0dxKLMAHZBd6GJtUjbJWYVkFbn5aMsx5ny3j5N2R+2ekBANkFKKe++9l2PHjvHRRx+hnVplon///pjNZsxmM/fddx/dunVj+fLl1RLD/x3YyJ/WfsriQ1v4NDmRmRu/4s/rP+Nkkb1ajieEqH+kD6iokKHtoukUE8z21Dxyi1w0DjVzeeMQbnhnI/lOO14FOh0EGfXEhZgJtwSw40Qe2UUuNMDrVRS5PBzPdbDpaA7D20u/MCGqilKKadOmsWXLFlatWoXVai2zrk6nQ6nSZvy9eO/u20i2qwi31wsa5Lud7MlJY9Wxfdzcslu1HFMIUb/UiRbQ7Oxsxo8fT3BwMHFxcbz88stl1n3vvfdo164dVquVgQMHsmfPHr/t8+fPp3HjxlitVm688UaysrKqOfpLT+NQM8PbRzO+a2P6xYeT6/AQaNSh1zQ0DfSahjlAT5BJz56TdvIdHkx6HUpBvtNNZoGLgxn5/Ho4E7tDlvIUoqpMnz6dtWvX8u233xISEuIrT05OZs2aNTidTpxOJ//973/ZsGGD7/Z8Vct2FpLvcuBSHlxeDy7lxeX1UOh2VcvxhBD1T51IQKdPn47D4SAlJYVvv/2WuXPnlnpr6Oeff+aBBx7gf//7H9nZ2QwaNIgxY8bgdhcnMStXrmT27Nl88cUXpKamotfruffee2v6dC45Px/KQK9ptIoIIijAgMmgw6jTSMkpwu7wUOh2U+jy4FXg9hYPRgow6Nh3Mp8XfjhQ2+EL0SAkJSXx2muvsXPnTpo2beqb73Pu3LnY7Xbuv/9+wsPDiYmJ4d133+XLL7+kVatW1RJLpCkIhTq1aIXC6/USagxkYEzLajmeEKL+qfVb8Pn5+SxZsoRNmzYREhJC586dmTp1KgsWLGDEiBF+dT/77DPGjh3L5ZdfDsBjjz3G3LlzWbNmDVdffTXvvPMOd9xxh2/7nDlz6NixIzk5OYSGyhx01SUlp4hNKTkoBUEBenQaBOh1hAcFYA0wsOloNka9Dp0GTk9xncahgWiaxt6T+RzPLSImJLC2T0OIeq158+bnvaW+ZcuWGotFAerUamnq1KOjrRGtQyJrLAYhRN1W6y2ge/fuxev1+k0V0rVrVxITE0vUVUr5fcGe/v9t27YBkJiYSNeuXX3b27RpQ0BAALt37y7z+JWdZkSckZRZgEFX/FHS6zR0Oo1BrSNpFRFEtDUAq8mAXtMICtATYjIQGmhAO/VaDQiUyemFaFCO5mfDqcV8NUDT4Ne0JDZnyDRsQohitZ6A2u32Eq2TNpuNvLyS0/Rce+21LFmyhPXr1+N0OnnyySdxu92+hNFut2Oz2cq1r9MqO82IOCPf5aVLbAjNbGaahAbSJS6Uga0iTv3waHSIDsISoMdqMhJsMqDTNCJOrZ40qE0kNrOxdk9ACFGlHB43LuX1tX6evhW/Mf1IbYcmhKgjav0WvNVqJTc3168sJyeH4ODgEnUHDRrE888/z+233056ejqTJk2iY8eONGnSxLevnJyccu1LVJ1ezWys3JtOE5sZKF7x6IbLYunR1Maagxn8mpSFhkaOw41RpyMiyMgNnWNoE2nl8ibSNUKIhsbucfo9P52ERgbKKmhCiGIVbgF1u93cdNNNHD58uEoCaNu2LZqmsWPHDl/Zli1b/G7Jn+3uu+9m9+7dpKen89hjj3H48GGuuOIKABISEvz6Oe3fvx+Hw0H79u2rJFZRulsub8K1HaJpFBxA59hg/ja4NWajnnbRVu7q3RwURFpNtIoIolmYmaAAAx0aBdO9qc03R6EQouEwajrOdLQp5vC6GBLXtpYiEkLUNRVOQA0GA9nZ2VUWQFBQEGPHjmXWrFnk5eWRmJjIW2+9xZQpU0rUdTgcbN26Fa/Xy4kTJ7jrrru44YYbaNeuHQCTJ09m4cKFbN68GbvdzmOPPcYNN9wgA5CqWYBBx63dm/DSdQk8ek0bWkb4t3Kcvt3uV2YpWXY2p9vLgfR8CpwyTZMQ9U0Laxhe8OvrHWsOITlfpsUTQhSrVB/QsWPH8t577+HxeKokiFdffRWj0UhsbCxDhgxh5syZvhHwVquVNWvWAOB0Opk0aRIhISEkJCQQFxfHG2+84dvPkCFDeOKJJxg5ciQxMTE4nU5ef/31KolRVN7Ebo0x6M60hkQFGXllzSFmfb2LdUklf5ASU3O59+Nt3L1kK9cv3Mg3u9NqMlwhxEXqHtmUQL0BDQ0dGma9EaNOz69p51/6Uwhx6dBUJZbCeOSRR1i7di1Wq5XWrVtjNpv9tj/77LNVFmBtGDJkCCtXrqztMBqUrAInicfzOJJdyJc7zySUGvDk8Ha0jixuNVVKcd8n2/l+XzpOjxcoXoN+5T29aRYmg8OEqG3l+X58dOPXrE7dx9as1OLVkIAIk4V5PUdxW+seNRGmEKKOq1QLqMViYdCgQVxxxRWEh4f71hc+/RDiXGGWAAa0jOBYjv/67wrYkJzte17o8vD70WwKXWda1x1uLx9vSy2xzx/2pzN96Xbu+HALb69LxnUqYRWiPnA6nWzfvp2kpJKtgg6Hgy+//LIWoqoaV8W2wu5y4vZ6UaemY8p3Ozman3uBVwohLhWVGgU/e/bsqo5DXCLCLSWnXIoIKi4rdHl44YeDpOQUkV3oxmTQEWwyoNfBL4cyOZhRQOvIIG7r3oTcIjdvrk327eO7fenYzAZuvCyuxs5FiMo6cuQI06dP59ixY2iaRvfu3fnHP/5BZGTxRO12u52nnnqKUaNG1XKklTOscTuetwRztCAHL4pAnQGz3siunBO1HZoQoo6o9XlAxaVldKdG2Mxn/u4xG3Wk5zvZdCSbr3aeYHeanQ6NrOh1xQmp2+tFp2kUuDxkFrhYn5zNSz8dZHtqyZaUrcfKbl3JLXKx+WgO6XZHmXWEqCn/+te/iI+PZ+XKlXzyyScEBgZy5513cvz48doOrcpcE9saW4CZ8AALFkMAmqZxWVhsbYclhKgjan0eUHFpibKa+OeYTmxOyeGngxn8fjSHD35PQa9pmI169DqNIpeXAL0OlBev8hISaCLMfGbU/MGMAq5sGVFi33FlLOe5LimL1349jMuj0Glwa/cmDG8fXW3nKMSFbN26lddeew2bzYbNZuOll17iueee46677uL111+v912ZlFLEW8MJ0OvJdhQSbDRxWXgcd7frXduhCSHqCGkBFTUu0KinexMbW4/lsvVYLjuO57EtNZeNR7JIyixgy7FcCl1ePApigs3kOdx4vGf6dwYadPSPD6NXM5uvLMoawI2XlWxd8XoV7248gstT3A/Nq+B/m1PId8j0TqL2OBwO9Hr/JWgfeeQRBg4cyD333ENycnIZr6wfPk/ewRdHd9EtojFdIuLoYGvEm33HEhJQ+h+JQohLj7SAilrh9no5mFFAkftMYunyKNLzi2+Ra1rx7fmcIjdNQgNxeRT6U38uje8ahznAwAMDW3IkqxC7003bKCt6XclJ7YvcHrIL/ZNNl0eRUeDEoxRmox6jXv4OEzWrefPm7Ny5k/j4eL/yhx9+GKUUDz30UC1FVjV+STsMgF7TERVoBWBH9nGuMreuxaiEEHWJJKCiVlgCDIRbjBzPO9Mn02oyEBEUgEGnYXd60DQNr1J0jQtl5jWtOZCRT+vIIGLPutXeNOz8tyotAQZaR1rYn17gKws26Vmw/gh7T+ZjNuoY1yVObsmLGnX11Vfz7bffMnLkyBLbHnnkEZRSfPLJJ7UQWdUIMQYCOSilfKudFZcJIUSxSs0DCrB27Vo2bdpEZmYmXq//9Df1fZS8zANaMz7ZdoyXfjxIrsONNcBAk9BAIoKMpNmdHM4sJM/hpoktkI9u606U1VTp46TbHby9/gi70+w0DzOj12BXWr5fnXmjO9A4tH73uxOiJpTn+3FzxlHu+fUTThbZMer09I+O57/9xqHXyd0GIUSxSrWA/uc//2HhwoUkJCQQGRkp63mLSvlDQix2h4cfD2RgCdBzfUIMl8WF8O6Go0RZ82kVYWFSj6YXlXxC8Tr0jww6c+vvgU8TcXm85DncWIx6Ao169p3MlwRU1Bn79+/n7rvv5vvvv6/tUCpla2YqTYNCCTIEEKDTY3c52Jtzkg5hjWo7NCFEHVGpBHTZsmU89dRTDB8+vKrjEZcQvU5jUs+mTOrZ1K/8r1e3AmB/ej6r9p0kzGzkylYRWAKqpseIQaex8Wg2p9v+m9nMxIfLKkui7vB6vdjt9toOo9K2Z6Vi0hsJMXrZlJGC3eUgpSCHIY3b8mCngYQGyB97QlzqKv2L3rFjx6qMQwg/G5Kzefmng5zuH/L9/gzmXtv+ogcMKaXIKXRhNurJd7hxehX5LjdWk3SHFqKqNAmysSvrBN+l7sOpirtobcw4is1k5sODW7infZ9ajlAIUdsq9Wt+88038/HHH1d1LEL4fLHzOGd3Tk7JKeL3ozkXvV+3V1Hg8tI+yopep8Oo08jIdzFr+W6Kzlr+UwhReTfFd2VbVqov+QRwej38npHC/rz0WoxMCFFXVKrZZ/fu3axbt46ff/6Zli1bYjD47+bZZ5+tkuDEpev0vJ1nc1bBWu9GvY7OscF8lngcr1LoNI0AvY6cQhc/HcxgaDsZDS+qX1FR0Xm3OxwVX7HL4XAwbdo0vvvuO9LT02nWrBmzZs1i4sSJACQmJnLXXXexbds2WrRowfz58xk0aFCl4r+QWEsIblXyD7ocZyGtgyOr5ZhCiPqlUgmoxWLh6quvrupYhPAZ1DqShRuO+J6HBhro0dRWJfu+r28LfjucyUm7E0uAnpYRFhRwOLPAb9oYIarLgAEDzvs5q8zn0O12ExcXx3fffUd8fDy//PILI0eOJD4+nh49ejB69GimTp3Kjz/+yKeffsr111/Pvn37iI6unj+6os3BpBXl+93JsAWYuall1zJfI//+hLh0VHoapoZMpmGqG345lMm65CzCzEZGdmhEdPDFjYY/208HMnj9tyQA0vMdHMwooEOjYNpGBfHnAS0vOL+oEBdj06ZN5arXvXv3izrOtddeyzXXXMNll13GrbfeSmpqKrpTUyH169ePiRMnMm3atArts7zfj2/u/o371y3DeWqaPqOm8Ua/cdzR5ooSdT1eL+8d2MQPxw8QoNMzplknRjWVcQZCNGQXNfLi6NGjHDp0CID4+HiaNGlSJUEJAdAvPpx+8eGVem1ukYsfD2RQ6PLSLz6sxBRLA1tFYHe6+Wb3Sban5hIfXrx978l8Xvv1EOO6xNEo2CRTM4lqcbGJZXnk5+ezceNGHnjgARITE+ncubMv+QTo2rUriYmJ593HvHnzmDdvnl9ZmzZtynX8KLOVtiHRHCvIRaGItYQQYSp9tolvU/aw4theoLiv6KKDm2lhDSMhrOTyukKIhqFSCWheXh5PPfUUP/zwA0ajESi+/XPllVcye/ZsrFZrlQYpREXYHW7+9vVuMgtcAHy58wSPDWlD2yj/z+W1HRrRPMzC8bwidh63Y3cW4PJ4+flQJgczCtDrdAxpE0GXxqFYjHraRVvl9qCoF7xeL5MnT6Znz54MHTqU9evXY7PZ/OrYbDaSkpLOu58ZM2YwY8YMv7IhQ4Zc8PhKKb49ugcvihCjCZfXQ77LyYJ9GxgU2war0f9uxvas1BL72J51XBJQIRqwSo2Cf/755zly5AjvvPMOv/76K7/++isLFizgyJEjJf5aFqKm/Xwo05d8QvHI9+W70kqt2zzMzEm7E7uzeL34fKcHpRQn8pwUujw88/0Bnlq5l6dW7mPud/twV8FAKCGqk1KKe++9l2PHjvHRRx+haRpWq5WcHP9ZJHJycggODq6WGD5L3sH27FRSC3JJLcwj21mEVyny3U4+PLSlRP0mQbaSZZbQaolNCFE3VCoB/fnnn5k1axadOnXylSUkJPDoo4/y008/VVlwQlSG010ySXSUUgbF6893axyKQaf5lRW4PBzNKcLp8VLkKn7tjuN21iVnV0vMQlQFpRTTpk1jy5YtLF++3Hc3KiEhge3bt/stm7xlyxYSEhKqJY6lSdtJyc+lyOPC4XWT73GS4yxEp+nYk1Pyj8HRTTvSPCjM9/zy8Mb0jW5RLbEJIeqGSt2Cd7vdBAYGligPDAzE45G5FEXt6tMijE8Tj1N0VtI5qE3ZU79c3zmGY7lFuDxekrIKSc93YjMbOZ5bhF7TCAk0UOTyYNDrSLNXfHocIWrK9OnTWbt2Ld999x0hISG+8quuugqz2czzzz/Pgw8+yOeff8727dtZunRptcRxJD+bfLeDIo/bV5brdrD55BE62UouxxkSEMjc7iM4mJdBgM5AU6utWuISQtQdlWoB7dGjBy+++CIZGRm+svT0dF5++WV69OhRZcEJURkGnUanGCv5TjcWo44HB8afdwqn/vHh3NA5ljBzAJ1jQxjSNoqmtkA6NLLSMsJC4vE8fk/JYeORbLLOurUvRF2SlJTEa6+9xs6dO2natClWqxWr1crcuXMxGo18/vnnLFu2DJvNxt///neWLl1abVMwtQ6OJM/l4NwpVuweJzpK70etaRqtQiIl+RTiElGpFtCHH36Yhx56iFGjRhEbW9xJPDU1lfj4eGbPnl2lAQpREV6v4ulV+0jNdRAUYKDA5WVXmp2ezcLKfI2maYzvGsf4rnEl9nXHh1tIyirEbNTT1BbIqn3pDG0XRRObjI4XVWPdunWsWrWK48eP43L5/4Hz+uuvl3s/zZs353yz6nXu3Jl169ZVOs6KGBd/GYsOlJxqSo8OqzGgRmIQQtRtlWoBjY2NZdGiRbz00kuMHz+e8ePH89JLL/HBBx8QExNT4f1lZ2czfvx4goODiYuL4+WXXy6z7uLFi+nYsSPBwcG0adOGt99+27ft8OHDvg73px/33ntvZU5R1FN7T9pJzfW/Tf7D/owyap+fTld8+71HUxvdGocSGVQ8cvdgRsFFxykEwKeffspDDz2E2+1m06ZNREdH43A42LVrF23btq3t8CptaON2tAkt2e3FqNMzskmHWohICFHXVHoeUE3T6N27N717977oIKZPn47D4SAlJYWkpCSuueYa2rVrx4gRI/zqJScnc+utt/LJJ58watQo1q1bx+DBg7n88svp1q2br156enqpfVRFwxdo1JcoM5dSVl5to6z8lpTlV9Y6MqjM+gfS89mXnk/ryKDz1hMC4IMPPuDRRx9l1KhRfPfdd9x99900adKEV155BbfbfeEd1GE3Nr+MnVkncJ11Iz7abKXAI91YhBCVbAFdvHgxq1ev9j2fM2cOffr0YcKECRw+fLhC+8rPz2fJkiXMmTOHkJAQOnfuzNSpU1mwYEGJukeOHMFmszF69GhfAtyhQ4cLTqYsLh0twi10jQvxK/tDwvlb5bMLXaxNyuJodmGJbbd0b0zryOLJswMNOib1aEJcaOl/3Czbnsrj3+zh/zYe5e/f7OGTbccqeRbiUnHs2DHfpPQBAQEUFBS3rl9//fUsX768NkO7aBqaXx9QDchzOXh049e8smMNxwtyays0IUQdUKkEdNGiRYSGFs/RtnHjRlasWMFTTz1F69at+ec//1mhfe3duxev1+s3HUhZK3T06tWLdu3asWzZMrxeLz///DOHDh1i4MCBfvVat25NXFwcN910E0eOHCmxn7PNmzeP6Ohov8fpHwFRPz10ZUv+2Lc5f0iIYfbQtgxpF8WhjAL+uzaJ1389zP70fF/djUey+dOyRP615hAPf7mLjzan+O0r3BLAU8Pb89qNnXl97GUMa1/6oI0il4fPEo/7lX2+4wQFzvrdiiWqV0REhG9+ztjYWHbu3AkU96k/e8qk+mhN2iE07cxPjAJOFOVxssjOuvRk5m77Hk89P0chROVVKgE9efIkcXHFAzZ+/PFHBg8ezJAhQ5g6dWqFWyPtdrsvmT3NZrORl5dXoq7BYGDSpEncfvvtBAQEcPXVV/Pcc8/RvHlzACIjI9mwYQOHDx9m27ZtBAUFMXr06PNODTVjxgzS0tL8HhZL6cvFifrBoNfRv2UE47vG0S7aypGsQp5YsYfV+zP46WAmT67Y40tC39t0FLf3TDvN5ztOkJHvLLFPm9lIgKHsfy5Fbi9Oj/8AEJdHUeCUaclE2Xr06MGaNWsAGD16NC+88AJ/+tOfePTRR7nyyitrObqL4/R48Cj/z79S+FpF0x357MtNr/nAhBB1QqX6gFqtVtLS0oiJieHXX3/lnnvuAYr7hVZ0HlCr1Upurv+tmLJW6FixYgUzZsxgxYoV9OrVi127djFq1ChiYmIYOXIkVqvVNw1UZGQkr732GsHBwezbt4/27dtX5lRFA/DjwQxcZyWHHi98vy+dluEW0u3+yaYC0vOdRARVbKSuzWykfbSV3Wl2X1nbqCAirabzvEpc6v72t7/5WjrHjx9PaGgo27Zto2/fvtx44421HN3FaR0SwS9ph/CqM62cATo9Jv2Zn53QAOmrL8SlqlIJ6KBBg5g1axZNmzYlKyuLvn37AsW30xs3blyhfbVt2xZN09ixY4dvZaWyVujYtm0b/fr1o0+fPgB06tSJa6+9luXLlzNy5MgS9TVNQ9O0805NIho+fSnrtxt0GjqdRtfGIWxOOfMHkM1soFVE5VrA/zwwno+2HGN/ej6tIoKYcM60TkKcKz09nUaNzkzMPmzYMIYNG1a8HOyJE5WaVaSuuCw8jiCjiWznmb7Vbq8Hi94IwJWNWhJrCSnr5UKIBq5St+D/8pe/MGHCBFq2bMmrr77qW+7t5MmTjBs3rkL7CgoKYuzYscyaNYu8vDwSExN56623mDJlSom6V1xxBb/++isbNmwAYM+ePXz99dd06dIFKJ5Pb9euXXi9XrKzs5k+fTqtW7eu19OZiIs3qE0kZuOZj7rJoGNI2ygA7u3Tgr4twggJNNCxkZWHr26NQV+pfxaEBBqZ2rs5z43qyN19mhNqNlZJ/KLhGjNmDFlZWSXKc3JyGDNmTC1EVHWynQUUuf1HvCsgyBDA3y4bxN3tLn4GFSFE/VXhFlC3280///lPbr31Vl8/0NMmTpxYqSBeffVVpk6dSmxsLMHBwcycOdM3BZPVamX58uUMGDCAgQMHMnfuXG655RZSU1MJCwvj1ltv5c477wTg4MGDzJo1ixMnThAcHEz//v358ssv0esrPw2PqP8aBZt45toOrN6fjlfBVa0jiA0pvvUXHGhgev/4Wo5QXKrKujtTVFREQED9nbDd5fXw1dHduNU5g4w0SM7PIiEstnYCE0LUGRVOQA0GA1999RW33HJLlQVhs9lYsmRJqdvsdrvf83vvvbfMyeVvvvlmbr755iqLSzQc0cEmJnSrWPcQIarLK6+8AhR3E/rvf//rN2+x1+tl+/bt9frOzZJDW8lxFJUo16Fh1sudASFEJfuADhw4kDVr1nDTTTdVdTxCCNHgnZ5uSSnFnj17MBrPJGVGo5FWrVpx22231VZ4F+33jBTiLCHsz/Mf5W41mLipZbcyXiWEuJRUKgFt2bIlb775Jlu3bqVDhw6Yzf7rYle0H6gQQlxK3njjDQCefPJJ/vKXv/j60TcU0WYrqYW5KPy7GFgMRvpEFa9Zr5UyOFAIcenQVCWGiJ+vc7ymaXz22WcXFVRtGzJkCCtXrqztMIQQos4pz/fj4bxM+n71b04U2vGelYRqQI+IptzfsT+3te5e4nW/pR1mWVIihR4XAxu1ZGyLyyRRFaKBqlQL6Oeff17VcQghxCVh5syZ5a777LPPVmMk1adFcDi9o5qxLHmHX7kCfs84wgcHf+fmll0x6IoHiBa5XaQW5jJ/1y++dHVZciLBRhPDm8gczkI0RJVKQIUQQlTOuV2WGqpuEU34LHkH5y626QG2ZqSAgv256byx5zdSCnIpcDnxorAazyzesDH9qCSgQjRQlU5ADx8+zPfff8/x48dxufzneps9e/ZFByaEEA3RpfL9GBkYRKgxkCxXydHwOc4iXF43/9r5M+mO4mVx7W4nR/OzuTyiMZy67R5tblh9Y4UQZ1Rqxu0ff/yRm2++mXXr1vH555+TmprK+vXrWb16NYWFhRfegRBCCJ+dO3eyYsUK3/en3W4v8Yd9fRMWYKbAU/o5uJWXcT+8xzdHd5OSnwMUJ6wBej1Or8f3+j8061Rj8QohalalWkDfeOMNpk2bxq233srAgQN57LHHiImJ4amnnqJFixZVHKIQtWdPmp3/bU4hze7g8sah3Nq9CYHG6l3YwONVbDiSTXq+k+5NQn2T5ouGJz09nYceeog9e/aglGLp0qU0adKE+fPno9PpePjhh2s7xEr7LGkHjlPJ5Lk8SrEt6zh5riIcHjdGnZ5os5V+0S34Y/t+OL0eOofFEKCXXmJCNFSVagFNTk7m6quvBornrCsqKsJgMDBp0qQyJ5QXor4pcLp57vv97D2ZT3ahm+/3Z/B/G49W6zGVUjzz3T7+teYQi35P4eEvd7L5aE61HlPUnnnz5hEbG8vq1av9JqO/5pprWLt2bS1GdvHWpyeXuU2naWhAoN6IS3nIdBZg0RuZ2rY3ncNj6R7ZRJJPIRq4SiWgISEhvltFUVFRHDx4EIDCwkIKCgqqLjohatGO43aK3P5DKDYdza7mY+ax88SZ1b88Xvhke2q1HlPUnk2bNnHvvfdisVj8yhs3bsyJEydqKaqqEXieFY9Ozw9q0hloFRzJLS278WqfG7g8sklNhSeEqGWVSkC7du3K+vXrgeK/1F988UWeeeYZZs2axRVXXFGlAQpRWyKDSq7FHW01lVKz6uQ53OUqEw2D2136e5uWllYiKS2P+fPn06NHD0wmU4mV6lq0aIHZbMZqtWK1WunUqXr7Vw5tXPZSonpNh6K4JbR9aDRT2lyBSVo8hbikVCoBffjhhxk0aBAAd955JzfffDNpaWlcddVVPP7441UaoBC1JT7CwlWtInzPTQYdN1fzevJdG4diNfn3MR0QH16txxS1p3fv3nz00Ue+55qmUVhYyH//+1/69u1b4f3FxcXx2GOPMXXq1FK3L1u2DLvdjt1uZ8eOHaXWqSpjmnXEWMZPTKQpiIgAC21DoxjauC22gEtjaiohxBmV+pPTZrP5/l+n0zF58uQqCkeIuuXuPs0Z0jaKE3YHCTHBWE3V20pjNup5fHBbliWmkp7vpGdTGyM7NKrWY4ra8+CDD3L//fdz00034XA4mD17NsnJyVitVp588skK7++GG24AYMuWLaSnp1+gdvXKdBSW2cKR73bS1GrDajSxNGk7Rwuy8SpFdGAwo5p2IMxU8dZfIUT9Uulf0/T0dJYvX87Ro0e57777sNlsbN68maioKJo0kX48ouGIj7AQH1FzP4hNw8z8aUDLGjueqD2NGjVi0aJFrFixgn379lFYWMjIkSMZMWKE36CkqjJp0iS8Xi+dOnVizpw59OvX77z1582bx7x58/zK2rRpU65jfZe6D0eJaeiL5bocbMk8RlRgEGa9kT05J2lviwbg94yjvNBzNHpdpW7QCSHqiUoloImJiUybNo2WLVuya9cubrvtNmw2G5s2beLAgQM888wzVR2nEEI0SAaDgWuvvbbaj/P+++/TvXvx+uvvvPMOI0aMYPv27TRv3rzM18yYMYMZM2b4lQ0ZMqRcx0uyZ5W5TaFwedxkFBXgRdHEEurbdqLITmL2cbqEx5XrOEKI+qlSCejLL7/MpEmTmDJlCgMHDvSV9+7dm2XLllVZcEII0ZCdXsAjNbV4poO4uDgGDRpEjx49qvxY/fv39/3/fffdx4cffsjy5cu59957q/xYAEZNhw7KaAOFbGch0WYDRk1HnCXE/7XS+ilEg1epf+V79+5l2LBhJcrDw8PJyir7r14hhBDFnn76aaZNm8bWrVuxWCyYzWY2b97MH//4R+bMmVPtx9fpdCilqm3/HWyNMGhl/8S4vF6CDSae6DbMb87P1sERBOoM/JZ2mBynrKwnRENVqRbQoKAgMjMzadzYf0Tw3r17iYqKqpLAhBCioVq1ahUrVqzg1VdfLTF13dq1a3n44Yfp3bs311xzTYX263a7fQ+v10tRURF6vZ7U1FSSkpLo1asXAO+++y4bNmzgrbfeqrJzOlff6Ba8Y9rI8aK8Urd7ULiVlz7Rzeke2ZjNGcdoZLZyKC+TxzZ/CxS3oj7YaSBdI6p39gkhRM2rVAvo0KFD+fe//01WVhaapgHF/UJfeeUVRowYUaUBCiFEQ/PFF18wefLkUudN7t27N7fffjtffPFFhff79NNPYzabmTNnDkuWLMFsNjN16lTsdjv3338/4eHhxMTE8O677/Lll1/SqlWrqjidUg1t3I7owKAyt+uAHFcR36fuo6MthltaXU5HWyNWHz/gq+NSXhYd3FxtMQohak+lWkCnTZvG888/z4gRI/B4PIwfPx63283gwYO56667qjpGIYRoUPbs2cN9991X5vb+/fvz8ccfV3i/TzzxBE888USp27Zs2VLh/V0Mk95A0yAb27KPl1nH7nKQ6Tizet7JovwSdU4W2UuUCSHqv0oloAEBATz22GPcdddd7N+/n8LCQtq2bXve0ZRCCCGK5eTkEBkZWeb2iIgIcnNzazCi6nH+kfAQoNMTpD+z4lj70GishgDsbqevrEdk0+oMUQhRSyp0C97r9fLOO+8wZcoUbr/9dpYsWUKPHj0YMmTIRSWf2dnZjB8/nuDgYOLi4nj55ZfLrLt48WI6duxIcHAwbdq04e233/bb/vHHH9OqVSssFgvXXHMNSUlJlY5LCCGqg8fjQa/Xl7ldp9OVuUxnfeL0ln0OCggJCPTN/wnFraaPdL6aDqHR2AICubJRS+5o0/OiYsh1FvHLiUPszk67qP0IIapWhVpAFyxYwMKFC7n22msxmUx8+umnZGZmMnv27IsKYvr06TgcDlJSUkhKSuKaa66hXbt2JfqTJicnc+utt/LJJ58watQo1q1bx+DBg7n88svp1q0bu3btYvLkySxdupQBAwbwt7/9jfHjx7Nu3bqLik8IIaqSUop//OMfBAQElLrd6XSWWl7fRJiCIC+jzO0p+bk8uWUlKQW53NLqcow6Pa1CInm8a/nmGr2QPTlpPLvtexxeDwA9I5vyYKeBF3iVEKImVCgB/eqrr3j88ccZPnw4AMOHD2fq1Kk8/vjj6Co5b1t+fj5Llixh06ZNhISE0LlzZ6ZOncqCBQtKJKBHjhzBZrMxevRooLizfocOHUhMTKRbt268//77DB8+nKFDhwLw1FNPERUVxY4dO+jUqVOl4hNCiKo2atSo8243m82MHDmyhqKpPmaD8bzbXcrD2rQk9uacJMNRwF8SrqzS4390aIsv+QTYkH6EXdkn6GCT5W2FqG0VSkCPHz/O5Zdf7nuekJCATqfj5MmTNGpUuX/Qe/fuxev1kpCQ4Cvr2rUrS5cuLVG3V69etGvXjmXLlnHdddfx66+/cujQId9k+ImJifTseeZ2TXBwMK1atSIxMbHMBPRilpoTQojKuNi7RvVFocd1wToeFOmOfJ7Z+h0trGGMadYJo67s7gkVkVFUUKIsvZSBTkKImlehZkuPx4PR6P8XrV6vv6i+Sna7ndDQUL8ym81GXl7JueMMBgOTJk3i9ttvJyAggKuvvprnnnvO1//Ubrdjs9nKta/TZsyYQVpamt/DYqm5db+FEKKhigwo33epAjKdBfxl/RdMWP1elU2Q3zPKfwCTSacv9xKfHq8Xp6f+98MVoq6qUAtoaf2WHA4Hzz//PGaz2Vf27LPPlnufVqu1xGjPnJwcgoODS9RdsWIFM2bMYMWKFfTq1Ytdu3YxatQoYmJiGDlyJFarlZycnHLtSwghRPWyGEzlrqtRPKf0xoyjPP77N4SbLHQNj+Pq2Na++abPZ09OGpszUogxB9OvUTxGnZ4J8V05nJfFd6n7iAoM4qGOAwkJCCzx2v256WQ48ukcFovFEMAnh7fx1dFduLwe+kW34M62vaqkVfaL5B2sOraPAL2e0U07MjCm+uZhFaKuq1ACWlq/pWuvvfaiAmjbti2apvn109yyZYvfLfnTtm3bRr9+/ejTpw8AnTp14tprr2X58uWMHDmShIQEv7nu7HY7Bw4cKHVfQgghqle6oyK3uxVZjgI8KL5L3Ud8cAQbM46S7shnfHzX875ydep+/rv3zGDTNScO8XjXIaw5fpCdOSeIPbXW/AcHf6dLRJxfMvnKjjWsS08GwKI3cl2zTnyStN23/acTh4izhDKm2cWNI/jx+AH+d2iL7/nre9YSHRjsNwuAEJeSCiWg1dFvKSgoiLFjxzJr1izee+89kpKSeOutt1i4cGGJuldccQVz585lw4YN9OzZkz179vD111/zt7/9DYBbb72Vnj17smrVKvr378/s2bO57LLLZACSEELUAo3y30r3ArluB3pNR0zgmbtWq47tO28CqpTijd2/caQgh2CjiXCThV05aezNOcm3KXv86p4osrMlI4WeUc0A2Jl93Jd8AhR4XLx/4Hf05wyq3Zl94qIT0I3pR0qWZRyRBFRcsio3dL2KvfrqqxiNRmJjYxkyZAgzZ870jYC3Wq2sWbMGgIEDBzJ37lxuueUWgoODGTJkCDfffDN33nknAB06dGDhwoXcfffdhIeHs3nzZhYvXlxr5yWEEJeyEKP5wpXO4VFekvKzUEpxOC+TdSeTeWrLCnZlnyi1/oJ969mUcZSU/Bx2Z6f5Jr8vKscAqNJWXipNc2tY+U+gDFGB1hJl0aWUCXGpqNRKSFXNZrOxZMmSUrfZ7f7LsN17773ce++9Ze5r3LhxjBs3rkrjE0IIUXHdIuJYeiSxXHVP9/LUAUfzcyj0uMkoyqeZNYxd2Wk8t301L19xHTbTmaQ211nE96n7iTYHk5Jf3P//WEEu3cLjaBcSxfAm7f1uzTcKtNI1orHv+WVhsRg1HS7l9ZWNbNqBPJeDX9IOA9AmJJLRTTtW7gKcZVTTjmxKP8rJU90S4q3hDGzU8qL3K0R9VScSUCGEEA3P5ZFNMaLhKseteAXo0AgxmggPtHC8IBezIYD0onz0mo5WIRFszTrGlWcN3HF43SigeZCNAJ2eYwU5ZDkK+S0tiZt+fJ9hjdvxYKeBbM5IIcJkYUhcW7/+n2EmC39NuIrFh7eS4cjnishm3NyyG0adnvHxXXB43DQJslXJtQg3WXjhitFsy0zFqNOREBaDTqsTNyGFqBWSgAohhKgWXSPiaB/aiO05x8tVP1Cnp7k1jByXg3CThXy3kxyXizy3g8jAICJM/tM6RQVa6RjaiJ05JzDpDeS5nLiVlwxHAemOfHJdDiJNQdzdrneZx+wcHkvn8NgS5aXdMr9YRp2e7pFNqny/QtRH8ueXEEKIahFnCaVXdLNy1/cCxwry0ACdppHncuD0uClyu0gtyKWlNaLEax5KGMjoph3RgGCjiWCDiXy3k3RHPtszU3lu+2r25ZyssnMSQlQNSUCFEEJUm8YVaEks8rpJd+RjdzlIKyru/x+oNxBnCaFtSCRv7V3HO/s2sOHkmZHrFkMAN7fsxrVNOtDcasOLosDtBFU8t6hRp+f/Dmyq8vMSQlwcuQUvRAOUlFnAe5uOciS7kISYECb1bEJI4PnX5RaiOjQJDq9QfS8Kl9dDoM6AAw9GnYGuEY3Zk3uSbGchtgAzK47t5YbmCYxt0cX3umubtGdD+hGO5ueQ6ShAr+kICzTT2BJKSkHOeY4ohKgNkoAK0cB4vIrnVx8gq7B4GprfkrIocHl4ZFDrWo5MXIpGVWIEeY6riGBjIHpNw+31cCgvkyxHAQVuFw7PSazGAJRS3NC8s28gT6uQSP7QrBN2VxGFbhdNg2w0sdoA6BJWvuU3hRA1R27BC9HAHMos8CWfp207lovb4y3jFUJUnwO56QRWcLS3F8h3ObC7nRh0OuwuByeL8ilwO1Eo7C4nO7NPcPaS8T+m7mfBvg3oNB3tbY2Kb+U7HfSIaMIdbXpWOG6v8rIp/ShfH9nFMWlBFaLKSQuoEA1MhMWITgPvWT/O4RYjet2F19MWoqoF6A14KvgaDXCjMChFpqOAfLcTh8dNamEeATo9ATo9EYEWvCj0wMG8DGZu+prUgjzQINQYiNPrIctZSIajAIfXXeG4X0z8kc2Zx8h0FGB3ObirzRVMatOzXOvSCyEuTFpAhWhgwiwBXJcQ43uu18Gt3ZvID6eoFSa9AV0FP3vaqWnp3cqLWymKPG48KDzKi9PrLm4FdRbx8eGt7M4+wbv7N+Lyeovrez0cyMvA4XFj0Ok4ZM/k7b3rK3T8/bnpbM48xuG8THZnp3E0P4fnEn9g4b4NFdqPEKJskoAK0QCN6xLHvNEd+NOAeP59fWd6Nb/4pQRF3Td//nx69OiByWTipptu8tuWmJhI7969sVgsdOzYke+//75GYlJKEWQIqNBrvGdNXK9QuM9aqcijFLkuBwfyMrn/t0+57ruFLD5YPJF8trOQTEcBbq8HvU7zzeW5N9d/GqZMRwG/nDjEEXt2qcfPdRbh8XpJLcz1lbm9HlYe20taQV6FzkUIUTq5BS9EA9U41Ezj0IqvxS3qr7i4OB577DFWrVpFenq6r9zlcjF69GimTp3Kjz/+yKeffsr111/Pvn37iI6OrtaYYiwhtAyJJDP9SJXu14Mix1VEvseJ1+tFr9MTbAzAq4r7iHayxWDQ6VBKYTOayXQUEG6ysDYtiVd3/4LnVAfSMU070jkslkxHAV3C4wgJCKRTWAzBRpOvj6k61fr6W1oS96/7lJFNO3Bbq+5yV0GIiyAJqBBCNBA33HADAFu2bPFLQH/44QcKCgqYOXMmOp2OCRMm8K9//YslS5Ywbdq0ao2pkTmYm+O7srGKE1AAj/Li8RS3l+q8Cq9TYTMVJ5A6NArcLpLsmRS4ndz+0yK6hjfmYG46Xk0r7hagFM9v/4H44HBMegMmnZ5HL7uGIEMAHW3RbEw/gt3txKTTY3c5CQs0o9fp+CZlD21CIukT3aLKz0mIS4UkoEII0cAlJibSuXNndLozva66du1KYmLieV83b9485s2b51fWpk2bCh//Tx3785cNX1T4dRdy+ka9DgjQ9Bh1OgJ0ekKNJlxeDxvSj6DXdOQ4CslyFvLdsf3oNIgxh9IlIpYij5tMRwGNg0Ix6Q04vB4W7ttAWpGdQo+LdrZojuXnYHc50evcBOoNOD1uAvQG9uScLFcCqpRi7ckk9uWm0yYkkt5RzaXlVAgkARVCiAbPbrdjs9n8ymw2G0lJSed93YwZM5gxY4Zf2ZAhQyp8fINOjx4Nz1l9O6uSArxeLw4UGYUFZBYlo9fpKHC78CqFQqGhoU4dPzk/C4/y0DI4ggC93q+PamJWKsEBgQDoNR0GnZ5cdxFepTheYCfLUUS3iMa0LOcE+3O2ruKH4wcIDTBj0hvYmX2CO9v2qvJrIER9I4OQhBCigbNareTk+M9lmZOTQ3BwcI0cXylFkNFUpfvU+cbKF4+ad+PFqOloZAlGAQXu4rlw1am003vqv+pUWVpRPkGGALqGxfmN0m8bGuV3nNTCXEKNZmym4v7UDq+bZkE2+kXHXzDGedtX8/a+9ezPzeD3jBSyHIWsTj2A3eW4+AsgRD0nCagQQjRwCQkJbN++Ha/3zGjyLVu2kJCQUCPH1zSN7hGNq3ifYNYbMfgmuddwKy9WQ/GSs97ztLZqCiJMFsJNFp7oNowOodE0CrQyumlHZna+mmDDmWRZj0acJYSOtkZcHtmY7hFNuLllN/RndWfYnZ3GP7as5KH1n/PRwS14vF7256az/uSRMwOZlCI5PwsvCrdXFoUQQhJQIYRoINxuN0VFRbjdbrxeL0VFRbhcLq666irMZjPPP/88DoeDJUuWsH37dsaNG1djsb3db/xF7+PsnpMeVTw9k+fUFE06TcOrILUwD6NOj1LFI9fVqded/Vq9Tkd0oBVN0+hgi+bxrkN4qdd13NyyG1HmYG5q2RWP10u+y8n1zTsTcuqWfKDeSKvgCC4Li/XtK9tRyLPbv2dXThrHC/P47MgOliZtJ9NRgFGvJ9xk8dV1ej10j2jsa00V4lImCagQQjQQTz/9NGazmTlz5rBkyRLMZjNTp07FaDTy+eefs2zZMmw2G3//+99ZunRptU/BdLb4kIiL+sHR8E8kdWi4vadGwAMmTY9ZbyTf7SLYaCLWHIxJZ8CgFd+stwUEEqgzoNd0GHV6UgtzOZCbzopje/2Ok3Rq4nq9TkeQMYCD9kxGNG7HkLg23Bzfldldh/i1fm7OTMHp9V/raX16MglhMVj0RtqGRNLMaiPMZGZE43ZM79D/Iq6CEA2HDEISQogG4oknnuCJJ54odVvnzp1Zt25dzQZ0loyifMJNFtIdBZV6vYaGUdOh1+lQKEw6A27lxeFxE2Qwomk6Ct1OUODxesl1O05NtQRoUOB209gSQlqRHZfXS4HbRXJ+Ns9vW02fqOa+Vs51J5NL3L4/XpjHjM5XlxrX2S2cp4UFWLAYAvhbl2v46NAWYgtD6B7ZhAnxXTHq9JU6fyEaGklAhRBCVLs39qw9a9hQxSkUDuUBT3FrY6HHjV7TCNDpfaPdPSg04HhRHjpN8418N2o6LAYjOa5CPF6FXtNweT3kOovYlXOCz5MTmdCyG0adnhBjYIljhwaUfcv8srBYWljD2JWdhtlgxKTTM7bFZQC0DI7g0cuuqfQ5C9GQyS14IYQQ1W5H9nGsFzESvrQhRR6l8Hi9OJUXt2+0e3Hd031ElSou02kaRs2ABy9OrweX14Pd7STP5eB/h7YwY8OXZDsKGRjTkgK3kx9TD/D9sf0k5xXfgi+N0+Pm+e2rOZSXicPrJkgfwPM9RpUYSS+EKKlOJKDZ2dmMHz+e4OBg4uLiePnll0ut98EHH2C1Wn2PoKAgNE1j6dKlQPFqHzqdzq/O3Llza/BMhBBClKaJJZQOoVXf59Spzj+i3EtxkqoDTjrsxQmp8hYnpyjCTRaCjYGkFdn5NmUPO7KOsyv7BJ5TEzelFOayOvWA3z4L3S4O52Wy8tg+tmalomkatgAz+R4n69KTq/wchWiI6sQt+OnTp+NwOEhJSSEpKYlrrrmGdu3aMWLECL96t9xyC7fccovv+fLly7npppsYPny4ryw6Oprjx4/XWOxCCCEu7I42V/Dijh+x6ozYva4aPbZCUeBxoxSnEstiRp2OzmGxvnlA0x35/J5xlAK3iwCdAR0aSsGnyYnc3qYHAG/tWcdLO34i3+1Ep2m0DY0k3BTkO1aSPatGz02I+qrWW0Dz8/NZsmQJc+bMISQkhM6dOzN16lQWLFhwwdcuWLCACRMmYLGU7AQuhBCi7mhvi2Z+7+vpHd28xo55etS8XtMINgagaRoGTcN4asS8SWfwDT4CiLOE8G3KXtKL8skosmN3OwCFSV/cVpNWkMez278n3+0EoMDt5PeMFL9jdrI1IttRyIqUPfxy4hBOj7uGzlaI+qXWW0D37t2L1+v1mxC5a9euvtvqZcnIyODzzz/nhx9+KFEeExODyWRi+PDhPPPMM4SHl71kWlWtdSyEEOL8THoD17fozA8nDuK+wK3zytAAAxruU22cyld+KhVV4EahaQqlFGFGC8fzc8h0FpLvcfHt0d2nWkkVHq8Ct4sQo4k7214BwNr0JL9J5E06A/keJ8ZTk+FfE9eG5kFhPLThc4pOJZ5NLaE8dflwXxIrhChW6y2gdrud0NBQvzKbzUZeXt55X/fBBx/QqlUr+vTp4ytr3749W7Zs4dixY/z8888cPXqUSZMmnXc/M2bMIC0tze8hLapCCFE97mnXm3hr+dZRryi9ptEjqhmB+uL5P/VoxRPUo8goKvD16/Sq4gQzvcjOuvQj7MpJ44g9ixxnIXa3g0BNj9lgpJHZyk0tuhFnCcHuctAjoil63ZmR/Jqm0T40mrf7j+et/uO5vXUPvji605d8ZjsL+fLoLm747h3+s/tXCt0V63qQ5Sjgs6REPj68leMFuVV6rYSobbX+J5nVaiU31/8fVnnWKF64cCF33HGHX1lMTAwxMTEANG3alPnz59O6dWsKCgokqRRCiDpAr9PzSu/ruHbl21W+b49S6JWGSWfAq4qnZkIp0DR0GmhKOzVWvri11OX14qa4RfN0z1CP8uL2ODEpPTp0rEjdy9fHdhFqNPN4l8Hc07Y3L+9cQ5HHTbjJwpzLR2A4a27PXGfxOu8ur4fd2Wl4lSLP7WTNiUPo0LinfR/KI9NRwN82fU3uqXXjvzqyiye6DaV5NSXvQtS0Wm8Bbdu2LZqmsWPHDl/ZhdYo3rx5M4mJidx2223n3bdOp0Op4lstQggh6oarY1rTO7JJle9XAb9npZDncuD0evCgiqdlUpy65a/8V1PStNN35n2P0/txeD0czs9kT85JMh2F7M9L565fFvPD8QN0DW9Mr6hm9IhoQkpBjl8M/Ru1ACDXWYRXKTSteN15KF41qbx+PH7Al3xyKp5vju6p8DURoq6q9QQ0KCiIsWPHMmvWLPLy8khMTOStt95iypQpZb5m4cKFjBgxwtfaedrq1as5fPgwSimOHz/On/70J4YOHUpQUFAZexJCCFHTAg1G5nYfWS37LvC48PjaOYt5UXiU8ks0FQodoNNK/xk0aDqUApfXjd3lIMdZSFqRnS0Zx9idk4ZZb0Sn07EqdZ/vNbuz03B4PYxs0p52oVHYTIF0tDXCbDACEGM+/529szk9npJl3pJlQtRXtZ6AArz66qsYjUZiY2MZMmQIM2fO9E3BZLVaWbNmja+u0+lk0aJFpSaomzdvZsCAAQQFBdG9e3ciIyN57733auw8hBBClE+/RvFcFhZbK8cO0HQE6QMw6PSEGQOxBQQScM4SmW7lPTU9k4bd5aDI48btLZ4/1KsUJ4rsQPGa9AAfHPidp7au5N39G/nq6G5GNu3InzoM8K2iZNEbmdjy8nLH2L9RvG9wExS32l4d2+riTlyIOkRTcn+6hCFDhrBy5craDkMIIeqcqvx+/C5lH3/4biF2j7NK9leW07fc1Vn/b9TpMGkGLIYAcpyFFClPidfYAswUely+ke9WQwBupTAbDLQKjiA+OIJxLS7j6pjW/PG3pRwvzCXX5SDcZKGpJZQ3+43jWEEuaUV2v5bQ8jqQm86KY3txejwMim1N5/DaSdjP5fF6OWzPJCrQ6jeNlRAVUeuDkIQQQlyarmnchr1jZxL30VPVdgyN4tvpp9eKP93i4vR6ceKkyOvGc047jEZxy2ZMoJVCr5tAnZ4cp4N8txOP14PHqwMFN8V3YUyzBI7mZ/N7xlGS7dl4lBeDTke70ChcXg9NrTYama2kFuYSaw4hoALTMbUKieS+kMgquxZV4VBeBvMSfyDbWYRe0xjfogujm3Wq7bBEPVQnbsELIYS4NMVaQnihGvqDnh5spACX8uI5q1eodlY9l/KetTbSGQGajmbWMEKMgRR6PEQEWjDq9QQZA+gR2ZT4kAgO5RWveqRDI8mehVO58eDF4XVzIC8DvU7H5owUpq9dxqObljNt7TJ+Tz9a5edak97Zv5FsZxFQPOvAh4e2cPJUdwQhKkISUCFEveJwe1hzMIPv9p4kr0hWmWkI/nLZ1Vxuq9rby2ePaj+bXiueG1QrZRsUJ5M6NKLNwYSbLLSwhmHU68hyFmI1mmgdEklEYPHA1ixnAQCJp9aD12s6NIr/61Fe0vLzeH33b+zNOcme7DT25Zzk9T2/4T5rMNHu7DTe3beBz5ISsZ816r2uSj5nqVEFHLFn10oson6TW/BCiHqjwOnm8W/2kJpb/EP90dZjPDmsHbEh0g+tvvt62FRafPQ0RVT9Ckln09DKnJpPo3hqphbWMLqGN4ZTa8THmkNwetzYAsyEB56ZVaVXVDMAmgSFodM0PAoMOh06NKzGQJQGGzOOkFFUnKhmOArIcznIdBQQbQ5m3ckk/rXzZ1+i/MPxAzzbY2SdXjWpo60RmzOP+Z4bNR1tQutWNwFRP0gLqBCi3vjxQIYv+QSwOzx8ufNEpfaVbnfw2i+HmfnlLt7beIQil0xxU5saWULYev1fq/044SYLVqOJRoFWzHpjidHvKEVLazhNgkJJtmex9mQyv2ekcNCeSVpRPjnOQhpbQpjQogvDG7fH6XHz371riTFb0aHhUQqzwch97foQbDRhd/kPsLK7nRh1xT+9Xx3Z5ddKe6LIzu8ZdfsW/Z1te9ExtBFQPL/p/R37E2yUPwBFxdXdP7OEEOIcuaXcci+t7EKUUjz7/X6OnUpmk7MLySxw8cDAlhcdo6i8trZoDoydScePn8NR6g30i6MHOoRGExlo4VhBHkl5mWS5itDjBQ30mg6z3siO7BMMimnN1sxUijwudJqGUQsky1lA65AI7m7bmzahUQBsyjjKiSI73SObEmPOIcdVRIIthse6DKbQ46J9aBR7ck9S5HYTqDfQJjTSt3LSuYOfAL+15uuicJOFx7oOxulxY9Tp0bSyOjMIcX7SAiqEqDd6Nw9Dd87vXd8WYRXez+HMQl/yedqGI9m4PHX7x/9S0DI4kj3jHq2WfcdZQjlZlM/XR/ewIf0IJxx2lPJiOJV46tBweT1kOQt5LvEHHB5X8dKcXi9ZzsLiFZW8XtLOGnTjPZVEepUiMjCIBFsMLYMj0Ot0WI0mRjbtSLfwxvSMakq3iMaMatLR12I4JK6NX3xhAWZ6RDatlnOvagF6gySf4qJIC6gQot5oHm5hxtWt+GpnGg63l0FtIujTouJrYweb9L4R0qcFBegxnJvdilrR3BrO671vYNraZX6j1y9WgdvFsYJcvKemY9Kh4UZh0Rko8np8k8oHGYxknRrpDcWrJhV5XGQ6CtiZfYLjhXm+bd0jmpDpyGdPzkmUArPByE3xXXzb72p7Be1Do9ifm0HrkAj6N4r3bbsqtjVWo4n/7lnLwbxMzDoD36XuY1TTjlV2zkLUVZKACiHqlS5xoXSJC72ofURaTVzTJpJV+9J9ZeO6xDXoFp3JkyezaNEiAgICfGU7d+6kWbNmtRhV2e5s14vggEBmb/qG/fmZVbJPu9tRYnS8UgqTwYjXU9yKGWcJIcJkYWPG0VPLcWr+SbCm+PjwNlqHRNIlPI6UghysBlPxQCWvh8jAIJLzz6wPr9N0DIxpxcCY0lcx8igveW4nUebiOUcXHdxMdKCVK6Lq5vsiRFWRBFQIcUma0qsZvZqHkZxVQMdGwTQPt9R2SNXuoYce4tlnn63tMMrFoNMzsdXlTIjvyk0/vMfHSdsvep+OUtZS1zSNQL2BGHMwxwtzOVFoJ9NRCKp4jlCFQgMCdHrCTRaK3B5yXUVsyUihS3gch+1ZBOgNtAg+0xJ/yJ5JpqMAk05PkNF03pg2Z6SUWiYJqGjoJAEVQlyyOsUE0ykmuLbDEOeh1+n48KrbmLnpK17b8QsFqurmfvWiUEphMRgxG4w4vB4cXrevJbw4+dRQKFxeDxmOAqzGAFxeL4UeF/87uJlAnQGllO81bq+Hg3kZTF+7DL2mMaxxO25t1b3MGGLMIaWUle8z6fZ6WJGyl905abSwhnNtk/YEVnC5TyFqiwxCEkKIS8Sbb75JeHg4Xbp0YcGCBResP2/ePKKjo/0eBQUFNRCpP71Ox7yeo9l2/V/pE9m8yve/Py+DlIIcdGiEmyyYTo1SN2i6syat1/AqL3aXg0CdntWpB/jiyE6WJG0j3GQhyBBwqpaG8axR7l8f3c3Ws+bNPNeQuDY0DzozkK55UBhDGrctV9xv7V3H+wd/Z2PGUT5O2sY/d/xUqfMXojZoqqwZeS9hQ4YMYeXKlbUdhhBCVJnff/+dZs2aYbPZWLNmDePGjeONN97gxhtvrNB+avv70eX18Pru33h1188k2bNxKU+p0xlVhAZYDAG0sIZhdzkp8rhIdxRg1HRoWnHfUJ2m0dwaRiOTlUBjAOEm/y4bj3S+im+O7mFp0na8StHCGo5RX5yIjm1+GTe06Fzm8ZVS7Mopns+2Q2ijcvVFdnjc3PXL4hLn/tIVY2hUzhZUIWqT3IIXQohLwOWXX+77/6uvvppp06axZMmSCiegtc2o03N/x/7c37E/WzJS+N/BzXxzdDfbso9Xep8KCKB4GU2721G8kpHeiFFvwKO8aECL4HCsBhOH7Zm4vF5izMG0DolEr9OBUry66xfsbhd6nY4T+bk4vR46hcUA0Dok4rzH1zSNjraYCsWsUTyK/9xZAvSa3NgU9YMkoEIIcQnS6XRlLklZX3SNaEzXiMY813MUbo+HVSl7mfDDe+R6nBd+8Tmy3EVkZR/HeOq2e4fQRvyheScO5GVyMC+DfLeTzRkpGDQdmqZxIC8Do05Hy5BIjDo9608ewen1YAswYwsIJNtZhA4Y2bQjl4XHVfm5B+gNDG3cjq+O7vKV9YlqTuRZS4UKUZdJAiqEEJeAxYsXM2LECIKCgvj111+ZP38+//73v2s7rCpj0OsZ3qwDObfPpdnip0nJz8FbiTlEPSiMmp59eSfpGdEUh8fD8qO7yHYW4VFedJqGVR+AUWcgx+Xg6phWrDy2lxxXEUoVzzXaNCiUruFxzO99A9YAE0opfjh+gI3pR4gKtDK6aUciqiBRnNiyG61DItidnUZ8cDj9ouMv/CIh6ghJQIUQ4hIwf/587r77bjweD82aNePpp5/mpptuqu2wqsWhsX/jtV2/8NvJJDIdhTiVh80ZR8l1FnG+ta404HSjsFLw710/szkzlTyXA48qTme9SlHk9RBoMNI6OIIsRyFbM1NxeNx4vMXrwGc4Cniy2zCsAcVTMH2evIOPDm/1Hef3jBRevGK0b7BSZWmaRq+o5vSKqvqBWUJUN0lAhRDiEvDTT5fOCGm9Tsf9nQZwPwNwez3sz83AFhBIkj2Ld/au5/OjOznpKEAphVd5sRgCKHK78FA82Oj0PgL0Boq8Lrzgm44JigcN6TUdncNiWH18Pw6PmwCdAbdWPBH94LjWDDprmc3vU/f7xZfuyGdbZirdI5sAsO5kEt+m7AFgeOP2FZoDdG/OSfbkpNHcGlYtt/qFqC6SgAohhGiwDDo97W3RAMRYQugV3Zz/ABmF+RzIS+eH4wdYfHgrJwrt5DqLQNMIMhh5sOMAErNPYDOaKXK78WqgUxqaBmEmM2OadSLOEsohexahAYHkOIswaHoMOh2TW1/hF0OAvmRLp+lU2Y6s47yy82df+Z6ck8zqcs15ByWlF+Wz8thefktL4lBeBsEBxWvLD41ry+Q2PS/2kglRIyQBFUIIccmJMAcRYQ7iiujmPHzZIKC4ZTOtyI4twIxJb2DJoa3kuRxsSE8mvagAg15HZ1sMd7S9grva9uLz5B2nRrA3IttZiNPj4fbW3X0J72ljmnbiP3t+8z1vFRxBp1MJ5i9ph33lbq+HYwW5zP59BZPb9GB44/bFo+zPkuMs5LHfl5PtLGL9yWS8StHRFo3NZGHlsb38oVkCNpO5mq6aEFVHElAhhBCC4j6VZ8+hOS6+C53CGrEvJ50Ik4XLwuMwG4y+vpuD49rwa9phjhTkEGay0NgSUup8nwNiWtLIHMzG9CNEm60MaNTSN9dnyFlLde7OSSPXWbxe/QcHN5NSkMvd7Xr77evXtMPkuhx4lfINskotzMNmsqCAfLdTElBRL0gCKoQQQpShoy2mzNvhQUYTc7tfy45Tc5B2ssWUaLE8rW1oFG1Do0qUD23cjjUnDpFSkEOu00GAXu9binPNiYPc0aan32Cl04OkDDodYQFmshyFvm3x1nAaB4VW6jzLq8jtIrUwl8aWUAL0kkKIyqsTM9ZmZ2czfvx4goODiYuL4+WXXy613gcffIDVavU9goKC0DSNpUuX+urMnz+fxo0bY7VaufHGG8nKyqqhsxBCCHGp0et0XBYex2XhcWUmn+cTbrIwr+cobm/VnVYhEXQNb4zpVGJn0HScuyZS3+gWWE8t+9kmJIo4Swg9IptyTWxrZiRcdZFnc35r05L449qlzPr9G6avXca28ywxKsSF1IkEdPr06TgcDlJSUvj222+ZO3cuy5cvL1HvlltuwW63+x4ff/wxISEhDB8+HICVK1cye/ZsvvjiC1JTU9Hr9dx77701fTpCCCFEuVkMAdzQ4jLGNO2E4awkdkST9hjOmarJZjIzp/sIRjbpwJC4NizoP4H/9hvHnW17Veutd4fHzVt711HkcQNgdzv57951eNX5JrYSomy13n6en5/PkiVL2LRpEyEhIXTu3JmpU6eyYMECRowYcd7XLliwgAkTJmCxFK/J+84773DHHXf4lpybM2cOHTt2JCcnh9DQ6r0tIYSoX9LyHGxLzSUm2ESnmOByrb8tRHWa3qEf3dLiSLJn09HWyDdN07miAq3c0uryUrdVl5NFdgo8Lr+yDEcBdpeTkFOj8EXDUeR2sTHjKBrQPaIJgQZjlR+j1hPQvXv34vV6SUhI8JV17drV77Z6aTIyMvj888/54YcffGWJiYl+SWubNm0ICAhg9+7d9OrVq9T9zJs3j3nz5vmVtWnTptS6QoiGYUNyNv/6+SCeU403fZqHcf8AWUVG1C69TsfAmFa1HUapYs0hviVGT2tsCZHkswHKdhTy983fku7IByDSFMRT3YZVeQt7rd+Ct9vtJVonbTYbeXl5533dBx98QKtWrejTp4/fvmw2W4X2NWPGDNLS0vwep1tUhRAN0+Ktx3zJJ8BvSVkcziyovYCEqOP0Oh0PdBzgGyDV1BLK9A79ajkqUR2+TdnjSz6heOGElcf2Vvlxar0F1Gq1kpub61eWk5NDcHBwGa8otnDhQu64444S+8rJyanwvoQQl5asAleJsuzCkmVCiDPahUbzYs/RFHpcWE4NhBINT7azsFxlF6vWW0Dbtm2Lpmns2LHDV7Zlyxa/W/Ln2rx5M4mJidx2221+5QkJCWzZssX3fP/+/TgcDtq3b1/lcQsh6q/ezW1+z0MCDXRsJH+oCnEhmqZJ8tnA9Y5uXrIsqmTZxar1BDQoKIixY8cya9Ys8vLySExM5K233mLKlCllvmbhwoWMGDGCmBj/udkmT57MwoUL2bx5M3a7nccee4wbbrhBBiAJIfzc3qMpozpG0zg0kO5NQpl1TRsCDLX+dSiEELWuS3gc09r3pVVwBK2DI5jWvi+dw2Or/Di1fgse4NVXX2Xq1KnExsYSHBzMzJkzfYOJrFYry5cvZ8CAAQA4nU4WLVrEW2+9VWI/Q4YM4YknnmDkyJHk5uYydOhQ3n777Ro9FyFE3Rdg0DHx8iZMrNmBxEIIUS/0axRPv0bVOzBTU+r0ugritCFDhrBy5craDkMIIeoc+X4UQlQFueckhBBCCCFqlCSgQgghhBCiRkkCKoQQQgghapQkoEIIIYQQokZJAiqEEEIIIWqUjIIvRa9evQgJCSlX3YKCgnq3dKfEXDMk5ppR0zEHBgbyxRdf1Njx6pqG/v14MS618wU550tBRc63It+PkoBepOjoaNLS0mo7jAqRmGuGxFwz6mPMl4pL7b251M4X5JwvBdV1vnILXgghhBBC1ChJQIUQQgghRI2SBFQIIYQQQtQoSUAv0owZM2o7hAqTmGuGxFwz6mPMl4pL7b251M4X5JwvBdV1vjIISQghhBBC1ChpARVCCCGEEDVKElAhhBBCCFGjJAEVQgghhBA1ShJQIYQQQghRoyQBFUIIIYQQNUoS0HNkZ2czfvx4goODiYuL4+WXXy6z7o8//khCQgIWi4WePXuydetWv+3z58+ncePGWK1WbrzxRrKysup0zD/88AM6nQ6r1ep7zJ07t1ZjdjqdjB07lhYtWqBpGt98802JOnXtOl8o5rp4ndeuXcuwYcOIiIggIiKCkSNHsm/fPr86de06XyjmmrzO4oyKfB/VR5MnTyYgIMDvc5WcnOzbfuTIEYYOHUpQUBDx8fF8+OGHtRht5c2fP58ePXpgMpm46aab/LYlJibSu3dvLBYLHTt25Pvvv/fb/vHHH9OqVSssFgvXXHMNSUlJNRl6pZzvfFu0aIHZbPa93506dfLbfqFcoC5yOBzcddddxMfHExwcTKdOnVi0aJFve428x0r4ueWWW9SYMWNUTk6O2rZtm4qKilJff/11iXrp6ekqNDRUvfvuu6qoqEi9+OKLqlmzZqqoqEgppdSKFStUeHi42rRpk8rNzVXjxo1T48ePr9Mxr169WjVq1KhaYqxszA6HQ7300kvqp59+Uk2aNFHLly/3214Xr/OFYq6L1/nrr79WH374ocrOzlYOh0M9/PDDqn379r7tdfE6XyjmmrzO4ozyvn/11aRJk9QjjzxS5vZ+/fqp++67TxUUFKjVq1crq9Wqtm/fXoMRVo1PPvlELVu2TE2bNk1NmDDBV+50OlWLFi3UnDlzVFFRkfrwww9VSEiIOnHihFJKqZ07d6qgoCD17bffqoKCAvXnP/9ZXXHFFbV1GuVW1vkqpVTz5s1LfI+fdqHf1brKbrerxx9/XB04cEB5vV61Zs0aFRISon799dcae48lAT2L3W5XAQEBfl8Wf/vb39TYsWNL1H3zzTdV9+7dfc+9Xq9q0qSJ+vLLL5VSSk2cOFH95S9/8W3fu3evMhgMKjs7u87GXFM/2BWJ+WylfQnUxet8oZjr+nVWSqkTJ04oQKWnpyul6v51Li1mSUBr3sW8f/XF+RLQ0/8uMjMzfWUTJ05Uf/3rX2sqvCo3e/Zsv4RsxYoVKjo6Wnk8Hl9Z37591fz585VSxe/3jTfe6NuWm5urTCaTSkxMrLmgL8K556vU+RPQC/2u1icjRoxQL7zwQo29x3IL/ix79+7F6/WSkJDgK+vatSuJiYkl6iYmJtK1a1ffc03TuOyyy3x1z93epk0bAgIC2L17d52NGSAjI4OYmBiaN2/OPffcQ2ZmZpXGW9GYL6QuXufyqOvX+ccffyQmJoaIiAigflznc2OGmrnO4oyq/ndSV7355puEh4fTpUsXFixY4CtPTEykefPmhIWF+coa2vknJibSuXNndLoz6cPZ53jud0VwcDCtWrWq99dg0qRJREVFcdVVV/HLL7/4ysvzu1of5Ofns3HjRhISEmrsPZYE9Cx2u53Q0FC/MpvNRl5eXql1bTZbmXUvtL0uxty+fXu2bNnCsWPH+Pnnnzl69CiTJk2q0ngrGnN59lXXrvOF1PXrfPDgQaZPn+7Xd6+uX+fSYq6p6yzOqMp/J3XVn/70J/bu3UtaWhovv/wyDz/8MJ988glQc/9OalNd+e2rSe+//z6HDx8mOTmZCRMmMGLECF+fx4Zwvl6vl8mTJ9OzZ0+GDh1aY++xJKBnsVqt5Obm+pXl5OQQHBxcat2cnJwy615oe12MOSYmhk6dOqHT6WjatCnz58/n66+/pqCgoNZiLs++6tp1vpC6fJ2PHDnC4MGDeeSRR5gwYYLfvurqdS4r5pq6zuKMqvx3UlddfvnlREZGYjAYuPrqq5k2bRpLliwBau7fSW2qK799Nal///6YzWbMZjP33Xcf3bp1Y/ny5UD9P1+lFPfeey/Hjh3jo48+QtO0GnuPJQE9S9u2bdE0jR07dvjKtmzZ4nc76bSEhAS2bNnie66UYtu2bb66527fv38/DoeD9u3b19mYz6XT6VDF/YRrLeYLqYvXuaLqynU+evQogwYN4u677+ahhx7y21ZXr/P5Yj5XdV1ncUZ1/jupq05/rqD430lSUhLZ2dm+7Q3t/BMSEti+fTter9dXdvY5nvtdYbfbOXDgQIO6Bue+5xX5Xa1LlFJMmzaNLVu2sHz5cqxWK1CD73Gle6s2UBMnTlTXXXedys3NVdu3b1eNGjU674jy9957zzfquWnTpn6j4CMiItTvv/+u8vLy1IQJE6pt1HBVxfz999+rQ4cOKa/Xq1JTU9WYMWPUsGHDajVmpZQqKipShYWFqlmzZurzzz9XhYWFvs7RdfE6XyjmunidU1JSVOvWrdUTTzxR6n7q4nW+UMw1eZ3FGRX5d1IfffTRRyo3N1d5PB61Zs0aFRkZqf73v//5tvft21dNmzZNFRQUqB9//FEFBwfXy1HwLpdLFRYWqlmzZqlx48apwsJC5XQ6fSOkn3nmGVVUVKQWL15c6gjplStXqsLCQvXQQw/Vi1HwZZ1vUlKS+umnn5TD4VAOh0O9+eabKigoSO3fv18pdeHf1brsj3/8o+rWrZvfoDmlVI29x5KAniMrK0uNHTtWBQUFqZiYGPXSSy/5tgUFBamffvrJ93z16tWqU6dOKjAwUPXo0UNt3rzZb1///ve/VWxsrAoKClLXX399iTe5rsX84osvqiZNmiiz2azi4uLUlClTVFpaWq3H3Lx5cwX4PVavXu3bXhev8/lirovX+YknnlCACgoK8nskJSX56te163yhmGvyOoszzvf+NQQDBgxQoaGhymq1qo4dO6rXX3/db3tycrIaPHiwMpvNqnnz5mrRokW1FOnFmT17donvsEmTJimllNq2bZu64oorVGBgoGrfvr1atWqV32sXL16s4uPjldlsVldffbU6fPhwLZxBxZR1vjt27FBdunRRQUFBKiwsTPXr18/v90epC+cCddHhw4cVoEwmk9/355w5c5RSNfMea0rJ/SghhBBCCFFzpA+oEEIIIYSoUZKACiGEEEKIGiUJqBBCCCGEqFGSgAohhBBCiBolCagQQgghhKhRkoAKIYQQQogaJQmoEEIIIYSoUZKACiGEEEKIGiUJqBBCCFEN7r77bl5++WXf89GjR/PRRx/VXkAN0MaNG+nRowcFBQVVvu833niD2267rcr3K4pJAiouabNmzeL999+v7TAu2p133sn3339f22EI0WCkpaUxd+5cRo0aRZ8+fRg9ejQzZ85k586dtR1avXbs2DF69OjB/v37azsUPz169GDNmjV+Zbfddhv//ve/aymihk8SUFHjnnjiCXr06MFzzz1XYtvjjz9Ojx49/FoNjhw5wsyZMxk2bBh9+/Zl1KhRPPLII2RmZgJnvtBKexw+fLjMOPbs2cP69esZO3ZsVZ9ijbvzzjuZP38+Xq+3tkMRot47evQot956K/v27WPmzJksWbKEF154gZYtW/p9N9V3SincbneNHa+qj1XdsVssFmw2W7Ue41ImCaioFY0aNWLFihU4nU5fmd1uZ/Xq1TRq1MhX5nK5mD59Og6HgxdffJGPP/6YJ554gkaNGlFYWOi3zzfeeINvvvnG79G0adMyY1i8eDFDhgwhMDCw6k+whvXu3ZvCwkLWrl1b26EIUe89++yzREZG8tZbb9G/f3+aNGlCu3btuPvuu3nxxRd99fbv38/9999P//79GTZsGP/4xz+w2+3lOoZSijfeeIORI0fSp08frr32Wl599dUy65++HbxkyRJGjBhB//79efzxxykqKvLV8Xq9LFiwgNGjR9OvXz9uueUWfv75Z9/207erf/31VyZOnEjv3r3Zs2dPiWOd/qN+xYoVTJ48mb59+zJx4kQSExP96m3evJkpU6bQr18/Ro0axSuvvOL3nT569GgWLFjA448/zsCBA3nxxRcZM2YMADfddBM9evTg7rvvBkp2V4DiFsg33njD97xHjx4sXbqUP//5z/Tr149Fixb5xTJhwgT69u3LXXfdRXJysm9bcnIyDz74IEOHDmXgwIHccccdbNmyxS9OgAcffJAePXr4np97C97r9fLGG28wYsQI+vTpw2233camTZtKXN/169czceJE+vfvzx//+EfS0tJKXGMhCaioJZ06dSIsLIyffvrJV7ZixQrat29P48aNfWUHDx4kJSWFRx99lISEBOLi4ujRowcPPfSQXz2A0NBQIiMj/R56vb7U43s8Hr777jsGDBjgV96jRw+WLVvG/fffT79+/ZgwYQK7d+9m3759TJ48+f/bu/O4qMr9D+CfYREGBllFQmURVxiQa+CKRa6YAmqSyzUlt6uhKRrmdQO9lik3EzVNMbUkUzFUzFxCc0/MmyigJSpo7huCA7LO/P7wx8mRQUEPMwx83q/XvF7Oc5bne476ne8855xn4Ofnh7CwMGH0tcy2bdswYMAAdOrUCSEhIUhMTFRbvmTJEvTv3x+dO3dGcHAwvv76a7XRyqioKEybNg1r165Fjx490KNHD6xYsUJY/qIPKwMDA3Tu3Bk///zzi049ET3Hw4cPkZycjOHDh2vMHxYWFgCAR48eYfz48fDw8EBcXByWLFmCq1evIioqqlL97N+/Hxs3bsSMGTOwbds2fPbZZ3B2dn7uNleuXMHhw4exdOlSfPHFFzh9+jRiYmKE5evWrcOePXswc+ZMbN68Ge+88w4iIiLwxx9/qO1n+fLlCA8Px9atW+Hk5FRhf8uWLUNoaCji4uLg5uaG8PBw4Yv/tWvXMGnSJPTs2RObNm3Cf/7zHxw7dqzcJetvv/0WrVu3xsaNG/HPf/4T33zzDYC/Bwyio6Mrdb7KrFq1Ct27d8eWLVvQs2dPtVg/+ugjfPPNN6hXrx4iIiKEHJufn48uXbpgxYoV2LBhA7y8vDBp0iQ8fPhQiBEA5s2bhz179gjvn7Vx40Z8//33mDp1Kr7//nu0adMGkyZNKldgrl69Gh9//DHWrl2Le/fu1apRczGxACWdCQwMVCvUdu7cKXzzLGNtbQ0DAwPs378fpaWlovWdkZEBhUKB1q1bl1u2Zs0aBAcH47vvvkODBg0wZ84cfP755/jwww+xdu1a3Lp1S6043L17N2JjY/Hhhx8iPj5eGCU5ePCgsI5MJsPcuXMRHx+PSZMmYePGjdi+fbtav8nJycjOzkZsbCymTp2KdevW4ddffwVQuQ+r1q1b4/Tp06KdI6K66Nq1a1CpVHBxcXnueps3b4a7uzvGjRsHFxcXtG7dGjNnzsTBgwfLfUHV5NatW7C1tUX79u3h4OAALy8v9O3b97nbFBUVISoqCs2bN4evry8mT56M7du3Iz8/H0VFRVi3bh0iIyPRoUMHNG7cGAMGDIC/v3+5XPPBBx/A19cXTZo0EQpqTQYPHgx/f380bdoUs2bNgkqlwu7duwE8KXb79OmDwYMHo0mTJvjHP/6ByZMnY9u2bVCpVMI+2rdvj6FDh6Jx48Zo3LgxrK2tAfw9YGBpafnCc/W0t99+G3379kWjRo3g4OAgtP/rX/+Cr68vmjdvjnnz5uHKlSs4efIkAKBVq1YYMGAAmjVrBmdnZ4SHh6N+/fo4fvw4AAgxWVhYwM7OTnj/rLi4OLz//vvo3r07XFxcMHXqVDg6OiI+Pr7c+W3Tpg1atGiBwYMHq42S0t+MdB0A1V19+vTBqlWrcPfuXeTl5eHixYvo0aMHdu3aJaxjb2+PKVOmYNmyZfjqq68gl8vh6+uLPn36oEGDBmr7GzFiBAwM/v5OZWZmhr1792rs+9atW6hXr57G+3uCg4PRvXt3AE8uAYWFheGDDz5A27ZtAQD9+vXDDz/8IKy/atUqTJkyBf7+/gCARo0a4Y8//kBCQoLQNnr0aGF9R0dHZGRkICkpCQMGDBDara2tMWXKFEgkEri4uGDLli04deoUOnbsqPZhZWRkJHxgPc3Ozg63bt2q6HQTkYgyMjKQnJxc7ioK8KSItbGxee723bt3x8aNGxEcHIxOnTrBz88PXbp0Ucthz3J0dIStra3w3svLC8XFxbh27RoMDQ1RUFCAcePGqW1TXFwMHx8ftTZNX7w1kcvlwp9NTEzQokULZGZmAnhy/BkZGfjxxx+FdZRKJQoLC3H//n3Y2dlVqa/Kqmh/T8dqZ2eH1157DZmZmejQoQPy8/OxatUqHD16FPfv30dpaSkKCwurlC8VCgXu3buHNm3aCG0SiQReXl7lnjVo1qyZWiyV+UJSF7EAJZ2xs7ND+/btsWvXLuTm5qJr164wMzMrt97gwYPRt29fnDx5EqmpqdixYwfWr1+P1atXo0WLFsJ6ixYtUrucJJFIKuy7oKAAxsbGGpc1b95c+HNZsm/atKnQZmNjg+zsbADA48ePce3aNURGRmLu3LnCOiUlJXjttdeE9/v27cOmTZtw7do1PH78uNzysj6ejtnW1lZIXJX5sDIxMUFJSQlKSkpgZMT/2kQvo3HjxpBIJMjKykKrVq0qXO/x48fw9/dHWFhYuWXPfjnWxMHBAQkJCThx4gSSk5Pxn//8B61atcKyZcuem7ueFw/w5FL000Uq8CQ3PE0qlVZ5/8/Kz89HSEgIQkJCyi17+ot9ZfsyMDBQGzkFND9k9DKxL1myBL/99hsmTZqExo0bw8TEBJMmTUJxcXGV91UZT+dfiURS7rjoCX5KkU4FBgZi+fLlyM/Px/z58ytcTyaToWvXrujatSvCwsIwdOhQxMXFYd68ecI6DRs2fO5DR0+zsrJCfn4+SktLy93npal4ezahPH1vEQBERkaW+2Zets3Zs2cxe/ZsjB8/Hu3bt4e5uTl27NiBffv2PbffpxNXZT6scnNzIZPJWHwSvQIrKyu0a9cO3377LXr06FEuPzx69AgWFhZo2bIlDh06BEdHxwrvNX8RU1NT+Pv7w9/fH3369EFoaChu376tdmn5aTdu3MCDBw+E0dXU1FQYGxujcePGUKlUMDY2xq1bt+Dt7f1S8TwrLS1NGPErLCzEhQsXhKtDLVu2xOXLlyudc8uUffF/dsYOa2tr3L9/X3ifn5+Pa9euVSnWt956CwBw79493Lx5E66urgCAM2fOICgoSLgilZubi9u3b6ttb2Rk9NzbvGQyGezs7HDmzBnh/KpUKpw9e1bYL1UNP6lIp9544w18+umnMDc3x+uvv16pbYyMjNCoUaNyT8FXRcuWLaFSqZCZmal2uaSqbG1tYWdnh+vXr6vdEP+0s2fPolGjRggNDRXabty4UeW+XvRhlZmZiZYtW77UcRDR3z7++GOMGjUKY8aMwciRI+Hi4oLHjx/j8OHDSE5OxurVqxESEoKEhATMmjUL7733HurXr4+srCwcOnQIM2fOfGEfO3fuhFKphFwuh4mJCfbu3QsLC4vnXrqvV68eoqKiMHHiROTk5CAmJgZBQUHClaOhQ4fi888/R2lpKdq0aYNHjx7h999/R8OGDYXCsSq2bNmCxo0bw8nJCevWrQMABAQEAACGDx+OkSNH4r///S+CgoJgYmKCS5cuITU1FZMmTapwn9bW1jAxMcHx48dhZ2eHevXqQSaT4fXXX8eyZcvw66+/omHDhoiNja3SSPDq1athYWEBS0tLxMTEwMnJCe3atQMAODk54cCBA+jcuTOUSiW+/PLLcl8aHB0dcfLkScjlctSrVw/169cv18ewYcOwZs0aNGrUCM2aNcPWrVtx48YNjaPA9GIsQEmnjIyMsH37dkgkEo3J5s8//8Tq1avRt29fuLq6QiKR4MiRIzh+/Dhmz56ttm5OTg7u3bun1mZpaanxUru1tTVatmyJlJSUVypAgSf3d8bExMDMzAwdOnRAUVERUlNToVQqMXDgQDRp0gQ3btzAzz//jNatW+OXX37Br7/++tyb/59VmQ+rlJQUdOjQ4ZWOhYieFCwbNmzA119/jQULFuDBgwewtbWFp6cnpkyZAuDJ/elff/01li1bhg8++ADFxcVo1KgR3nzzzUr1YWFhgXXr1mHx4sVQqVRo0aIFlixZgnr16lW4jbOzMzp37oyJEydCoVDA399frdibMGECrK2t8fXXX+PGjRuoX78+3N3d1e5Br4oPPvgAX3/9NTIyMuDi4oIvvvhCKHZbtmyJr776CitXrsTIkSNhaGiIxo0bv/BBKiMjI0RERCA2NhYrVqyAt7c3Vq9ejeDgYPz555+YMWMGTE1NMXbs2CqNgIaFhWHRokX466+/4O7ujkWLFgm3KIWHh2Pu3Ll4//33YWNjg1GjRglPwJeZPHkyvvjiC/zwww+wt7fHzp07y/UxdOhQ5OXl4b///S9ycnLQrFkzxMTEVOqWCypPouLNCaRlUVFRyM/Px6JFizQuHzt2LNzd3TF58mQ8fPgQa9aswcmTJ3Hz5k0YGBigSZMmGDhwIPr16wfgyWhi2dxyz1q6dCk6deqkcVl8fDz27duH2NhYoc3HxwdffPGF8GDBxYsXMXjwYCQmJsLR0REA8NNPP2HBggVqv5qxa9cuxMXFISsrC2ZmZmjRogWGDx+Ojh07AgBiYmKQmJiIkpISvPHGG2jatCkSEhKEJKfpnEybNg1mZmaIiorCwYMHsW7dOmRlZQkfVhMnThQuj927d0+YVYDJkKj2KXuIZsOGDdXeV1lO3bRp0yt/QSeqCAtQqrMKCgrwzjvv4L///a/oT2pq25dffomHDx9W6tIfEekfFqBU2/ASPNVZpqammDt3bq2YIsPS0hKDBw/WdRhERESVwhFQIiIiItIq/hISEREREWkVC1AiIiIi0ioWoERERESkVSxAiYiIiEirWIASERERkVaxACUiIiIirWIBSkRERERaxQKUiIiIiLSKBSgRERERaRULUCIiIiLSKhagRER6ZOzYsWjUqBHq168PFxcXfPrpp8IyFxcXSKVSyGQyyGQyeHh4qG176NAhyOVymJmZwdfXF2fOnNF2+EREAFiAEhHplcmTJ+PixYvIzc3FkSNHEBcXhy1btgjLt23bBoVCAYVCgfT0dKH9/v37CA4OxrRp05CdnY0hQ4YgKCgIhYWFujgMIqrjWIBqEBgYqOsQiIg0cnd3h1QqFd4bGBjg4sWLL9wuISEBzZo1w/Dhw2FiYoLw8HAolUokJSVVqX/mRyISAwtQDQoKCnQdAhFRhf7973/D3NwcTk5OyMvLw7Bhw4RlI0aMQIMGDeDv749jx44J7WlpafD29hbeSyQSeHl5IS0trcJ+oqOjYW9vr/Z68OBBtRwTEdUtLECJiPTMggULoFAocPLkSQwdOhTW1tYAgLi4OGRlZeHq1asYNGgQevfujStXrgAAFAoFrKys1PZjZWWFR48eVdhPREQE7ty5o/YyMzOrtuMiorqDBSgRkR6SSCTw9fWFqakpIiMjAQB+fn6QSqWQSqUYP348/vGPf2D37t0AAJlMhpycHLV95OTkwMLCQuuxExGxACUi0mMlJSW4dOmSxmUGBgZQqVQAALlcjpSUFGGZSqXC2bNnIZfLtREmEZGaGlmALl++HD4+PjAxMcHgwYM1rnPw4EFIJBJMnz5drX3r1q1wc3ODmZkZunXrJlx+IiLSd9nZ2diwYQNyc3OhVCpx7NgxrFy5Et27d8fVq1dx5MgRFBUVoaioCLGxsfjtt9/Qs2dPAMCAAQOQkZGBuLg4FBUVISYmBgDQvXt3XR4SEdVRNbIAdXR0xKxZszBmzBiNy4uKivDhhx+iffv2au3nz59HaGgoVq5cifv378PLywvvvvuuNkImIqp2EokE69atg7OzMywtLTFq1ChMnToVEyZMgEKhwMSJE2FjYwMHBwd88803+PHHH+Hm5gYAsLW1xfbt2/HZZ5/B0tIS3333HRITE2FiYqLjoyKiushI1wFoMmDAAABASkoK7t27V275ggUL0KdPH9y8eVOtPS4uDgEBAcI3/nnz5qFBgwZIT08vNyEzEZG+sbKywoEDBzQuc3d3V7vErom/v/9zn3onItKWGjkC+jwXLlzAxo0bMXv27HLLnp1mxMLCAm5ublWeZiQ/P786QiciIiIi6GEBOn78eCxatEjjVCCcZoSIiIio5quRl+ArEhcXB6lUiuDgYI3LOc1I7ZBzcgpKFZlqbYYyV1i2W6yjiIhI30zZkY7MB3lqba425lgczNuxiGoCvSpAk5KScPToUTg4OAB4UlwaGhri1KlTSEpKKjfNiEKhwKVLlzjNiJ4pVWSi5OE5XYdBRHos80Eezt1W6DoMIqpAjbwEX1JSgoKCApSUlECpVKKgoADFxcWIiYnBH3/8gZSUFKSkpCAoKAjvv/8+Nm/eDAAYNmwYdu/ejaSkJBQUFCAyMhJeXl58AImIiIioBqmRBej8+fMhlUrxySefID4+HlKpFGPGjIGlpSUcHByEl1Qqhbm5OWxtbQEArVu3xrp16zB27FjY2Njg9OnT2LJli46PhoiIiIieViMvwUdFRSEqKuqF661fv75cW0hICEJCQsQPioiIiIhEUSNHQImIiIio9mIBSkRERERaxQKUiIiIiLSKBSgRERERaRULUCIiIiLSKhagRERERKRVNXIaJqpd+NOaRERE9DQWoFTt+NOaRERE9DRegiciIiIirWIBSkRERERaxQKUiIjqBCMDia5DIKL/x3tAiYioTnCylmLKjnRkPshTa3e1McfiYA8dRUVUN3EElIhIj4wdOxaNGjVC/fr14eLigk8//VRYlpaWhg4dOsDMzAzu7u44cOCA2rZbt26Fm5sbzMzM0K1bN1y5ckXb4etc5oM8nLutUHs9W5ASUfVjAUpEpEcmT56MixcvIjc3F0eOHEFcXBy2bNmC4uJiBAYGIigoCNnZ2YiMjET//v1x584dAMD58+cRGhqKlStX4v79+/Dy8sK7776r46MhorqKBSgRkR5xd3eHVCoV3hsYGODixYs4ePAg8vPzMX36dJiYmGDQoEGQy+WIj48HAMTFxSEgIAA9e/aEVCrFvHnzcObMGaSnp+vqUIioDmMBSkSkZ/7973/D3NwcTk5OyMvLw7Bhw5CWlgZPT08YGPyd1r29vZGWlgbgyeV5b29vYZmFhQXc3NyE5ZpER0fD3t5e7ZWfn19tx0VEdQcLUCIiPbNgwQIoFAqcPHkSQ4cOhbW1NRQKBaysrNTWs7KywqNHjwDghcs1iYiIwJ07d9ReZmZmYh8OEdVBLECJiPSQRCKBr68vTE1NERkZCZlMhpycHLV1cnJyYGFhAQAvXE5EpE2chomISI+VlJTg0qVL6N27NxYuXAilUilchk9JScGQIUMAAHK5HCkpKcJ2CoUCly5dglwu10XYotE0rZKfq62OoiGiyqqRI6DLly+Hj48PTExMMHjwYKH9woUL6NevHxo2bAgrKyu8+eabOHXqlNq2nGaEiGqr7OxsbNiwAbm5uVAqlTh27BhWrlyJ7t27w9/fH1KpFIsWLUJhYSHi4+ORmpqKkJAQAMCwYcOwe/duJCUloaCgAJGRkfDy8oKHh37Pf6lpWqUbuY91HRYRvUCNLEAdHR0xa9YsjBkzRq394cOHePvtt5Geno779+9j4MCB6N27N/Lynnz75TQjRFSbSSQSrFu3Ds7OzrC0tMSoUaMwdepUTJgwAcbGxkhMTMS2bdtgZWWFOXPmICEhAfb29gCA1q1bY926dRg7dixsbGxw+vRpbNmyRcdHRER1VY28BD9gwAAATy4f3bt3T2hv164d2rVrJ7yfOHEiPv74Y/zxxx94/fXX1aYZAYB58+ahQYMGSE9P1/tv+UREVlZW5SaXf5qnpyeSk5MrXB4SEiKMiBIR6VKNHAGtrN9++w1KpRLNmjUDwGlGiIiIiPSB3hag9+/fx7BhwzBv3jxYWloC4DQjRERERPpALwvQnJwc9OrVCwEBAZg2bZrQzmlGajFJjbxbhIiIiF6C3n2qlxWfvr6+iImJUVtWW6cZqWlyTk5BqSJTrc3Y3g/K/Osa28VgaO6ksV9DmSss2y0WpQ8iIiLSjhpZgJaUlAgvpVKJgoICGBoa4vHjxwgICIC7uztWrFhRbrthw4bB19cXSUlJ8PPzqzXTjOhKRYVmqSITJQ/PqbUbyppW2C4WTfvnyCgREZH+qZGf3vPnz8fcuXOF9/Hx8RgxYgTeeustnDhxAmfPnlWbPmT37t3o0qWL2jQjt27dQocOHTjNyCuo7oJSDBwZJSIi0j81sgCNiopCVFSUxmUjRox47racZkRPiDhyqXFklIiIiGqsGlmAUu1X0cilWPeMEhERUc3FApR0Rh8u8RMREZH49HIaJiIiIiLSXyxAiYiIiEirWIASERERkVaxACUiIiIirWIBSkRERERaxafgqc7hxPVERES6xQKU6hxOXE9ERKRbvARPRERERFrFApSIiIiItIoFKBERERFpFQtQIiIiItIqFqBERHqisLAQo0ePhqurKywsLODh4YGNGzcKy11cXCCVSiGTySCTyeDh4aG2/aFDhyCXy2FmZgZfX1+cOXNG24dARASABSgRkd4oKSmBo6Mj9u/fj9zcXKxatQrjx4/Hr7/+Kqyzbds2KBQKKBQKpKenC+33799HcHAwpk2bhuzsbAwZMgRBQUEoLCzUxaEQUR3HApSISE+Ym5tj3rx5aNq0KSQSCfz8/NC5c2ccP378hdsmJCSgWbNmGD58OExMTBAeHg6lUomkpCQtRE5EpI4FKBGRnsrLy8OpU6cgl8uFthEjRqBBgwbw9/fHsWPHhPa0tDR4e3sL7yUSCby8vJCWllbh/qOjo2Fvb6/2ys/Pr5ZjIaK6hQUoEZEeUiqVCA0Nha+vL3r27AkAiIuLQ1ZWFq5evYpBgwahd+/euHLlCgBAoVDAyspKbR9WVlZ49OhRhX1ERETgzp07ai8zM7NqOyYiqjtELUC3bNmCX375RXj/ySefoGPHjhg0aBCysrIqvZ/ly5fDx8cHJiYmGDx4sNqytLQ0dOjQAWZmZnB3d8eBAwfUlm/duhVubm4wMzNDt27dhORLRKQrYuXGMiqVCuPGjcONGzewefNmSCQSAICfnx+kUimkUinGjx+Pf/zjH9i9ezcAQCaTIScnR20/OTk5sLCwePkDIyJ6SaIWoBs3boSlpSUA4NSpU9i3bx/mzZuHZs2aYfHiyv/OtqOjI2bNmoUxY8aotRcXFyMwMBBBQUHIzs5GZGQk+vfvjzt37gAAzp8/j9DQUKxcuRL379+Hl5cX3n33XfEOkIjoJYiVG4EnxWdYWBhSUlKwe/duyGSyCtc1MDCASqUCAMjlcqSkpKjt5+zZs2qX74mItEXUAvTu3btwdHQE8GS6j+7du6NHjx4YM2bMc+8zetaAAQPQr18/2NnZqbUfPHgQ+fn5mD59OkxMTDBo0CDI5XLEx8cDeHL5KSAgAD179oRUKsW8efNw5swZtSdBiYi0TazcCAATJkzAiRMnsHfvXtSvX19ov3r1Ko4cOYKioiIUFRUhNjYWv/32m3B5fsCAAcjIyEBcXByKiooQExMDAOjevbtIR6m/jAwkug6BqM4xEnNnMpkMd+7cgYODA44fP45//etfAJ7c7F5aWvrK+09LS4OnpycMDP6um729vYUEnpaWBl9fX2GZhYUF3NzckJaWVm4+vDLR0dGIjo5Wa2vevPkrx0pEVEas3HjlyhWsWLECJiYmaNKkidA+Y8YM9OvXDxMnTsTFixdRr149uLu748cff4SbmxsAwNbWFtu3b8eECRMwZswYyOVyJCYmwsTERNyD1UNO1lJM2ZGOzAd5au2uNuZYHKz5s4OIXo2oBWjXrl0xc+ZMNGnSBNnZ2ejUqRMA4MKFC2jUqNEr77+im+hf9Sb7iIgItbYePXq8cqxERGXEyo3Ozs7CJXVNnr7Erom/v3+VR1zriswHeTh3W6HrMIjqDFEL0KlTp8LR0RG3b9/GxIkThXuT7t69i5CQkFfe/4tuoudN9iSQiPpPm+iVVHduJCLSN6J+ShsZGeG9994r1z506FBR9i+Xy7Fw4UIolUrhMnxKSgqGDBkiLH96BEChUODSpUu8yb4OMjR3Qs7JKShVZKq1G9v76SgiqsuqOzcSEekb0YeJsrKy8Pvvv+PBgwflLhU9+1R7RUpKSoSXUqlEQUEBDA0N4e/vD6lUikWLFiE8PByJiYlITU1FQkICAGDYsGHw9fVFUlIS/Pz8EBkZCS8vrwrv/6TarVSRiZKH59TaDGVNdRQN1XVi5EYiotpC1AI0Pj4e0dHRsLW1hZ2dnTA3XZnKJtn58+dj7ty5avsdMWIE1q9fj8TERIwePRpz586Fi4sLEhISYG9vDwBo3bo11q1bh7Fjx+LWrVvo0KEDtmzZIt4BEhG9BLFyIxFRbSFqAbp+/Xp8+OGHGDZs2CvtJyoqClFRURqXeXp6Ijk5ucJtQ0JCeE8VEdUoYuVGIqLaQtR5QPPz8+Hv7y/mLomI9B5zIxGROlEL0MDAQOzfv1/MXRJpB5+ap2rE3EhEpE70T93169cjOTkZbm5uMDJS3/2kSZPE7o5IFBU9NW8oc4Vlu6r9VCKRJsyNRER/E7UA/fPPP9GiRQuUlpbiwoULasuevemeqKbR9NQ8kRiYG4mI1IlagK5atUrM3RER1QrMjURE6qrlxreCggJcu3YNANC4cWOYmppWRzdERHqFuZGI6AlRC9Di4mIsW7YMP/zwA4qKigAA9erVw8CBAzFx4sRy9z0REdUFzI1EROpEzXpLlizBwYMHMXfuXLRp0wbAk5/KXLJkCZRKJaZOnSpmd0REeoG5kYhInajTMO3btw9z5sxB9+7d0aBBAzRo0AA9evTA7NmzsXfvXjG7IiLSG8yNRETqRJ+IvuxnMZ9mb2+PvLw8MbsiItIbzI1EROpELUDlcjnWrFmD4uJioa2oqAhr1qyBp6enmF0REekN5kYiInWi3gM6depUTJw4EW+//TZatmwJ4Mn8d4aGhli+fLmYXRER6Q3mRiIidaIWoC1atMCOHTvw008/ISsrCwDQrVs39O7dm9ONEFGdxdxIRKRO9Lk/TE1NMWDAALF3S9VI009QGtv76SgaotqJuZGI6G+vXIAePXoUHTp0gJGREY4ePfrcdf38WNTURJp+gtJQ1lRH0RDVDsyNREQVe+UCNDw8HHv37oWNjQ3Cw8MrXE8ikeDkyZOv2h0RkV6ojtxYWFiIsLAw7N+/H/fu3YOTkxNmzpyJoUOHAgDS0tIwevRonD17Fi4uLli+fDm6du0qbL9161Z8/PHHuHnzJjp27Ii1a9fC2dn51Q6UiOglvHIB+ttvv2n8MxFRXVYdubGkpASOjo7Yv38/XF1dcezYMfTp0weurq7w8fFBYGAgxowZg0OHDmH79u3o378/MjIyYG9vj/PnzyM0NBQJCQno0qULZsyYgXfffRfJycmixEZEVBWiTsP0448/Cj8z97Ti4mL8+OOPYnZFRKQ3xMqN5ubmmDdvHpo2bQqJRAI/Pz907twZx48fx8GDB5Gfn4/p06fDxMQEgwYNglwuR3x8PAAgLi4OAQEB6NmzJ6RSKebNm4czZ84gPT1dtOMkIqosUQvQefPmQaFQlGvPy8vDvHnzxOyKiEhvVFduzMvLw6lTpyCXy5GWlgZPT08YGPyd1r29vZGWlgbgyeV5b29vYZmFhQXc3NyE5ZpER0fD3t5e7ZWfn//S8RIRlRG1AFWpVJBIJOXa7969C5lMJlo/WVlZ6NOnD2xsbGBvb4/hw4fj0aNHAIC//voLPXv2hLm5OVxdXbFp0ybR+iUiehnVkRuVSiVCQ0Ph6+uLnj17QqFQwMrKSm0dKysrITe+aLkmERERuHPnjtrLzMzspeIlInqaKNMwDR8+HMCTm+knTJgAQ0NDYZlSqcS1a9fQoUMHMboCAIwdOxYNGzbE9evXUVBQgHfeeQezZ8/GkiVLMGTIEHh5eWHHjh1ITk5GYGAg5HI55HK5aP0TEVVGdeVGlUqFcePG4caNG9i7dy8kEglkMhlycnLU1svJyYGFhQUAvHA5EZE2iVKAlk0hcv78ebRr1w5SqVRYZmxsjNdeew3dunUToysAQGZmJiZPngypVAqpVIp33nkHCQkJyMjIQHJyMnbu3AmpVAp/f38EBQXhm2++QXR0tGj9ExFVRnXkRpVKhbCwMKSkpCApKUkYQZXL5Vi4cCGUSqVwGT4lJQVDhgwRlqekpAj7USgUuHTpEr+cE5FOiFKAjh07FgDg6OiIHj16wMTERIzdVmjy5MnYuHEj3njjDRQUFCA+Ph59+/ZFWloanJ2dYW1tLazr7e2NAwcOVLiv6OjocsVp8+bNqy12Iqo7qiM3TpgwASdOnMD+/ftRv359od3f3x9SqRSLFi1CeHg4EhMTkZqaioSEBADAsGHD4Ovri6SkJPj5+SEyMhJeXl7w8PB45ZiIiKpK1HtA+/btW+3FJ/Ak0f7xxx+wtLREgwYNYGJigg8//JD3OBFRjSRWbrxy5QpWrFiBc+fOoUmTJpDJZJDJZPj0009hbGyMxMREbNu2DVZWVpgzZw4SEhJgb28PAGjdujXWrVuHsWPHwsbGBqdPn8aWLVteOSYiopch6k9xFhcXIzY2FklJSbh16xZKSkrUlosxEX1paSkCAgIwcuRIHD16FEVFRZg8eTKGDRuGIUOG8B4nIqpxxMqNzs7OUKlUFS739PR87ryeISEhCAkJqVzQRETVSNQR0OXLl2Pfvn0YOXIkJBIJPvroI7z33nuwtrbGzJkzRekjOzsb165dw8SJE2Fqaor69etj/Pjx+OmnnyCXy3HlyhU8fPhQWD8lJYX3OBGRTmkjNxIR6RNRC9D9+/fj3//+N/r27QtDQ0N07NgREydOxAcffICDBw+K0oednR2aNm2KFStWoKioCHl5eVi9ejXatGmD5s2bw9fXF7NmzcLjx49x+PBhJCYmYsSIEaL0TUT0MrSRG4mI9ImoBejDhw+F3xU2NzdHbm4uAMDHxwenTp0SrZ9t27bh0KFDcHBwgJOTE27cuIFvv/0WALBp0yb8+eefsLW1xfDhw7Fq1SqOgBKRTmkrN5K4jAzKz91KROIQ9R7QRo0a4ebNm3BwcICLiwv2798PDw8PHDt2TNT7ML28vLB//36Ny5o0aYKff/5ZtL6IiF6VtnIjicvJWoopO9KR+SBPrd3VxhyLgzl7ANGrELUA7du3L/744w/84x//QGhoKKZMmYLNmzejuLgY4eHhYnZFRKQ3mBv1V+aDPJy7Xf5nVIno1YhagL733nvCn9u3b4/4+Hj88ccfaNKkCefWJKI6i7mRiEidqAXosxwdHeHo6FidXRAR6R3mRiKq60QvQPfv34+NGzciKysLAODi4oIhQ4age/fuYndFRKQ3mBuJiP4magG6fv16xMbGIjg4WJjs+OzZs4iKisK1a9cQGhoqZndERHqBuZGISJ2oBejGjRuFue7KBAQEwN3dHUuXLmWSJaI6ibmRiEidqPOAFhUVwcvLq1y7l5cXCgsLxeyKiEhvMDcSEakTtQDt3bs3duzYUa59586d6NWrl5hdERHpDeZGIiJ1r3wJPiYm5u+dGRlh69at+PXXX4VfH0pPT8e1a9cQFBT0ql0REekN5kYiooq9cgF67tw5tfetWrUCAFy5cgUAIJPJ0KpVK2RkZLxqV0Q1Ss7JKShVZKq1GcpcYdlusY4iopqEuZGIqGKvXICuWrVKjDiI9E6pIhMlD8+9eEWqk5gbiYgqJtpT8CUlJejWrRvWrl0LNzc3sXZLItI0Ymds76ejaIjqBuZGIqLyRCtAjYyMYGlpCZVKJdYuSWSaRuwMZU11FA1R3cDcSERUnqhPwYeGhuLLL7+EQqEQc7dERHqNuZGISJ2oE9Fv374dWVlZ6NWrFxo1agRTU1O15d9++62Y3RER6QXmRiIidaIWoH5+fvDz4z2FVItIRP0vQnWUWLlx+fLlWL9+PVJTU9G/f39s2rRJWObi4oLbt2/D0NAQAODs7Iz09HRh+aFDhxAWFobLly/Dw8MDa9asQZs2bV45JiKilyHqp+vYsWPF3B2RzhmaO/HhLXplYuVGR0dHzJo1C0lJSbh371655du2bUNAQEC59vv37yM4OBhLly7FoEGD8OWXXyIoKAgXLlyAiYmJKLEREVWFqPeAEtVGZQ9vPf1S5t/QdVhUBw0YMAD9+vWDnZ1dlbZLSEhAs2bNMHz4cJiYmCA8PBxKpRJJSUnVFCkR0fOJOgLq6+sLiURS4fKTJ0+K1tcPP/yAyMhIZGZmws7ODl988QUGDBiAtLQ0jB49GmfPnoWLiwuWL1+Orl27itYvEVFVaSs3jhgxAkqlEh4eHvjkk0/QuXNnAEBaWhq8vb2F9SQSCby8vJCWloY+ffpUuL/o6GhER0ertTVv3lyUWImobhO1AP3iiy/U3peUlODPP//Ezp07MWbMGNH6OXDgACZPnoxNmzahY8eOuHfvHhQKBYqLixEYGIgxY8bg0KFD2L59O/r374+MjAzY29uL1j8RUVVoIzfGxcXh9ddfBwCsX78evXv3RmpqKpydnaFQKGBtba22vpWVFR49evTcfUZERCAiIkKtrUePHqLES0R1m+gPIT3L398fbm5uSExMRHBwsCj9zJkzB3PmzBG+3dvb28Pe3h4///wz8vPzMX36dBgYGGDQoEFYunQp4uPjERYWJkrfRERVpY3c+HQf48ePx6ZNm7B7926MGzcOMpkMOTk5auvn5OTAwsLilfslInoZWrkHtGXLlvj9999F2VdpaSlOnjyJBw8eoEWLFnB0dMT777+PnJwcpKWlwdPTEwYGfx+Wt7c30tLSKtxfdHS0UMCWvfLz80WJlYjoecTMjc8yMDAQJr+Xy+VISUkRlqlUKpw9exZyubxa+iYiehFRC9CCggK11+PHj3Ht2jXExsaiUaNGovRx+/ZtFBcXY9OmTThw4ADOnTuH27dvY/LkyVAoFLCyslJb/0WXmSIiInDnzh21l5mZmSixEhEB4uXGkpISFBQUoKSkBEqlEgUFBSguLsbVq1dx5MgRFBUVoaioCLGxsfjtt9/Qs2dPAE8eXsrIyEBcXByKiooQExMDAOjevXu1HC8R0YuIegm+S5cu5W60V6lUsLe3xyeffCJKH2XF4YQJE9C4cWMAwMyZM9GvXz/MmDGDl5mIqMYRKzfOnz8fc+fOFd7Hx8djxIgRmDZtGiZOnIiLFy+iXr16cHd3x48//ij89rytrS22b9+OCRMmYMyYMZDL5UhMTOQUTESkM6IWoF999ZXaewMDA1hbW6Nx48YwMhKnKysrKzRp0kTjE6VyuRwLFy6EUqkULsOnpKRgyJAhovStLzhvJVHNIlZujIqKQlRUlMZlT19i18Tf3/+5tyMREWmTqAVo2dQezybUkpIS/P7772jbtq0o/YwePRrLly/H22+/DXNzc3z22WcICgqCv78/pFIpFi1ahPDwcCQmJiI1NRUJCQmi9KsvyuatfJqhrKmOoqlj+MtJpIG2ciMRkb4Q9R7QcePGITc3t1y7QqHAuHHjROtnxowZ8PPzg7u7O9zc3IR5QI2NjZGYmIht27bBysoKc+bMQUJCAqdgIq0p++WkBwf6q71yTk7RdWikQ9rKjURE+kLU4RqVSqXx0nhubi6kUqlo/RgZGWHp0qVYunRpuWWenp5ITk4WrS+iqtI0Ak11m7ZyIxGRvhClAJ0+fTqAJ5eZ/vOf/6BevXrCstLSUmRkZMDLy0uMroiI9AZzY+1kZFDxr1oRUeWIUoCWfYNXqVQwMTGBqanp3x0YGSE4OBj9+/cXoysiIr3B3Fg7OVlLMWVHOjIf5Km1u9qYY3Gwh46iItIvohSgkZGRAABTU1NMmDAB5ubmYuyWiEivMTfWXpkP8nDutkLXYRDpLdEeQiotLUVCQgLu3r0r1i6JiPQecyMRUXmiFaCGhoZo0qQJ8vLyXrwyEVEdwdxIRFSeqNMwTZgwATExMcjKyhJzt0T67yXmB+V0TrUHcyMRkTpRp2GaP38+8vPz8e6775a74R4Afv75ZzG7I9IbZfODPvsLVYYyV1i2W6xxG07nVHswNxIRqRO1AJ08ebKYuyOqVVhQ1l3MjURE6kQtQPv27Svm7oiIagXmRiIidaLeAwoA9+7dw4YNG7BgwQI8fPgQAHD69Glcu3ZN7K6IiPQGcyMR0d9EHQFNS0tDWFgYmjZtivPnz+O9996DlZUV/ve//+HSpUtYsGCBmN0REekF5sZXp2nidz9XWx1FQ0SvStQR0CVLlmDEiBFYt26d2k/OdejQAWfPnhWzKyIivcHc+OrKJn5/+nUj97GuwyKilyRqAXrhwgX06tWrXLuNjQ2ys7PF7IqISG8wNxIRqRO1ADU3N8eDBw/KtV+4cAENGjQQsysiIr3B3EhEpE7UArRnz55YtmwZsrOzIZFIADy59ykmJga9e/cWsysiIr3B3EhEpE7UAjQsLAzOzs54++23hUmXR44cidatW2P06NFidkVEpDeYG4mI1InyFLxSqcS3336Lw4cPo6SkBL1790bXrl3x+PFjtGjRAs7OzmJ0Q0SkV5gbiYg0E2UEdO3atYiNjYWbmxu8vLxw6NAh7N+/Hz169GCCJaI6S+zcuHz5cvj4+MDExASDBw9WW5aWloYOHTrAzMwM7u7uOHDggNryrVu3ws3NDWZmZujWrRuuXLnySsdGRPQqRClAd+3ahdmzZ2PmzJn46KOPEBMTgz179kCpVIqxeyIivSR2bnR0dMSsWbMwZswYtfbi4mIEBgYiKCgI2dnZiIyMRP/+/XHnzh0AwPnz5xEaGoqVK1fi/v378PLywrvvvvvKx0dE9LJEKUBv3bqFtm3bCu/lcjkMDAxw9+5dMXZfoXv37sHOzg4dOnQQ2l40CkBEpC1i58YBAwagX79+sLOzU2s/ePAg8vPzMX36dJiYmGDQoEGQy+WIj48HAMTFxSEgIAA9e/aEVCrFvHnzcObMGaSnp7/8wRERvQJRCtDS0lIYGxurtRkaGqKkpESM3VcoIiIC7u7uwvsXjQIQEWmTtnJjWloaPD09YWDwd0r39vZGWlqasNzb21tYZmFhATc3N2F5RaKjo2Fvb6/2ys/PFzV2IqqbRHkISaVS4T//+Y/aL3wUFhZi0aJFkEqlQttnn30mRncAgEOHDiEjIwOjRo3CqlWrAKiPAhgYGGDQoEFYunQp4uPjERYWJlrfRESVoa3cqFAoYGVlpdZmZWUl3OdZ0fJHjx49d78RERGIiIhQa+vRo8crxUpEBIhUgPbt27dc29tvvy3GrjUqKirChAkTEBcXh9OnTwvtLxoF0CQ6OhrR0dFqbc2bNxc/aCKqc7SVG2UyGXJyctTacnJyYGFhUanlRETaJkoBGhkZKcZuKu2zzz5D9+7d0aZNG7UC9EWjAJrwGz4RVRdt5Ua5XI6FCxdCqVQKX8BTUlIwZMgQYXlKSoqwvkKhwKVLlyCXy7USHxHRs0SdiF4bLl68iPXr12Pu3LnllvFbPtV6ElG+M5KeKikpQUFBAUpKSqBUKlFQUIDi4mL4+/tDKpVi0aJFKCwsRHx8PFJTUxESEgIAGDZsGHbv3o2kpCQUFBQgMjISXl5e8PDw0PEREVFdpXefZkePHsWtW7fQokULAMDjx4/x+PFjODg4YNWqVUhNTa1wFIBI3xmaOyHn5BSUKjLV22WusGy3WEdRkbbMnz9f7ct3fHw8RowYgfXr1yMxMRGjR4/G3Llz4eLigoSEBNjb2wMAWrdujXXr1mHs2LG4desWOnTogC1btujqMIiI9K8AHTRoEAICAoT3mzdvxrfffotdu3bB1tZWGAUIDw9HYmIiUlNTkZCQoMOIicRVqshEycNzug6DdCAqKgpRUVEal3l6eiI5ObnCbUNCQoQRUaoeRgYSXYdApDf0rgCVSqVqT49aWlrC2NgYDg4OAPDcUQAiIqLq4mQtxZQd6ch8kKfW7mpjjsXBvN2B6Gl6V4A+KzQ0FKGhocL7F40CEBERVZfMB3k4d1uh6zCIajy9ewiJiIiIiPQbC1AiIiIi0ioWoERERESkVSxAiWoDzg9KRER6hJ9aRLUA5wclIiJ9wgKUSJdEHLnk/KBERKQvWIDqMU0jXsb2fjqKhl5GRSOX/HskIqLajAWoHtM04mUoa6qjaOhl8e+RiIjqGj6ERERERERaxQKUiIioGvE34onK4yV4IiKiasTfiCcqjwUoERFRNeNvxBOp4yV4IiIiItIqFqBEREREpFUsQImIiIhIq1iAEhEREZFWsQAlIiIiIq1iAUpERKQDnB+U6jIWoEREtUBoaCjq1asHmUwmvK5evSos/+uvv9CzZ0+Ym5vD1dUVmzZt0mG0BPw9P2j/dSfVXlN2pOs6NKJqp3cFaGFhIUaPHg1XV1dYWFjAw8MDGzduFJanpaWhQ4cOMDMzg7u7Ow4cOKDDaIl0TMKpfuuSKVOmQKFQCC8nJydh2ZAhQ9CsWTPcu3cP69atw5gxY5CWlqbDaAn4e37Qp1/PTlhPVBvp3adTSUkJHB0dsX//fri6uuLYsWPo06cPXF1d4ePjg8DAQIwZMwaHDh3C9u3b0b9/f2RkZMDe3l7XoRNpnaG5E3JOTkGpIlO9XeYKy3aLdRQVaVtGRgaSk5Oxc+dOSKVS+Pv7IygoCN988w2io6N1HR4R1UF6NwJqbm6OefPmoWnTppBIJPDz80Pnzp1x/PhxHDx4EPn5+Zg+fTpMTEwwaNAgyOVyxMfH6zpsIp0pVWSi5OE5tdezBSnVDqtXr4aNjQ3atGmDtWvXCu1paWlwdnaGtbW10Obt7f3CEdDo6GjY29urvfLz86stfiKqO/SuAH1WXl4eTp06BblcjrS0NHh6esLA4O/DelGSZYKlOomX5mudDz/8EBcuXMCdO3ewZMkSTJs2DT/88AMAQKFQwMrKSm19KysrPHr06Ln7jIiIwJ07d9ReZmZm1XUIRFSH6HUBqlQqERoaCl9fX/Ts2fOlkiwTLNVFZZfmHxzor/bKOTlF16HRS2rbti3s7OxgZGSEt956C2FhYcLVH5lMhpycHLX1c3JyYGFhoYtQiYj07x7QMiqVCuPGjcONGzewd+9eSCQSJlmiKii7NE+1k4GBAVQqFQBALpfjypUrePjwofAlPSUlBXK5XIcRElFdppcFqEqlQlhYGFJSUpCUlASZTAbgSZJduHAhlEqlcBk+JSUFQ4YM0WW4RETVbsuWLejduzfMzc1x/PhxLF++HMuWLQMANG/eHL6+vpg1axaio6Px22+/ITExEcePH9dx1OVN2ZFe7ilwP1dbHUVDRNVFLwvQCRMm4MSJE9i/fz/q168vtPv7+0MqlWLRokUIDw9HYmIiUlNTkZCQoMNoiYiq3/LlyzF27FiUlpbCyckJ8+fPx+DBg4XlmzZtwsiRI2Frawt7e3usWrWqRo6Alk1L9LSmtrwtiqi20bsC9MqVK1ixYgVMTEzQpEkToX3GjBmYMWMGEhMTMXr0aMydOxcuLi5ISEjgFExEVOsdPnz4ucubNGmCn3/+WUvREBE9n94VoM7OzsJ9TZp4enoiOTlZixERERGJhz/RSXWB3hWgRKTfODE+0fOV/UTns/fCutqYY3Gwh46iIhIXC1Ai0io+fU/0YpruheXIKNUmLECJiIj0AEdGqTZhAUpERKQnNI2MEukjvf4lJCIiIiLSPxwB1QOaHtowtvfTUTRERFST8N5Q0kcsQPWApoc2DGVNdRQN1WoSpgQifcN7Q0kf8dOmBuFIJ+maobmT5mmSLJrB0je6+joWsfDlNE9UF/HeUNI3LEBrEI50Uk1Q0b/D6izsKix8X2L/nOaJiKjmYwFKRJWisbATceSShSMRUd3BApSIXlpFI5fG9n5Q5l/nLSVERKQRC1AieiUVXbLnLSVEuvW8p+P50BLpGgtQIqq5+FQ+0Uur6Ol4P1dbPrREOsfsTkQ1lpgPJxHVRZoKzaa2ZjqKhuhvLECJqEbjw0lERLUPf4qTiIiIiLSKI6BEpH94byhRjcCHmehlMYuLgPeoEWnX86Z/IqKX8zK/Kc+Hmehl1boC9OHDhxg7dix2794NCwsLTJs2DZMnT67WPqt6jxo/OIleHad5qjpd5MeKVPR0NunO856av55TUOm/r6pO/9TMVoboIPeXjLpydNUvVazWFaATJkxAYWEhrl+/jitXrqBbt25o2bIlevfurd1AnnOJkB+cRFrGS/YAalB+BJ/Orqkq+nvJfJBf6b+vqk7/1NTWrMqX8qtaUIrVL4mnVmXlvLw8xMfH43//+x/q168PT09PjBkzBmvXrq3WBGsocy3XZtygA/LOL0Pp41vq7daeGtc3MHMEoGI729leDe0V/X80lDrAvPXEcuvXRrrKj8uOZuJWboFam+dr9eFqY15uXcf6UqhU5UfP2K5/7ddzCsq1A6jw772i9avCob5Jlf+9idFvbaDpvDnUN8VEv/L1ilgkKpWqfLbWU6dPn0a7du1QXFwstMXHx2POnDk4f/68xm2io6MRHR2t1ubs7AwrKyuN6+fn58PMjN/SNeG50YznpWL6eG5MTU2xc+dOXYdRZdrIj7WRPv4bFROPn8dfleOvSn6sVSOgCoUClpaWam1WVlZ49OhRhdtEREQgIiKi0n3Y29vjzp07Lx1jbcZzoxnPS8V4brRHG/mxNqrr/0Z5/Dz+6jr+WjUPqEwmQ25urlpbTk4OLCwsdBQREVHNwPxIRDVJrSpAW7RoAYlEgvT0dKEtJSUFcrlch1EREeke8yMR1SS1qgA1NzfHwIEDMXPmTDx69AhpaWlYs2YNRo4cqevQiIh0ivmRiGqSWlWAAsCXX34JY2NjvPbaa+jRowemT58u6hOedf1+qOfhudGM56ViPDfaVd35sTaq6/9Gefw8/upSq56CJyIiIqKar9aNgBIRERFRzcYClIiIiIi0igUoEREREWkVC1AiIiIi0ioWoERERESkVSxAiYiIiEirWIC+wOzZs9GgQQNYWlpi9OjRKCws1LjenTt3MHToUDRq1Aj169eHj48P9u3bp+Voq9fDhw/x7rvvwsLCAo6OjliyZEmF6x46dAhyuRxmZmbw9fXFmTNntBeollX2vJw4cQK9evWCra0tbG1t0adPH2RkZGg3WC2ryr+ZMuvXr4dEIsFXX31V/QFSnVfZf6NFRUUYOHAgXFxcIJFIsGfPHu0GWk3qev6q7PFfvnwZ7du3h42NDaysrNCpUyccPXpUu8FWE13laRagz7FmzRp89913SE5OxuXLl/HHH39gzpw5GtdVKBRo27YtTp48iYcPH2LGjBkYMGAArly5ouWoq8+ECRNQWFiI69evY+/evfj000+xe/fucuvdv38fwcHBmDZtGrKzszFkyBAEBQVVWLzru8qel+zsbIwcORKXL1/GzZs3IZfLERQUpIOItaey56bM/fv3sWDBAnh4eGgxSqrLqvJv1M/PDxs2bEDjxo21HGX1qev5q7LH36BBA3z33Xe4d+8esrOz8dFHHyEwMBBFRUU6iFpcOsvTKqpQp06dVMuWLRPe79u3T9WgQYNKb9+6dWvV1q1bqyM0rVMoFKp69eqpUlNThbYZM2aoBg4cWG7d1atXq15//XXhvVKpVDVu3Fj1448/aiVWbarKeXnW7du3VQBU9+7dq84QdeZlzk1oaKjqq6++Ur355puqlStXaiNMqsNe9v+vs7Ozavfu3dUdXrWr6/nrZY+/tLRUtX37dhUA1fXr16s7zGqlyzzNEdDnSEtLg7e3t/De29sbd+/exe3bt1+47Y0bN3Dx4sVaM5Jz4cIFKJVKyOVyoc3b2xtpaWnl1n32vEkkEnh5eWlcV99V5bw869ChQ3BwcICtrW11hqgzVT03hw4dwvnz5zFmzBhthUh13Kv8/60N6nr+epnjd3Z2homJCfr164f3338fjo6O2gi12ugyTxu98h5qMYVCASsrK+F92Z8fPXqEhg0bVrhdQUEB3n33XYwePRqtWrWq5ii1Q6FQwNLSUq3NysoKjx490riutbV1pdbVd1U5L0+7fPkyJkyYgKVLl1ZneDpVlXNTVFSEsLAwfPvttzAw4Pdi0o6X/f9bW9T1/PUyx3/lyhUUFBRg06ZNkEgk1R1itdNlnq6zmX7gwIGQSCQVvgBAJpMhJydH2KbszxYWFhXut6ioCO+88w4aNmyo9/85nyaTyZCbm6vWlpOTo/FcPHvenreuvqvKeSnz119/oXv37vj4448xaNCg6g5RZ6pybhYtWgR/f3+0bdtWW+ERvdT/39qkruevl/37NzU1RWhoKObPn6/3D9jqMk/X2QJ069atUKlUFb4AQC6XIyUlRdgmJSUFDRo0qHD0s6ioCCEhITA0NMSmTZtgZFR7BphbtGgBiUSC9PR0oS0lJUVt2L7Ms+dNpVLh7NmzGtfVd1U5LwBw7do1dO3aFWPHjsWUKVO0FaZOVOXcJCUl4fvvv4eDgwMcHBxw/PhxfPzxx3j//fe1GTLVMVX9/1vb1PX89ap//0VFRbh8+XJ1hacVOs3TL333aB2wevVqlZubm+ry5cuq+/fvq7p06aKaNm2axnWLiopUwcHBql69eqkKCgq0HKl2DB06VBUcHKzKzc1Vpaamqho2bKj66aefyq137949laWlpWrDhg2qwsJC1RdffKFq0qRJnT8v169fVzVr1kwVFRWlgyh1o7Ln5v79+6qbN28Kr44dO6oWLlyoys7O1n7QVKdU9t+oSqVSFRQUqB4/fqxycnJSJSYmqh4/fqwqLS3VcsTiquv5q7LHv3//ftXJkydVxcXFqry8PNXcuXNVMplM7x9CUql0l6dZgD6HUqlUzZw5U2Vra6uqX7++auTIkWpFVEBAgOqTTz5RqVQq1cGDB1UAVFKpVGVubi684uLidBW+6LKzs1UDBw5UmZubqxwcHFRffPGFsMzc3Fx1+PBh4f0vv/yi8vDwUJmamqp8fHxUp0+f1n7AWlLZ8xIVFaUCoPbvw9zcXHXlyhUdRV79qvJv5ml8Cp60pSr/Rp2dnVUA1F6//PKL9oMWUV3PX5U9/u3bt6vc3d1V5ubmKhsbG9Vbb72lOnLkiI6iFpeu8rREpfr/681ERERERFpQZ+8BJSIiIiLdYAFKRERERFrFApSIiIiItIoFKBERERFpFQtQIiIiItIqFqBEREREpFUsQImIiIhIq1iAEhEREZFWsQAlqiZFRUUIDg7G+fPnK1znyJEj8PHxqdJ+fXx8cOTIkVeKLScnB7169cKdO3deaT9EVDtVJn/pA+a6mosFKL2yqKgo+Pj4wMfHBx06dEC/fv0QGxuLkpISXYemUz/88ANcXFzQunXrauvjxo0b8PHxwcWLF6u0naWlJfr27YtVq1ZVU2REVFVluXThwoXlls2ePRs+Pj5YsmRJuWWJiYlo164dli5dWm7Zzp07hfz89KtXr17PjUUb+UsbmOtqLhagJIouXbpgz5492L59O/71r39h/fr12LBhQ7X1V1OK2+LiYo3tKpUK8fHxCAoK0nJElRcYGIg9e/bg0aNHug6FiP5fw4YNsW/fPhQVFQltCoUCv/zyCxo2bKhxm8TERLz33nv46aefUFpaWm65paUl9uzZo/batGlThTHoQ/6qCua6mslI1wFQ7WBsbAw7OzsAQO/evfG///0Phw8fxvvvv4/CwkKsWLECe/fuRV5eHpo3b47w8HB4enoCALKzs7Fo0SKkpKQgNzcXzs7OGDt2LPz9/YX9BwYGon///sjMzMShQ4fQp08fTJkyBYsXL8aBAwfw6NEjNGjQAEOGDMHgwYMBPBkdjI6Oxm+//QYjIyN06dIFERERqF+/PoAnow35+flo1aoVvv/+ewBA//798cEHH1R4nGXbNG/eHFu3boWlpSW2bNlSbr3z58/j+vXr6Ny5s1r7kSNHsHjxYty5cwfe3t7o2rVruW0PHjyI1atXIysrC/b29ujfvz/ee+89GBiU/75Y9gFRdsxt27bF6tWrkZaWhhUrVuDPP/9EaWkp3N3dMXXqVLi5uQnburi4wN7eHocOHULfvn0rPGYi0h4PDw9cvnwZhw8fRvfu3QEA+/btQ6tWrTTmgKtXryIjIwPLly/HkSNHcOzYMbzxxhvl1ivLz5WhKX+dOnUK48aNw7JlyxATE4O//voLr7/+Oj755BOcOHECy5cvR05ODgICAhAREQFDQ0MAEC3/v/POO7h8+TJ++eUX2NjYYNKkSUL+zM3NxcKFC3HixAkUFBTAwcEB48ePF84fc13NxBFQqhYmJibC6GB0dDTS09Px2Wef4fvvv0enTp0QFhYm3JNTWFgIDw8PLFmyBJs3b0bv3r3x8ccfIzMzU22f3377LVq3bo2NGzfin//8JzZt2oTDhw9j4cKF+OGHHzB79mw0aNAAAKBUKjF16lTk5eVhzZo1iImJwYULFzB37ly1fSYnJyM7OxuxsbGYOnUq1q1bh19//fW5x3bixAncuHEDK1euxGeffaZxndOnT8PFxQWmpqZC282bNzFt2jT4+/vju+++Q69evbBy5cpy20VFReGf//wntmzZgoiICGzevBmbN2/W2M8333wDAFi1ahX27NmD6OhoAEBeXh4CAwPx9ddfY82aNbC1tcWUKVPURlUAoHXr1jh9+vRzj5eItCswMBCJiYnC+507dyIwMFDjuomJifD394epqSkCAgLUtntZmvJXmdjYWMyYMQOrV6/GpUuXMG3aNOzZsweff/45PvnkE+zcuRP79u0T1hcr/8fFxaFt27bYuHEj/P39ERkZiYcPHwIAVq5ciczMTCxbtgxbtmzB1KlTIZPJ1LZnrqt5OAJKolKpVEhPT8fu3bsRHByMW7duYefOnfjpp59ga2sLABg9ejSOHj2K3bt3Y8SIEXBwcMCwYcOEfbz33ns4cuQI9u/fj9GjRwvt7du3x9ChQ4X3t27dgpOTE9q0aQOJRILXXntNWHby5ElcvnwZP/74o1CUzp49G6Ghobh69SqcnJwAANbW1pgyZQokEglcXFywZcsWnDp1Ch07dqzwGM3NzTFz5kwYGVX83+fWrVtCv2XK7qmaNGkSgCffys+fP4/4+HhhndjYWIwcORJ9+vQBADRu3BgjR47E5s2bMWTIkHL9WFtbA3hyie3pEY727durrTd79my8+eabOHfuHLy9vYV2Ozu7Kt8/SkTVq0+fPli1ahXu3r2LvLw8XLx4ET169MCuXbvU1istLcWuXbsQGRkJ4MnVp9jYWDx48AA2NjbCejk5OejSpYvatp06ddJ4rymgOX+VCQsLg5eXlxDn+vXrsW/fPlhZWaFZs2bw9fXFqVOn0Lt3b1Hzf5cuXdCvXz8AwPjx4/H999/j3Llz6NSpE27duoWWLVvC3d0dANCoUaNycTPX1TwsQEkUhw4dQpcuXVBaWorS0lIEBARg7Nix+N///ofS0lIhcZQpKipC8+bNATxJomvXrkVSUhLu3r2L4uJiFBUVwdnZWW2bZ2+G79OnD8LCwvDOO++gc+fOeOONN+Dr6wsAyMzMhKOjo1oS9fDwgLGxMTIzM4UCtGnTppBIJMI6tra2ePDgwXOPtXnz5s8tPgGgoKAA9erVU2vLysqCXC5Xa/Py8lIrQC9cuIAzZ84gNjZWaFMqlVAqlc/t71n379/HihUr8Pvvv+PBgwdQKpUoLi7GrVu31NYzMTFBQUFBlfZNRNXLzs4O7du3x65du5Cbm4uuXbvCzMys3Hq//vorlEqlkPdee+01eHh44KefflIr6urXry9cLSkjlUor7F9T/ipTlrcBwMbGBra2trCyslJrK8uhFy9eFC3/N2vWTPizqakpLCwshH4GDBiA6dOn488//0THjh3RtWtXeHh4qG3PXFfzsAAlUbRv3x4RERHCvaBlBVp+fj6MjIzw3XffqRV6wJORRADYsGGDcNmkadOmkEqlmD9/frkHfJ5NmO7u7khMTMSxY8eQnJyMqVOnomfPnpg1a1al4362kJRIJFCpVM/dRtNlqWdZWVkhKyur0nGUefz4McaPH48333yzyts+LSoqCrm5uYiIiICDgwOMjY0xZMiQcuc0NzdXGEUlopojMDAQy5cvR35+PubPn69xnR07duDBgwfo1KmT0KZUKpGbm6tWgEokEjRp0qTSfT8vfz2dMyUSyXNzqJj5X9OX/rJ+unTpgsTERBw9ehQnTpzA6NGjMXr0aIwaNUpYl7mu5mEBSqIwNTXVmOBatGiBkpISPHz4ULhs86wzZ87A398fAQEBAJ484X7t2jW1S+oVsbCwQEBAAAICAtCxY0fMnj0bM2bMgKurK27cuIG7d+8Ko6Dp6ekoLi6Gq6vrKxxp5bRs2RLbtm1Ta3NxccHx48fV2lJTU8ttd+XKlUp/WBgbGwNAuRHSM2fOYMaMGcIHU1ZWlsZv/5mZmVWeh5SIqt8bb7yBTz/9FObm5nj99dfLLX/48CGOHDmCBQsWwMXFRWgvLCzEqFGjkJaWVu6KS2Vpyl8vo7rz/9NsbW0RHByM4OBgrF+/Htu2bVMrQJnrah4WoFStXFxc0KNHD8yePRvh4eFo3rw5srOz8euvv6Jt27Z4/fXX0aRJE/zyyy9ITU2FmZkZ1q9fD4VC8cJ9f/fdd2jQoAFatGgB4MnT405OTjAwMEC7du3QtGlTzJo1C+Hh4SgsLMSnn36KN998U7j8Xp18fHygUCiQlZUlfDgMGDAAcXFxWLZsGYKCgpCamoqff/5ZbbtRo0ZhypQpaNiwofCE559//okbN26oJdMy1tbWMDExwfHjx2FnZ4d69epBJpOhSZMm+Omnn9CqVSvk5OQgJiZGKFbLFBYW4vz585gwYUL1nAQiemlGRkbYvn07JBJJudFDANi1axdsbW3RrVu3cst9fHywY8cOtQL03r175fZR0ZPxmvLXy6jO/P+0VatWoVWrVnBzc8Pjx49x4sQJtUv4zHU1EwtQqnbz5s1DbGwsPv/8c9y9exc2Njbw8vISJkIeNWoUrl+/jg8++ABmZmYICQkp9xCNJlKpFOvXr8dff/0FQ0NDeHp6YtGiRQAAAwMDfP7554iOjsbo0aNhaGiIN954AxEREdV6rGWsrKzg7++PPXv2YNy4cQAAR0dHfPbZZ1iyZAm+//57eHt741//+pfak/SdO3fG4sWLsWbNGqxbtw7GxsZo2rQpQkJCNPZjZGSEiIgIxMbGYsWKFfD29sbq1asxe/ZsfPrppxg6dChee+01TJo0qdwMAEeOHIGDg0OFIxNEpFvPPsn9tJ07d+Ktt97SWJx27doVMTExmDp1KgAI0yM969ixYzAxMSnXril/vazqyv9PMzQ0xLJly3Dz5k2YmprCx8cHH330kbCcua5mkqhedMMbEb2UCxcu4MMPP8T27dsrdd+oto0aNQohISEaP5iIqG6r6fmrKpjraibOA0pUTVq0aIEPPvgAN2/e1HUo5eTk5OCNN9544c/xEVHdVJPzV1Uw19VcHAElIiIiIq3iCCgRERERaRULUCIiIiLSKhagRERERKRVLECJiIiISKtYgBIRERGRVrEAJSIiIiKtYgFKRERERFrFApSIiIiItIoFKBERERFpFQtQIiIiItKq/wPvMpbTOMz0YgAAAABJRU5ErkJggg==", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Extended biological figures saved under: /root/final/baseline/outputs/conditional_flow_matching_method/figures\n" - ] - } - ], - "source": [ - "! pip install -q scikit-learn\n", - "\n", - "from sklearn.decomposition import PCA\n", - "from scipy import sparse as sp_sparse\n", - "from scipy.cluster import hierarchy\n", - "from scipy.spatial.distance import pdist, squareform\n", - "\n", - "\n", - "def _as_dense_rows(X, max_rows: int | None = None, rng: np.random.Generator | None = None) -> np.ndarray:\n", - " if sp.issparse(X):\n", - " arr = X.toarray()\n", - " else:\n", - " arr = np.asarray(X, dtype=np.float64)\n", - " if max_rows is not None and arr.shape[0] > max_rows:\n", - " rng = rng or np.random.default_rng(0)\n", - " idx = rng.choice(arr.shape[0], size=max_rows, replace=False)\n", - " arr = arr[idx]\n", - " return arr\n", - "\n", - "\n", - "def _group_mean_matrix(\n", - " X,\n", - " obs_series: pd.Series,\n", - " conditions: list[str],\n", - " max_conditions: int = 48,\n", - ") -> tuple[np.ndarray, list[str]]:\n", - " rows = []\n", - " labels: list[str] = []\n", - " for c in conditions[:max_conditions]:\n", - " m = obs_series.astype(str).values == c\n", - " if m.sum() < 1:\n", - " continue\n", - " block = X[m]\n", - " if sp.issparse(block):\n", - " v = np.asarray(block.mean(axis=0)).ravel()\n", - " else:\n", - " v = np.asarray(block, dtype=np.float64).mean(axis=0)\n", - " rows.append(v)\n", - " labels.append(c)\n", - " if not rows:\n", - " return np.zeros((0, 0)), []\n", - " return np.stack(rows, axis=0), labels\n", - "\n", - "\n", - "_bio_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", - "_rng_bio = np.random.default_rng(RANDOM_SEED + 8800)\n", - "setup_science_style()\n", - "\n", - "gene_names = np.array(adata.var_names.astype(str))\n", - "ctrl = np.asarray(control_mean, dtype=np.float64).ravel()\n", - "\n", - "real_mean_gene = np.asarray(real.X.mean(axis=0)).ravel()\n", - "pred_mean_gene = np.asarray(pred.X.mean(axis=0)).ravel()\n", - "\n", - "# --- Fig A: global gene-mean calibration (log1p) ---\n", - "fig, ax = plt.subplots(figsize=(3.2, 3.2), constrained_layout=True)\n", - "xr = np.log1p(real_mean_gene)\n", - "xp = np.log1p(pred_mean_gene)\n", - "ax.scatter(xr, xp, s=4, alpha=0.35, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\", rasterized=True)\n", - "lim = [float(np.min([xr.min(), xp.min()])), float(np.max([xr.max(), xp.max()]))]\n", - "ax.plot(lim, lim, ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.9)\n", - "ax.set_xlabel(\"log1p(mean expression), held-out truth\")\n", - "ax.set_ylabel(\"log1p(mean expression), predicted\")\n", - "ax.set_title(\"Gene-wise mean calibration\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_A_gene_mean_calibration_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_A_gene_mean_calibration_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - "plt.show()\n", - "\n", - "# --- Fig B: MA-style (mean abundance vs log2 ratio) ---\n", - "eps = 1e-6\n", - "M = 0.5 * (real_mean_gene + pred_mean_gene)\n", - "A = np.log2((pred_mean_gene + eps) / (real_mean_gene + eps))\n", - "ok = np.isfinite(A) & np.isfinite(M)\n", - "fig, ax = plt.subplots(figsize=(3.4, 3.0), constrained_layout=True)\n", - "ax.scatter(M[ok], A[ok], s=3, alpha=0.3, c=SCIENCE_COLORS[\"orange\"], edgecolors=\"none\", rasterized=True)\n", - "ax.axhline(0.0, color=SCIENCE_COLORS[\"gray\"], lw=0.8)\n", - "ax.set_xlabel(\"Mean abundance (avg pred & truth)\")\n", - "ax.set_ylabel(\"log2(pred / truth)\")\n", - "ax.set_title(\"MA-style bias vs abundance\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_B_MA_plot_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_B_MA_plot_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - "plt.show()\n", - "\n", - "# --- Fig C: per-gene mean error distribution ---\n", - "ge = pred_mean_gene - real_mean_gene\n", - "fig, ax = plt.subplots(figsize=(3.2, 2.6), constrained_layout=True)\n", - "ax.hist(ge, bins=80, color=SCIENCE_COLORS[\"green\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", - "ax.axvline(0.0, color=SCIENCE_COLORS[\"dark\"], lw=0.9)\n", - "ax.set_xlabel(\"Pred − truth (gene mean)\")\n", - "ax.set_ylabel(\"Genes\")\n", - "ax.set_title(\"Global gene-level error\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_C_gene_error_hist_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_C_gene_error_hist_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - "plt.show()\n", - "\n", - "# --- Fig D–E: perturbation × gene heatmaps (delta vs control, z-scored genes) ---\n", - "R_mat, perts = _group_mean_matrix(real.X, real.obs[\"perturbation_gene\"], eval_conditions_to_run, max_conditions=40)\n", - "P_mat, perts_p = _group_mean_matrix(pred.X, pred.obs[\"perturbation_gene\"], eval_conditions_to_run, max_conditions=40)\n", - "if R_mat.size and P_mat.size and R_mat.shape == P_mat.shape:\n", - " d_real = R_mat - ctrl\n", - " d_pred = P_mat - ctrl\n", - " var_g = np.var(d_real, axis=0)\n", - " top_k = min(60, d_real.shape[1])\n", - " gi = np.argsort(var_g)[-top_k:]\n", - " Z = lambda D: (D[:, gi] - D[:, gi].mean(axis=1, keepdims=True)) / (D[:, gi].std(axis=1, keepdims=True) + 1e-8)\n", - " Zr, Zp = Z(d_real), Z(d_pred)\n", - " for name, Zm in ((\"truth\", Zr), (\"predicted\", Zp)):\n", - " fig, ax = plt.subplots(figsize=(5.5, 4.2), constrained_layout=True)\n", - " im = ax.imshow(Zm, aspect=\"auto\", cmap=\"RdBu_r\", vmin=-2.5, vmax=2.5, interpolation=\"nearest\")\n", - " ax.set_yticks(np.arange(len(perts)))\n", - " ax.set_yticklabels(perts, fontsize=5)\n", - " ax.set_xlabel(\"Genes (high-variance subset, z-scored per perturbation)\")\n", - " ax.set_ylabel(\"Perturbation\")\n", - " ax.set_title(f\"Perturbation effects ({name})\")\n", - " plt.colorbar(im, ax=ax, fraction=0.035, pad=0.02, label=\"Row z-score\")\n", - " fig.savefig(FIGURE_DIR / f\"bio_DE_heatmap_delta_{name}_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - " fig.savefig(FIGURE_DIR / f\"bio_DE_heatmap_delta_{name}_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - " plt.show()\n", - "\n", - "# --- Fig F: perturbation–perturbation correlation of delta profiles (truth) ---\n", - "if R_mat.shape[0] >= 3:\n", - " d_real = R_mat - ctrl\n", - " C = np.corrcoef(d_real)\n", - " fig, ax = plt.subplots(figsize=(4.8, 4.2), constrained_layout=True)\n", - " im = ax.imshow(C, cmap=\"RdBu_r\", vmin=-1, vmax=1, aspect=\"auto\")\n", - " ax.set_xticks(np.arange(len(perts)))\n", - " ax.set_xticklabels(perts, rotation=90, fontsize=4.5)\n", - " ax.set_yticks(np.arange(len(perts)))\n", - " ax.set_yticklabels(perts, fontsize=4.5)\n", - " ax.set_title(\"Perturbation similarity (Pearson of Δ vs control)\")\n", - " plt.colorbar(im, ax=ax, fraction=0.046, pad=0.04)\n", - " fig.savefig(FIGURE_DIR / f\"bio_F_perturb_corr_truth_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - " fig.savefig(FIGURE_DIR / f\"bio_F_perturb_corr_truth_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - " plt.show()\n", - "\n", - "# --- Fig G: hierarchical clustering on subset of perturbations (truth delta) ---\n", - "if R_mat.shape[0] >= 4:\n", - " d_real = R_mat - ctrl\n", - " sub = min(24, d_real.shape[0])\n", - " cond_dist = pdist(d_real[:sub], metric=\"correlation\")\n", - " y = hierarchy.linkage(cond_dist, method=\"average\")\n", - " order = hierarchy.leaves_list(y)\n", - " labels_sub = [perts[i] for i in order]\n", - " Csub = np.corrcoef(d_real[:sub][order])\n", - " fig, ax = plt.subplots(figsize=(4.5, 4.0), constrained_layout=True)\n", - " im = ax.imshow(Csub, cmap=\"RdBu_r\", vmin=-1, vmax=1)\n", - " ax.set_xticks(np.arange(sub))\n", - " ax.set_xticklabels(labels_sub, rotation=90, fontsize=5)\n", - " ax.set_yticks(np.arange(sub))\n", - " ax.set_yticklabels(labels_sub, fontsize=5)\n", - " ax.set_title(\"Clustered perturbation similarity (subset)\")\n", - " plt.colorbar(im, ax=ax, fraction=0.046, pad=0.04)\n", - " fig.savefig(FIGURE_DIR / f\"bio_G_perturb_corr_clustered_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - " fig.savefig(FIGURE_DIR / f\"bio_G_perturb_corr_clustered_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - " plt.show()\n", - "\n", - "# --- Fig H: cell-level PCA (pred vs real, subsampled) ---\n", - "n_sub = min(6000, real.n_obs)\n", - "idx = _rng_bio.choice(real.n_obs, size=n_sub, replace=False) if real.n_obs > n_sub else np.arange(real.n_obs)\n", - "Xr = _as_dense_rows(real.X[idx], max_rows=None)\n", - "Xp = _as_dense_rows(pred.X[idx], max_rows=None)\n", - "lab = real.obs[\"perturbation_gene\"].astype(str).iloc[idx].to_numpy()\n", - "uniq = np.unique(lab)\n", - "if len(uniq) > 24:\n", - " keep = set(_rng_bio.choice(uniq, size=24, replace=False))\n", - " mask = np.array([g in keep for g in lab])\n", - " idx = idx[mask]\n", - " Xr, Xp, lab = Xr[mask], Xp[mask], lab[mask]\n", - "Z = np.vstack([Xr, Xp])\n", - "pca = PCA(n_components=2, random_state=RANDOM_SEED)\n", - "Z2 = pca.fit_transform(Z)\n", - "nr = int(Xr.shape[0])\n", - "Z_truth = Z2[:nr]\n", - "Z_pred = Z2[nr:]\n", - "fig, axes = plt.subplots(1, 2, figsize=(6.8, 3.0), constrained_layout=True)\n", - "for ax, name, Zpart in zip(axes, [\"truth\", \"pred\"], [Z_truth, Z_pred]):\n", - " for j, g in enumerate(np.unique(lab)):\n", - " m = lab == g\n", - " if not m.any():\n", - " continue\n", - " ax.scatter(Zpart[m, 0], Zpart[m, 1], s=3, alpha=0.45, label=g if j < 12 else None)\n", - " ax.set_title(f\"PCA — {name}\")\n", - " ax.set_xlabel(\"PC1\")\n", - " ax.set_ylabel(\"PC2\")\n", - "axes[0].legend(bbox_to_anchor=(1.02, 1), loc=\"upper left\", fontsize=5, markerscale=2)\n", - "fig.suptitle(\"Expression PCA (subsampled cells)\", y=1.02)\n", - "fig.savefig(FIGURE_DIR / f\"bio_H_cell_PCA_truth_pred_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_H_cell_PCA_truth_pred_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - "plt.show()\n", - "\n", - "# --- Fig I: per-cell L2 error distribution ---\n", - "diff = _as_dense_rows(pred.X[idx], max_rows=None) - _as_dense_rows(real.X[idx], max_rows=None)\n", - "l2 = np.linalg.norm(diff, axis=1)\n", - "fig, ax = plt.subplots(figsize=(3.2, 2.6), constrained_layout=True)\n", - "ax.hist(l2, bins=60, color=SCIENCE_COLORS[\"magenta\"], alpha=0.88, edgecolor=\"white\", linewidth=0.3)\n", - "ax.set_xlabel(\"L2(pred − truth) per cell\")\n", - "ax.set_ylabel(\"Cells\")\n", - "ax.set_title(\"Cell-level reconstruction error\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_I_cell_L2_error_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_I_cell_L2_error_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - "plt.show()\n", - "\n", - "# --- Fig J: metrics overview (ridge / strip) ---\n", - "if len(metrics) > 0:\n", - " fig, axes = plt.subplots(2, 2, figsize=(6.0, 4.8), constrained_layout=True)\n", - " axes[0, 0].scatter(metrics[\"mse_mean\"], metrics[\"pearson_mean\"], s=12, alpha=0.65, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\")\n", - " axes[0, 0].set_xlabel(\"MSE (means)\")\n", - " axes[0, 0].set_ylabel(\"Pearson r\")\n", - " axes[0, 1].scatter(metrics[\"n_cells\"], metrics[\"delta_l2\"], s=12, alpha=0.65, c=SCIENCE_COLORS[\"green\"], edgecolors=\"none\")\n", - " axes[0, 1].set_xlabel(\"Cells per perturbation\")\n", - " axes[0, 1].set_ylabel(\"Delta L2\")\n", - " axes[1, 0].hist(metrics[\"pearson_delta\"].dropna(), bins=40, color=SCIENCE_COLORS[\"orange\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", - " axes[1, 0].set_xlabel(\"Pearson r (delta)\")\n", - " axes[1, 0].set_ylabel(\"Perturbations\")\n", - " axes[1, 1].hist(metrics[\"mae_mean\"], bins=40, color=SCIENCE_COLORS[\"blue\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", - " axes[1, 1].set_xlabel(\"MAE (means)\")\n", - " axes[1, 1].set_ylabel(\"Perturbations\")\n", - " fig.suptitle(\"Perturbation-level metric panels\", y=1.02)\n", - " fig.savefig(FIGURE_DIR / f\"bio_J_metric_panels_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - " fig.savefig(FIGURE_DIR / f\"bio_J_metric_panels_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - " plt.show()\n", - "\n", - "# --- Fig K: high cell-wise variance genes — pred vs truth scatter ---\n", - "if sp.issparse(real.X):\n", - " mu = np.asarray(real.X.mean(axis=0)).ravel()\n", - " mu2 = np.asarray(real.X.power(2).mean(axis=0)).ravel()\n", - " var_per_gene = mu2 - mu**2\n", - "else:\n", - " var_per_gene = np.var(np.asarray(real.X, dtype=np.float64), axis=0)\n", - "top_idx = np.argsort(var_per_gene)[-12:][::-1]\n", - "fig, axes = plt.subplots(3, 4, figsize=(7.2, 5.4), constrained_layout=True)\n", - "axes = axes.ravel()\n", - "for ax, gi in zip(axes, top_idx):\n", - " if sp.issparse(real.X):\n", - " vr = real.X[:, gi].toarray().ravel()\n", - " else:\n", - " vr = np.asarray(real.X[:, gi], dtype=np.float64).ravel()\n", - " if sp.issparse(pred.X):\n", - " vp = pred.X[:, gi].toarray().ravel()\n", - " else:\n", - " vp = np.asarray(pred.X[:, gi], dtype=np.float64).ravel()\n", - " if vr.size > 8000:\n", - " si = _rng_bio.choice(vr.size, size=8000, replace=False)\n", - " vr, vp = vr[si], vp[si]\n", - " ax.scatter(vr, vp, s=2, alpha=0.25, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\", rasterized=True)\n", - " lim = [min(vr.min(), vp.min()), max(vr.max(), vp.max())]\n", - " ax.plot(lim, lim, ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.7)\n", - " ax.set_title(gene_names[gi], fontsize=7)\n", - " ax.set_xlabel(\"Truth\", fontsize=6)\n", - " ax.set_ylabel(\"Pred\", fontsize=6)\n", - "fig.suptitle(\"High-variance genes: cell-level pred vs truth\", y=1.01)\n", - "fig.savefig(FIGURE_DIR / f\"bio_K_topvar_gene_scatters_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", - "fig.savefig(FIGURE_DIR / f\"bio_K_topvar_gene_scatters_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", - "plt.show()\n", - "\n", - "print(\"Extended biological figures saved under:\", FIGURE_DIR)\n" + "data": { + "image/png": "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", + "text/plain": [ + "
" ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" + }, + "metadata": {}, + "output_type": "display_data" }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.12.3" + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Extended biological figures saved under: /root/final/baseline/outputs/conditional_flow_matching_method/figures\n" + ] } + ], + "source": [ + "! pip install -q scikit-learn\n", + "\n", + "from sklearn.decomposition import PCA\n", + "from scipy import sparse as sp_sparse\n", + "from scipy.cluster import hierarchy\n", + "from scipy.spatial.distance import pdist, squareform\n", + "\n", + "\n", + "def _as_dense_rows(X, max_rows: int | None = None, rng: np.random.Generator | None = None) -> np.ndarray:\n", + " if sp.issparse(X):\n", + " arr = X.toarray()\n", + " else:\n", + " arr = np.asarray(X, dtype=np.float64)\n", + " if max_rows is not None and arr.shape[0] > max_rows:\n", + " rng = rng or np.random.default_rng(0)\n", + " idx = rng.choice(arr.shape[0], size=max_rows, replace=False)\n", + " arr = arr[idx]\n", + " return arr\n", + "\n", + "\n", + "def _group_mean_matrix(\n", + " X,\n", + " obs_series: pd.Series,\n", + " conditions: list[str],\n", + " max_conditions: int = 48,\n", + ") -> tuple[np.ndarray, list[str]]:\n", + " rows = []\n", + " labels: list[str] = []\n", + " for c in conditions[:max_conditions]:\n", + " m = obs_series.astype(str).values == c\n", + " if m.sum() < 1:\n", + " continue\n", + " block = X[m]\n", + " if sp.issparse(block):\n", + " v = np.asarray(block.mean(axis=0)).ravel()\n", + " else:\n", + " v = np.asarray(block, dtype=np.float64).mean(axis=0)\n", + " rows.append(v)\n", + " labels.append(c)\n", + " if not rows:\n", + " return np.zeros((0, 0)), []\n", + " return np.stack(rows, axis=0), labels\n", + "\n", + "\n", + "_bio_tag = f\"{SPLIT_STRATEGY}_{EVAL_SPLIT}\"\n", + "_rng_bio = np.random.default_rng(RANDOM_SEED + 8800)\n", + "setup_science_style()\n", + "\n", + "gene_names = np.array(adata.var_names.astype(str))\n", + "ctrl = np.asarray(control_mean, dtype=np.float64).ravel()\n", + "\n", + "real_mean_gene = np.asarray(real.X.mean(axis=0)).ravel()\n", + "pred_mean_gene = np.asarray(pred.X.mean(axis=0)).ravel()\n", + "\n", + "# --- Fig A: global gene-mean calibration (log1p) ---\n", + "fig, ax = plt.subplots(figsize=(3.2, 3.2), constrained_layout=True)\n", + "xr = np.log1p(real_mean_gene)\n", + "xp = np.log1p(pred_mean_gene)\n", + "ax.scatter(xr, xp, s=4, alpha=0.35, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\", rasterized=True)\n", + "lim = [float(np.min([xr.min(), xp.min()])), float(np.max([xr.max(), xp.max()]))]\n", + "ax.plot(lim, lim, ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.9)\n", + "ax.set_xlabel(\"log1p(mean expression), held-out truth\")\n", + "ax.set_ylabel(\"log1p(mean expression), predicted\")\n", + "ax.set_title(\"Gene-wise mean calibration\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_A_gene_mean_calibration_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_A_gene_mean_calibration_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig B: MA-style (mean abundance vs log2 ratio) ---\n", + "eps = 1e-6\n", + "M = 0.5 * (real_mean_gene + pred_mean_gene)\n", + "A = np.log2((pred_mean_gene + eps) / (real_mean_gene + eps))\n", + "ok = np.isfinite(A) & np.isfinite(M)\n", + "fig, ax = plt.subplots(figsize=(3.4, 3.0), constrained_layout=True)\n", + "ax.scatter(M[ok], A[ok], s=3, alpha=0.3, c=SCIENCE_COLORS[\"orange\"], edgecolors=\"none\", rasterized=True)\n", + "ax.axhline(0.0, color=SCIENCE_COLORS[\"gray\"], lw=0.8)\n", + "ax.set_xlabel(\"Mean abundance (avg pred & truth)\")\n", + "ax.set_ylabel(\"log2(pred / truth)\")\n", + "ax.set_title(\"MA-style bias vs abundance\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_B_MA_plot_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_B_MA_plot_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig C: per-gene mean error distribution ---\n", + "ge = pred_mean_gene - real_mean_gene\n", + "fig, ax = plt.subplots(figsize=(3.2, 2.6), constrained_layout=True)\n", + "ax.hist(ge, bins=80, color=SCIENCE_COLORS[\"green\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", + "ax.axvline(0.0, color=SCIENCE_COLORS[\"dark\"], lw=0.9)\n", + "ax.set_xlabel(\"Pred − truth (gene mean)\")\n", + "ax.set_ylabel(\"Genes\")\n", + "ax.set_title(\"Global gene-level error\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_C_gene_error_hist_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_C_gene_error_hist_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig D–E: perturbation × gene heatmaps (delta vs control, z-scored genes) ---\n", + "R_mat, perts = _group_mean_matrix(real.X, real.obs[\"perturbation_gene\"], eval_conditions_to_run, max_conditions=40)\n", + "P_mat, perts_p = _group_mean_matrix(pred.X, pred.obs[\"perturbation_gene\"], eval_conditions_to_run, max_conditions=40)\n", + "if R_mat.size and P_mat.size and R_mat.shape == P_mat.shape:\n", + " d_real = R_mat - ctrl\n", + " d_pred = P_mat - ctrl\n", + " var_g = np.var(d_real, axis=0)\n", + " top_k = min(60, d_real.shape[1])\n", + " gi = np.argsort(var_g)[-top_k:]\n", + " Z = lambda D: (D[:, gi] - D[:, gi].mean(axis=1, keepdims=True)) / (D[:, gi].std(axis=1, keepdims=True) + 1e-8)\n", + " Zr, Zp = Z(d_real), Z(d_pred)\n", + " for name, Zm in ((\"truth\", Zr), (\"predicted\", Zp)):\n", + " fig, ax = plt.subplots(figsize=(5.5, 4.2), constrained_layout=True)\n", + " im = ax.imshow(Zm, aspect=\"auto\", cmap=\"RdBu_r\", vmin=-2.5, vmax=2.5, interpolation=\"nearest\")\n", + " ax.set_yticks(np.arange(len(perts)))\n", + " ax.set_yticklabels(perts, fontsize=5)\n", + " ax.set_xlabel(\"Genes (high-variance subset, z-scored per perturbation)\")\n", + " ax.set_ylabel(\"Perturbation\")\n", + " ax.set_title(f\"Perturbation effects ({name})\")\n", + " plt.colorbar(im, ax=ax, fraction=0.035, pad=0.02, label=\"Row z-score\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_DE_heatmap_delta_{name}_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_DE_heatmap_delta_{name}_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig F: perturbation–perturbation correlation of delta profiles (truth) ---\n", + "if R_mat.shape[0] >= 3:\n", + " d_real = R_mat - ctrl\n", + " C = np.corrcoef(d_real)\n", + " fig, ax = plt.subplots(figsize=(4.8, 4.2), constrained_layout=True)\n", + " im = ax.imshow(C, cmap=\"RdBu_r\", vmin=-1, vmax=1, aspect=\"auto\")\n", + " ax.set_xticks(np.arange(len(perts)))\n", + " ax.set_xticklabels(perts, rotation=90, fontsize=4.5)\n", + " ax.set_yticks(np.arange(len(perts)))\n", + " ax.set_yticklabels(perts, fontsize=4.5)\n", + " ax.set_title(\"Perturbation similarity (Pearson of Δ vs control)\")\n", + " plt.colorbar(im, ax=ax, fraction=0.046, pad=0.04)\n", + " fig.savefig(FIGURE_DIR / f\"bio_F_perturb_corr_truth_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_F_perturb_corr_truth_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig G: hierarchical clustering on subset of perturbations (truth delta) ---\n", + "if R_mat.shape[0] >= 4:\n", + " d_real = R_mat - ctrl\n", + " sub = min(24, d_real.shape[0])\n", + " cond_dist = pdist(d_real[:sub], metric=\"correlation\")\n", + " y = hierarchy.linkage(cond_dist, method=\"average\")\n", + " order = hierarchy.leaves_list(y)\n", + " labels_sub = [perts[i] for i in order]\n", + " Csub = np.corrcoef(d_real[:sub][order])\n", + " fig, ax = plt.subplots(figsize=(4.5, 4.0), constrained_layout=True)\n", + " im = ax.imshow(Csub, cmap=\"RdBu_r\", vmin=-1, vmax=1)\n", + " ax.set_xticks(np.arange(sub))\n", + " ax.set_xticklabels(labels_sub, rotation=90, fontsize=5)\n", + " ax.set_yticks(np.arange(sub))\n", + " ax.set_yticklabels(labels_sub, fontsize=5)\n", + " ax.set_title(\"Clustered perturbation similarity (subset)\")\n", + " plt.colorbar(im, ax=ax, fraction=0.046, pad=0.04)\n", + " fig.savefig(FIGURE_DIR / f\"bio_G_perturb_corr_clustered_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_G_perturb_corr_clustered_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig H: cell-level PCA (pred vs real, subsampled) ---\n", + "n_sub = min(6000, real.n_obs)\n", + "idx = _rng_bio.choice(real.n_obs, size=n_sub, replace=False) if real.n_obs > n_sub else np.arange(real.n_obs)\n", + "Xr = _as_dense_rows(real.X[idx], max_rows=None)\n", + "Xp = _as_dense_rows(pred.X[idx], max_rows=None)\n", + "lab = real.obs[\"perturbation_gene\"].astype(str).iloc[idx].to_numpy()\n", + "uniq = np.unique(lab)\n", + "if len(uniq) > 24:\n", + " keep = set(_rng_bio.choice(uniq, size=24, replace=False))\n", + " mask = np.array([g in keep for g in lab])\n", + " idx = idx[mask]\n", + " Xr, Xp, lab = Xr[mask], Xp[mask], lab[mask]\n", + "Z = np.vstack([Xr, Xp])\n", + "pca = PCA(n_components=2, random_state=RANDOM_SEED)\n", + "Z2 = pca.fit_transform(Z)\n", + "nr = int(Xr.shape[0])\n", + "Z_truth = Z2[:nr]\n", + "Z_pred = Z2[nr:]\n", + "fig, axes = plt.subplots(1, 2, figsize=(6.8, 3.0), constrained_layout=True)\n", + "for ax, name, Zpart in zip(axes, [\"truth\", \"pred\"], [Z_truth, Z_pred]):\n", + " for j, g in enumerate(np.unique(lab)):\n", + " m = lab == g\n", + " if not m.any():\n", + " continue\n", + " ax.scatter(Zpart[m, 0], Zpart[m, 1], s=3, alpha=0.45, label=g if j < 12 else None)\n", + " ax.set_title(f\"PCA — {name}\")\n", + " ax.set_xlabel(\"PC1\")\n", + " ax.set_ylabel(\"PC2\")\n", + "axes[0].legend(bbox_to_anchor=(1.02, 1), loc=\"upper left\", fontsize=5, markerscale=2)\n", + "fig.suptitle(\"Expression PCA (subsampled cells)\", y=1.02)\n", + "fig.savefig(FIGURE_DIR / f\"bio_H_cell_PCA_truth_pred_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_H_cell_PCA_truth_pred_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig I: per-cell L2 error distribution ---\n", + "diff = _as_dense_rows(pred.X[idx], max_rows=None) - _as_dense_rows(real.X[idx], max_rows=None)\n", + "l2 = np.linalg.norm(diff, axis=1)\n", + "fig, ax = plt.subplots(figsize=(3.2, 2.6), constrained_layout=True)\n", + "ax.hist(l2, bins=60, color=SCIENCE_COLORS[\"magenta\"], alpha=0.88, edgecolor=\"white\", linewidth=0.3)\n", + "ax.set_xlabel(\"L2(pred − truth) per cell\")\n", + "ax.set_ylabel(\"Cells\")\n", + "ax.set_title(\"Cell-level reconstruction error\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_I_cell_L2_error_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_I_cell_L2_error_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "# --- Fig J: metrics overview (ridge / strip) ---\n", + "if len(metrics) > 0:\n", + " fig, axes = plt.subplots(2, 2, figsize=(6.0, 4.8), constrained_layout=True)\n", + " axes[0, 0].scatter(metrics[\"mse_mean\"], metrics[\"pearson_mean\"], s=12, alpha=0.65, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\")\n", + " axes[0, 0].set_xlabel(\"MSE (means)\")\n", + " axes[0, 0].set_ylabel(\"Pearson r\")\n", + " axes[0, 1].scatter(metrics[\"n_cells\"], metrics[\"delta_l2\"], s=12, alpha=0.65, c=SCIENCE_COLORS[\"green\"], edgecolors=\"none\")\n", + " axes[0, 1].set_xlabel(\"Cells per perturbation\")\n", + " axes[0, 1].set_ylabel(\"Delta L2\")\n", + " axes[1, 0].hist(metrics[\"pearson_delta\"].dropna(), bins=40, color=SCIENCE_COLORS[\"orange\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", + " axes[1, 0].set_xlabel(\"Pearson r (delta)\")\n", + " axes[1, 0].set_ylabel(\"Perturbations\")\n", + " axes[1, 1].hist(metrics[\"mae_mean\"], bins=40, color=SCIENCE_COLORS[\"blue\"], alpha=0.85, edgecolor=\"white\", linewidth=0.3)\n", + " axes[1, 1].set_xlabel(\"MAE (means)\")\n", + " axes[1, 1].set_ylabel(\"Perturbations\")\n", + " fig.suptitle(\"Perturbation-level metric panels\", y=1.02)\n", + " fig.savefig(FIGURE_DIR / f\"bio_J_metric_panels_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + " fig.savefig(FIGURE_DIR / f\"bio_J_metric_panels_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + " plt.show()\n", + "\n", + "# --- Fig K: high cell-wise variance genes — pred vs truth scatter ---\n", + "if sp.issparse(real.X):\n", + " mu = np.asarray(real.X.mean(axis=0)).ravel()\n", + " mu2 = np.asarray(real.X.power(2).mean(axis=0)).ravel()\n", + " var_per_gene = mu2 - mu**2\n", + "else:\n", + " var_per_gene = np.var(np.asarray(real.X, dtype=np.float64), axis=0)\n", + "top_idx = np.argsort(var_per_gene)[-12:][::-1]\n", + "fig, axes = plt.subplots(3, 4, figsize=(7.2, 5.4), constrained_layout=True)\n", + "axes = axes.ravel()\n", + "for ax, gi in zip(axes, top_idx):\n", + " if sp.issparse(real.X):\n", + " vr = real.X[:, gi].toarray().ravel()\n", + " else:\n", + " vr = np.asarray(real.X[:, gi], dtype=np.float64).ravel()\n", + " if sp.issparse(pred.X):\n", + " vp = pred.X[:, gi].toarray().ravel()\n", + " else:\n", + " vp = np.asarray(pred.X[:, gi], dtype=np.float64).ravel()\n", + " if vr.size > 8000:\n", + " si = _rng_bio.choice(vr.size, size=8000, replace=False)\n", + " vr, vp = vr[si], vp[si]\n", + " ax.scatter(vr, vp, s=2, alpha=0.25, c=SCIENCE_COLORS[\"blue\"], edgecolors=\"none\", rasterized=True)\n", + " lim = [min(vr.min(), vp.min()), max(vr.max(), vp.max())]\n", + " ax.plot(lim, lim, ls=\"--\", color=SCIENCE_COLORS[\"gray\"], lw=0.7)\n", + " ax.set_title(gene_names[gi], fontsize=7)\n", + " ax.set_xlabel(\"Truth\", fontsize=6)\n", + " ax.set_ylabel(\"Pred\", fontsize=6)\n", + "fig.suptitle(\"High-variance genes: cell-level pred vs truth\", y=1.01)\n", + "fig.savefig(FIGURE_DIR / f\"bio_K_topvar_gene_scatters_{_bio_tag}.pdf\", bbox_inches=\"tight\", facecolor=\"white\")\n", + "fig.savefig(FIGURE_DIR / f\"bio_K_topvar_gene_scatters_{_bio_tag}.png\", bbox_inches=\"tight\", facecolor=\"white\", dpi=300)\n", + "plt.show()\n", + "\n", + "print(\"Extended biological figures saved under:\", FIGURE_DIR)\n" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" }, - "nbformat": 4, - "nbformat_minor": 5 + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.12.13" + } + }, + "nbformat": 4, + "nbformat_minor": 5 }