Upload ptm_data_preprocessing.ipynb

ptm_data_preprocessing.ipynb

@@ -0,0 +1,633 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 47,
+   "id": "945a82aa-1398-422a-98df-b3db93973271",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Entry</th>\n",
+       "      <th>Protein families</th>\n",
+       "      <th>Modified residue</th>\n",
+       "      <th>Sequence</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>A0A009GHC8</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>A0A009HTZ2</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>A0A009IVE2</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>A0A009MYL5</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>A0A009PHM9</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        Entry                                   Protein families  \\\n",
+       "0  A0A009GHC8  Precorrin methyltransferase family; Precorrin ...   \n",
+       "1  A0A009HTZ2  Precorrin methyltransferase family; Precorrin ...   \n",
+       "2  A0A009IVE2  Precorrin methyltransferase family; Precorrin ...   \n",
+       "3  A0A009MYL5  Precorrin methyltransferase family; Precorrin ...   \n",
+       "4  A0A009PHM9  Precorrin methyltransferase family; Precorrin ...   \n",
+       "\n",
+       "                                    Modified residue  \\\n",
+       "0  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "1  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "2  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "3  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "4  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "\n",
+       "                                            Sequence  \n",
+       "0  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...  \n",
+       "1  MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...  \n",
+       "2  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...  \n",
+       "3  MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...  \n",
+       "4  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...  "
+      ]
+     },
+     "execution_count": 47,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "# Load the TSV file\n",
+    "file_path = 'PTM/uniprotkb_family_AND_ft_mod_res_AND_pro_2023_10_02.tsv'\n",
+    "data = pd.read_csv(file_path, sep='\\t')\n",
+    "\n",
+    "# Display the first few rows of the data\n",
+    "data.head()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 48,
+   "id": "a03f8ff8-0612-4f8c-bccd-49fde3dce0f5",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Entry</th>\n",
+       "      <th>Protein families</th>\n",
+       "      <th>Modified residue</th>\n",
+       "      <th>Sequence</th>\n",
+       "      <th>PTM sites</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>A0A009GHC8</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>A0A009HTZ2</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>A0A009IVE2</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>A0A009MYL5</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>A0A009PHM9</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        Entry                                   Protein families  \\\n",
+       "0  A0A009GHC8  Precorrin methyltransferase family; Precorrin ...   \n",
+       "1  A0A009HTZ2  Precorrin methyltransferase family; Precorrin ...   \n",
+       "2  A0A009IVE2  Precorrin methyltransferase family; Precorrin ...   \n",
+       "3  A0A009MYL5  Precorrin methyltransferase family; Precorrin ...   \n",
+       "4  A0A009PHM9  Precorrin methyltransferase family; Precorrin ...   \n",
+       "\n",
+       "                                    Modified residue  \\\n",
+       "0  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "1  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "2  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "3  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "4  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "\n",
+       "                                            Sequence  \\\n",
+       "0  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...   \n",
+       "1  MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...   \n",
+       "2  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...   \n",
+       "3  MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...   \n",
+       "4  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...   \n",
+       "\n",
+       "                                           PTM sites  \n",
+       "0  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "1  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "2  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "3  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "4  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  "
+      ]
+     },
+     "execution_count": 48,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import re\n",
+    "\n",
+    "def get_ptm_sites(row):\n",
+    "    # Extract the positions of modified residues from the 'Modified residue' column\n",
+    "    modified_positions = [int(i) for i in re.findall(r'MOD_RES (\\d+)', row['Modified residue'])]\n",
+    "    \n",
+    "    # Create a list of zeros of length equal to the protein sequence\n",
+    "    ptm_sites = [0] * len(row['Sequence'])\n",
+    "    \n",
+    "    # Replace the zeros with ones at the positions of modified residues\n",
+    "    for position in modified_positions:\n",
+    "        # Subtracting 1 because positions are 1-indexed, but lists are 0-indexed\n",
+    "        ptm_sites[position - 1] = 1\n",
+    "    \n",
+    "    return ptm_sites\n",
+    "\n",
+    "# Apply the function to each row in the DataFrame\n",
+    "data['PTM sites'] = data.apply(get_ptm_sites, axis=1)\n",
+    "\n",
+    "# Display the first few rows of the updated DataFrame\n",
+    "data.head()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 50,
+   "id": "5d2e5043-e2f9-44ec-899b-7dad4f83f823",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "<div>\n",
+       "<style scoped>\n",
+       "    .dataframe tbody tr th:only-of-type {\n",
+       "        vertical-align: middle;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe tbody tr th {\n",
+       "        vertical-align: top;\n",
+       "    }\n",
+       "\n",
+       "    .dataframe thead th {\n",
+       "        text-align: right;\n",
+       "    }\n",
+       "</style>\n",
+       "<table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       "    <tr style=\"text-align: right;\">\n",
+       "      <th></th>\n",
+       "      <th>Entry</th>\n",
+       "      <th>Protein families</th>\n",
+       "      <th>Modified residue</th>\n",
+       "      <th>Sequence</th>\n",
+       "      <th>PTM sites</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <th>0</th>\n",
+       "      <td>A0A009GHC8</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>1</th>\n",
+       "      <td>A0A009HTZ2</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>2</th>\n",
+       "      <td>A0A009IVE2</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>3</th>\n",
+       "      <td>A0A009MYL5</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <th>4</th>\n",
+       "      <td>A0A009PHM9</td>\n",
+       "      <td>Precorrin methyltransferase family; Precorrin ...</td>\n",
+       "      <td>MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...</td>\n",
+       "      <td>MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...</td>\n",
+       "      <td>[0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table>\n",
+       "</div>"
+      ],
+      "text/plain": [
+       "        Entry                                   Protein families  \\\n",
+       "0  A0A009GHC8  Precorrin methyltransferase family; Precorrin ...   \n",
+       "1  A0A009HTZ2  Precorrin methyltransferase family; Precorrin ...   \n",
+       "2  A0A009IVE2  Precorrin methyltransferase family; Precorrin ...   \n",
+       "3  A0A009MYL5  Precorrin methyltransferase family; Precorrin ...   \n",
+       "4  A0A009PHM9  Precorrin methyltransferase family; Precorrin ...   \n",
+       "\n",
+       "                                    Modified residue  \\\n",
+       "0  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "1  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "2  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "3  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "4  MOD_RES 129; /note=\"Phosphoserine\"; /evidence=...   \n",
+       "\n",
+       "                                            Sequence  \\\n",
+       "0  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...   \n",
+       "1  MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...   \n",
+       "2  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...   \n",
+       "3  MDIFPISLKLQQQHCLIVGGGHIALRKANLLAKAGAVIDIIAPAIE...   \n",
+       "4  MDIFPISLKLQQQRCLIVGGGHIALRKATLLAKAGAIIDVVAPAIE...   \n",
+       "\n",
+       "                                           PTM sites  \n",
+       "0  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "1  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "2  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "3  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  \n",
+       "4  [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, ...  "
+      ]
+     },
+     "execution_count": 50,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "# Function to split sequences and PTM sites into chunks\n",
+    "def split_into_chunks(row):\n",
+    "    sequence = row['Sequence']\n",
+    "    ptm_sites = row['PTM sites']\n",
+    "    chunk_size = 1000\n",
+    "    \n",
+    "    # Calculate the number of chunks\n",
+    "    num_chunks = (len(sequence) + chunk_size - 1) // chunk_size\n",
+    "    \n",
+    "    # Split sequences and PTM sites into chunks\n",
+    "    sequence_chunks = [sequence[i * chunk_size: (i + 1) * chunk_size] for i in range(num_chunks)]\n",
+    "    ptm_sites_chunks = [ptm_sites[i * chunk_size: (i + 1) * chunk_size] for i in range(num_chunks)]\n",
+    "    \n",
+    "    # Create new rows for each chunk\n",
+    "    rows = []\n",
+    "    for i in range(num_chunks):\n",
+    "        new_row = row.copy()\n",
+    "        new_row['Sequence'] = sequence_chunks[i]\n",
+    "        new_row['PTM sites'] = ptm_sites_chunks[i]\n",
+    "        rows.append(new_row)\n",
+    "    \n",
+    "    return rows\n",
+    "\n",
+    "# Create a new DataFrame to store the chunks\n",
+    "chunks_data = []\n",
+    "\n",
+    "# Iterate through each row of the original DataFrame and split into chunks\n",
+    "for _, row in data.iterrows():\n",
+    "    chunks_data.extend(split_into_chunks(row))\n",
+    "\n",
+    "# Convert the list of chunks into a DataFrame\n",
+    "chunks_df = pd.DataFrame(chunks_data)\n",
+    "\n",
+    "# Reset the index of the DataFrame\n",
+    "chunks_df.reset_index(drop=True, inplace=True)\n",
+    "\n",
+    "# Display the first few rows of the new DataFrame\n",
+    "chunks_df.head()\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 52,
+   "id": "0e36e5bb-8e57-45af-a9da-9171875a0b88",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "% Test Data: 21.17% | % Test Families: 15.15%:  15%|█▌        | 661/4364 [00:05<00:30, 120.20it/s]\n"
+     ]
+    }
+   ],
+   "source": [
+    "from tqdm import tqdm\n",
+    "import numpy as np\n",
+    "\n",
+    "# Function to split data into train and test based on families\n",
+    "def split_data(df):\n",
+    "    # Get a unique list of protein families\n",
+    "    unique_families = df['Protein families'].unique().tolist()\n",
+    "    np.random.shuffle(unique_families)  # Shuffle the list to randomize the order of families\n",
+    "    \n",
+    "    test_data = []\n",
+    "    test_families = []\n",
+    "    total_entries = len(df)\n",
+    "    total_families = len(unique_families)\n",
+    "    \n",
+    "    # Set up tqdm progress bar\n",
+    "    with tqdm(total=total_families) as pbar:\n",
+    "        for family in unique_families:\n",
+    "            # Separate out all proteins in the current family into the test data\n",
+    "            family_data = df[df['Protein families'] == family]\n",
+    "            test_data.append(family_data)\n",
+    "            \n",
+    "            # Update the list of test families\n",
+    "            test_families.append(family)\n",
+    "            \n",
+    "            # Remove the current family data from the original DataFrame\n",
+    "            df = df[df['Protein families'] != family]\n",
+    "            \n",
+    "            # Calculate the percentage of test data and the percentage of families in the test data\n",
+    "            percent_test_data = sum(len(data) for data in test_data) / total_entries * 100\n",
+    "            percent_test_families = len(test_families) / total_families * 100\n",
+    "            \n",
+    "            # Update tqdm progress bar with readout of percentages\n",
+    "            pbar.set_description(f'% Test Data: {percent_test_data:.2f}% | % Test Families: {percent_test_families:.2f}%')\n",
+    "            pbar.update(1)\n",
+    "            \n",
+    "            # Check if the 20% threshold for test data is crossed\n",
+    "            if percent_test_data >= 20:\n",
+    "                break\n",
+    "    \n",
+    "    # Concatenate the list of test data DataFrames into a single DataFrame\n",
+    "    test_df = pd.concat(test_data, ignore_index=True)\n",
+    "    \n",
+    "    return df, test_df  # Return the remaining data and the test data\n",
+    "\n",
+    "# Split the data into train and test based on families\n",
+    "train_df, test_df = split_data(chunks_df)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 53,
+   "id": "0d5e7371-a6d0-4c5c-8587-dd0037f052f8",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [],
+   "source": [
+    "import pandas as pd\n",
+    "\n",
+    "# Assuming train_df and test_df are your dataframes\n",
+    "fraction = 0.105  # 10.5%\n",
+    "\n",
+    "# Randomly select 10.5% of the data\n",
+    "reduced_train_df = train_df.sample(frac=fraction, random_state=42)\n",
+    "reduced_test_df = test_df.sample(frac=fraction, random_state=42)\n",
+    "\n",
+    "# Split the reduced dataframes into sequences and PTM sites\n",
+    "#train_sequences = reduced_train_df['Sequence']\n",
+    "#train_ptm_sites = reduced_train_df['PTM sites']\n",
+    "#test_sequences = reduced_test_df['Sequence']\n",
+    "#test_ptm_sites = reduced_test_df['PTM sites']\n",
+    "\n",
+    "# Save the reduced data as pickle files\n",
+    "#train_sequences.to_pickle('train_sequences.pkl')\n",
+    "#train_ptm_sites.to_pickle('train_ptm_sites.pkl')\n",
+    "#test_sequences.to_pickle('test_sequences.pkl')\n",
+    "#test_ptm_sites.to_pickle('test_ptm_sites.pkl')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 55,
+   "id": "a5ac2515-2aaa-4417-b5bb-09b25ce31d44",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "['50K_ptm_data/train_sequences_chunked_by_family.pkl',\n",
+       " '50K_ptm_data/test_sequences_chunked_by_family.pkl',\n",
+       " '50K_ptm_data/train_labels_chunked_by_family.pkl',\n",
+       " '50K_ptm_data/test_labels_chunked_by_family.pkl']"
+      ]
+     },
+     "execution_count": 55,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import pickle \n",
+    "\n",
+    "# Extract sequences and PTM site labels from the reduced train and test DataFrames\n",
+    "train_sequences_reduced = reduced_train_df['Sequence'].tolist()\n",
+    "train_labels_reduced = reduced_train_df['PTM sites'].tolist()\n",
+    "test_sequences_reduced = reduced_test_df['Sequence'].tolist()\n",
+    "test_labels_reduced = reduced_test_df['PTM sites'].tolist()\n",
+    "\n",
+    "# Save the lists to the specified pickle files\n",
+    "pickle_file_path = \"50K_ptm_data/\"\n",
+    "\n",
+    "with open(pickle_file_path + \"train_sequences_chunked_by_family.pkl\", \"wb\") as f:\n",
+    "    pickle.dump(train_sequences_reduced, f)\n",
+    "\n",
+    "with open(pickle_file_path + \"test_sequences_chunked_by_family.pkl\", \"wb\") as f:\n",
+    "    pickle.dump(test_sequences_reduced, f)\n",
+    "\n",
+    "with open(pickle_file_path + \"train_labels_chunked_by_family.pkl\", \"wb\") as f:\n",
+    "    pickle.dump(train_labels_reduced, f)\n",
+    "\n",
+    "with open(pickle_file_path + \"test_labels_chunked_by_family.pkl\", \"wb\") as f:\n",
+    "    pickle.dump(test_labels_reduced, f)\n",
+    "\n",
+    "# Return the paths to the saved pickle files\n",
+    "saved_files = [\n",
+    "    pickle_file_path + \"train_sequences_chunked_by_family.pkl\",\n",
+    "    pickle_file_path + \"test_sequences_chunked_by_family.pkl\",\n",
+    "    pickle_file_path + \"train_labels_chunked_by_family.pkl\",\n",
+    "    pickle_file_path + \"test_labels_chunked_by_family.pkl\"\n",
+    "]\n",
+    "saved_files\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 57,
+   "id": "5ec5c5fc-7e9a-4c2c-a954-b2d2ad168b11",
+   "metadata": {
+    "tags": []
+   },
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "{'50K_ptm_data/train_sequences_chunked_by_family.pkl': 5132, '50K_ptm_data/test_sequences_chunked_by_family.pkl': 1378, '50K_ptm_data/train_labels_chunked_by_family.pkl': 5132, '50K_ptm_data/test_labels_chunked_by_family.pkl': 1378}\n"
+     ]
+    }
+   ],
+   "source": [
+    "import pickle\n",
+    "\n",
+    "def get_number_of_rows(pickle_file):\n",
+    "    with open(pickle_file, \"rb\") as f:\n",
+    "        data = pickle.load(f)\n",
+    "    return len(data)\n",
+    "\n",
+    "# Paths to the pickle files\n",
+    "files = [\n",
+    "    \"50K_ptm_data/train_sequences_chunked_by_family.pkl\",\n",
+    "    \"50K_ptm_data/test_sequences_chunked_by_family.pkl\",\n",
+    "    \"50K_ptm_data/train_labels_chunked_by_family.pkl\",\n",
+    "    \"50K_ptm_data/test_labels_chunked_by_family.pkl\"\n",
+    "]\n",
+    "\n",
+    "# Get the number of rows for each file\n",
+    "number_of_rows = {file: get_number_of_rows(file) for file in files}\n",
+    "print(number_of_rows)\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "71cc9d3d-bb35-4e2a-a382-7218bff5cb53",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "esm2_binding_py38b",
+   "language": "python",
+   "name": "esm2_binding_py38b"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.8.17"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}