File size: 14,963 Bytes
8454617 a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 8454617 a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 8454617 a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 95d2d5a a90e0da 8454617 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 |
---
library_name: scvi-tools
license: cc-by-4.0
tags:
- biology
- genomics
- single-cell
- model_cls_name:TOTALVI
- scvi_version:1.2.0
- anndata_version:0.11.1
- modality:rna
- modality:protein
- tissue:thymus
- annotated:True
---
TotalVI is a variational inference model for single-cell RNA-seq as well as protein data that can
learn an underlying latent space, integrate technical batches, impute dropouts,
and predict protein expression given gene expression or missing protein data given gene expression
and protein data for a subset of proteins.
The learned low-dimensional latent representation of the data can be used for visualization and
clustering.
TotalVI takes as input a scRNA-seq gene expression and protein expression matrix with cells and
genes.
We provide an extensive [user guide](https://docs.scvi-tools.org/en/1.2.0/user_guide/models/totalvi.html).
- See our original manuscript for further details of the model:
[TotalVI manuscript](https://www.nature.com/articles/s41592-020-01050-x).
- See our manuscript on [scvi-hub](https://www.biorxiv.org/content/10.1101/2024.03.01.582887v2)
how to leverage pre-trained models.
This model can be used for fine tuning on new data using our Arches framework:
[Arches tutorial](https://docs.scvi-tools.org/en/1.0.0/tutorials/notebooks/scarches_scvi_tools.html).
# Model Description
CITE-seq to measure RNA and surface proteins in thymocytes from wild-type and T cell lineage-restricted mice to generate a comprehensive timeline of cell state for each T cell lineage.
# Metrics
We provide here key performance metrics for the uploaded model, if provided by the data uploader.
<details>
<summary><strong>Coefficient of variation</strong></summary>
The cell-wise coefficient of variation summarizes how well variation between different cells is
preserved by the generated model expression. Below a squared Pearson correlation coefficient of 0.4
, we would recommend not to use generated data for downstream analysis, while the generated latent
space might still be useful for analysis.
**Cell-wise Coefficient of Variation**:
Modality: rna
| Metric | Training Value | Validation Value |
|-------------------------|----------------|------------------|
| Mean Absolute Error | 0.57 | 0.56 |
| Pearson Correlation | 0.76 | 0.76 |
| Spearman Correlation | 0.83 | 0.83 |
| R² (R-Squared) | -0.10 | -0.08 |
Modality: protein
| Metric | Training Value | Validation Value |
|-------------------------|----------------|------------------|
| Mean Absolute Error | 0.32 | 0.32 |
| Pearson Correlation | 0.53 | 0.53 |
| Spearman Correlation | 0.78 | 0.78 |
| R² (R-Squared) | -1.46 | -1.43 |
The gene-wise coefficient of variation summarizes how well variation between different genes is
preserved by the generated model expression. This value is usually quite high.
**Gene-wise Coefficient of Variation**:
Modality: rna
| Metric | Training Value |
|-------------------------|----------------|
| Mean Absolute Error | 26.96 |
| Pearson Correlation | 0.95 |
| Spearman Correlation | 0.99 |
| R² (R-Squared) | -0.25 |
Modality: protein
| Metric | Training Value |
|-------------------------|----------------|
| Mean Absolute Error | 4.30 |
| Pearson Correlation | 0.40 |
| Spearman Correlation | 0.73 |
| R² (R-Squared) | -6.19 |
</details>
<details>
<summary><strong>Differential expression metric</strong></summary>
The differential expression metric provides a summary of the differential expression analysis
between cell types or input clusters. We provide here the F1-score, Pearson Correlation
Coefficient of Log-Foldchanges, Spearman Correlation Coefficient, and Area Under the Precision
Recall Curve (AUPRC) for the differential expression analysis using Wilcoxon Rank Sum test for each
cell-type.
**Differential expression**:
Modality: rna
| Index | gene_f1 | lfc_mae | lfc_pearson | lfc_spearman | roc_auc | pr_auc | n_cells |
| --- | --- | --- | --- | --- | --- | --- | --- |
| CD14-positive monocyte | 0.95 | 2.11 | 0.59 | 0.91 | 0.09 | 0.02 | 120843.00 |
| CD16-positive, CD56-dim natural killer cell, human | 0.95 | 2.35 | 0.45 | 0.84 | 0.09 | 0.02 | 92848.00 |
| naive thymus-derived CD4-positive, alpha-beta T cell | 0.89 | 2.76 | 0.39 | 0.75 | 0.09 | 0.02 | 63096.00 |
| effector CD8-positive, alpha-beta T cell | 0.88 | 3.49 | 0.40 | 0.72 | 0.07 | 0.02 | 53534.00 |
| central memory CD4-positive, alpha-beta T cell | 0.93 | 2.59 | 0.34 | 0.74 | 0.06 | 0.02 | 49904.00 |
| naive B cell | 0.93 | 3.29 | 0.40 | 0.72 | 0.08 | 0.02 | 44136.00 |
| naive thymus-derived CD8-positive, alpha-beta T cell | 0.93 | 3.54 | 0.37 | 0.67 | 0.07 | 0.02 | 31175.00 |
| mature NK T cell | 0.91 | 3.62 | 0.44 | 0.63 | 0.04 | 0.01 | 21673.00 |
| effector memory CD8-positive, alpha-beta T cell | 0.82 | 4.47 | 0.37 | 0.56 | 0.07 | 0.02 | 18917.00 |
| T-helper 22 cell | 0.90 | 4.01 | 0.42 | 0.60 | 0.06 | 0.02 | 18379.00 |
| gamma-delta T cell | 0.88 | 4.52 | 0.39 | 0.50 | 0.05 | 0.01 | 15942.00 |
| platelet | 0.89 | 4.32 | 0.54 | 0.67 | 0.06 | 0.02 | 15847.00 |
| T follicular helper cell | 0.93 | 4.43 | 0.41 | 0.55 | 0.06 | 0.02 | 13608.00 |
| mucosal invariant T cell | 0.86 | 4.85 | 0.42 | 0.48 | 0.06 | 0.02 | 10992.00 |
| CD16-negative, CD56-bright natural killer cell, human | 0.85 | 5.29 | 0.38 | 0.44 | 0.05 | 0.02 | 10442.00 |
| class switched memory B cell | 0.89 | 5.17 | 0.45 | 0.49 | 0.08 | 0.02 | 7244.00 |
| immature B cell | 0.89 | 5.66 | 0.45 | 0.45 | 0.10 | 0.02 | 5238.00 |
| natural killer cell | 0.88 | 5.20 | 0.46 | 0.45 | 0.09 | 0.02 | 4963.00 |
| plasmacytoid dendritic cell | 0.90 | 5.18 | 0.46 | 0.46 | 0.05 | 0.02 | 4612.00 |
| CD14-low, CD16-positive monocyte | 0.91 | 4.38 | 0.58 | 0.59 | 0.10 | 0.02 | 4140.00 |
| plasmablast | 0.80 | 4.96 | 0.52 | 0.55 | 0.10 | 0.02 | 4121.00 |
| IgG plasma cell | 0.70 | 5.01 | 0.51 | 0.52 | 0.12 | 0.01 | 3527.00 |
| dendritic cell, human | 0.83 | 5.35 | 0.41 | 0.40 | 0.62 | 0.20 | 3357.00 |
| unswitched memory B cell | 0.92 | 5.08 | 0.49 | 0.47 | 0.17 | 0.02 | 3285.00 |
| myeloid dendritic cell | 0.82 | 5.44 | 0.47 | 0.45 | 0.13 | 0.02 | 3243.00 |
| B cell | 0.86 | 5.03 | 0.51 | 0.48 | 0.14 | 0.02 | 3024.00 |
| IgA plasma cell | 0.70 | 5.22 | 0.50 | 0.49 | 0.13 | 0.02 | 2699.00 |
| effector memory CD4-positive, alpha-beta T cell | 0.88 | 5.17 | 0.48 | 0.41 | 0.14 | 0.02 | 2634.00 |
| malignant cell | 0.94 | 5.40 | 0.48 | 0.45 | 0.28 | 0.02 | 2291.00 |
| CD34-positive, CD38-negative hematopoietic stem cell | 0.78 | 5.68 | 0.46 | 0.47 | 0.12 | 0.02 | 2238.00 |
| erythrocyte | 0.79 | 5.58 | 0.41 | 0.26 | 0.63 | 0.33 | 2232.00 |
| CD8-positive, alpha-beta T cell | 0.84 | 5.28 | 0.48 | 0.39 | 0.35 | 0.03 | 1355.00 |
| IgM plasma cell | 0.78 | 4.70 | 0.53 | 0.50 | 0.30 | 0.02 | 1163.00 |
| ILC1, human | 0.79 | 4.52 | 0.52 | 0.47 | 0.36 | 0.03 | 776.00 |
| erythroid progenitor cell, mammalian | 0.73 | 5.52 | 0.51 | 0.46 | 0.32 | 0.02 | 773.00 |
| monocyte | 0.89 | 4.21 | 0.57 | 0.51 | 0.38 | 0.02 | 649.00 |
| CD4-positive, alpha-beta T cell | 0.81 | 4.64 | 0.54 | 0.48 | 0.33 | 0.02 | 624.00 |
| dendritic cell | 0.75 | 5.15 | 0.50 | 0.41 | 0.47 | 0.03 | 585.00 |
| T-helper 1 cell | 0.86 | 4.12 | 0.55 | 0.51 | 0.35 | 0.02 | 481.00 |
| regulatory T cell | 0.77 | 3.97 | 0.56 | 0.54 | 0.36 | 0.02 | 329.00 |
| hematopoietic precursor cell | 0.68 | 4.34 | 0.60 | 0.57 | 0.30 | 0.02 | 180.00 |
| group 2 innate lymphoid cell, human | 0.66 | 3.08 | 0.58 | 0.63 | 0.26 | 0.02 | 93.00 |
| T-helper 2 cell | 0.65 | 2.71 | 0.62 | 0.67 | 0.18 | 0.02 | 55.00 |
| myeloid lineage restricted progenitor cell | 0.57 | 3.85 | 0.54 | 0.57 | 0.33 | 0.02 | 53.00 |
| megakaryocyte | 0.64 | 3.73 | 0.61 | 0.59 | 0.30 | 0.02 | 53.00 |
| T-helper 17 cell | 0.53 | 2.45 | 0.55 | 0.70 | 0.20 | 0.02 | 13.00 |
Modality: protein
| Index | gene_f1 | lfc_mae | lfc_pearson | lfc_spearman | roc_auc | pr_auc | n_cells |
| --- | --- | --- | --- | --- | --- | --- | --- |
| CD14-positive monocyte | 0.95 | 0.07 | 1.00 | 0.99 | 0.26 | 0.12 | 120843.00 |
| CD16-positive, CD56-dim natural killer cell, human | 0.95 | 0.06 | 0.99 | 0.98 | 0.22 | 0.12 | 92848.00 |
| naive thymus-derived CD4-positive, alpha-beta T cell | 0.84 | 0.06 | 0.99 | 0.98 | 0.37 | 0.13 | 63096.00 |
| effector CD8-positive, alpha-beta T cell | 0.95 | 0.05 | 0.99 | 0.97 | 0.17 | 0.09 | 53534.00 |
| central memory CD4-positive, alpha-beta T cell | 1.00 | 0.06 | 1.00 | 0.99 | 0.33 | 0.12 | 49904.00 |
| naive B cell | 1.00 | 0.08 | 1.00 | 0.96 | 0.20 | 0.12 | 44136.00 |
| naive thymus-derived CD8-positive, alpha-beta T cell | 0.95 | 0.07 | 0.99 | 0.95 | 0.13 | 0.08 | 31175.00 |
| mature NK T cell | 0.95 | 0.06 | 0.99 | 0.97 | 0.21 | 0.10 | 21673.00 |
| effector memory CD8-positive, alpha-beta T cell | 0.84 | 0.05 | 0.99 | 0.98 | 0.06 | 0.11 | 18917.00 |
| T-helper 22 cell | 0.95 | 0.06 | 0.99 | 0.98 | 0.11 | 0.08 | 18379.00 |
| gamma-delta T cell | 0.89 | 0.07 | 0.97 | 0.93 | 0.26 | 0.18 | 15942.00 |
| platelet | 0.79 | 0.10 | 0.97 | 0.95 | 0.21 | 0.11 | 15847.00 |
| T follicular helper cell | 1.00 | 0.07 | 1.00 | 0.99 | 0.24 | 0.12 | 13608.00 |
| mucosal invariant T cell | 0.89 | 0.08 | 0.97 | 0.94 | 0.15 | 0.09 | 10992.00 |
| CD16-negative, CD56-bright natural killer cell, human | 0.95 | 0.08 | 0.98 | 0.93 | 0.44 | 0.47 | 10442.00 |
| class switched memory B cell | 0.89 | 0.09 | 0.99 | 0.96 | 0.11 | 0.12 | 7244.00 |
| immature B cell | 0.89 | 0.13 | 0.98 | 0.93 | 0.26 | 0.14 | 5238.00 |
| natural killer cell | 0.89 | 0.06 | 0.98 | 0.98 | 0.68 | 0.70 | 4963.00 |
| plasmacytoid dendritic cell | 0.84 | 0.09 | 0.98 | 0.97 | 0.54 | 0.56 | 4612.00 |
| CD14-low, CD16-positive monocyte | 0.89 | 0.08 | 0.99 | 0.98 | 0.58 | 0.23 | 4140.00 |
| plasmablast | 0.79 | 0.08 | 0.99 | 0.97 | 0.47 | 0.49 | 4121.00 |
| IgG plasma cell | 0.89 | 0.08 | 0.99 | 0.95 | 0.47 | 0.51 | 3527.00 |
| dendritic cell, human | 0.79 | 0.10 | 0.97 | 0.90 | 0.94 | 0.90 | 3357.00 |
| unswitched memory B cell | 0.89 | 0.10 | 0.99 | 0.96 | 0.63 | 0.61 | 3285.00 |
| myeloid dendritic cell | 0.89 | 0.11 | 0.97 | 0.95 | 0.74 | 0.74 | 3243.00 |
| B cell | 0.89 | 0.10 | 0.98 | 0.92 | 0.58 | 0.59 | 3024.00 |
| IgA plasma cell | 0.89 | 0.10 | 0.97 | 0.91 | 0.47 | 0.48 | 2699.00 |
| effector memory CD4-positive, alpha-beta T cell | 0.95 | 0.08 | 0.99 | 0.95 | 0.79 | 0.80 | 2634.00 |
| malignant cell | 0.89 | 0.09 | 0.99 | 0.99 | 0.17 | 0.08 | 2291.00 |
| CD34-positive, CD38-negative hematopoietic stem cell | 0.84 | 0.09 | 0.97 | 0.95 | 0.37 | 0.35 | 2238.00 |
| erythrocyte | 0.89 | 0.07 | 0.99 | 0.98 | 0.21 | 0.25 | 2232.00 |
| CD8-positive, alpha-beta T cell | 0.63 | 0.09 | 0.92 | 0.87 | 0.68 | 0.52 | 1355.00 |
| IgM plasma cell | 0.95 | 0.09 | 0.98 | 0.92 | 0.42 | 0.45 | 1163.00 |
| ILC1, human | 0.84 | 0.10 | 0.97 | 0.89 | 0.53 | 0.55 | 776.00 |
| erythroid progenitor cell, mammalian | 0.68 | 0.14 | 0.95 | 0.93 | 0.23 | 0.23 | 773.00 |
| monocyte | 0.84 | 0.09 | 0.97 | 0.96 | 0.68 | 0.70 | 649.00 |
| CD4-positive, alpha-beta T cell | 0.84 | 0.12 | 0.91 | 0.85 | 0.58 | 0.59 | 624.00 |
| dendritic cell | 0.74 | 0.19 | 0.85 | 0.81 | 0.53 | 0.44 | 585.00 |
| T-helper 1 cell | 0.95 | 0.08 | 0.99 | 0.96 | 0.73 | 0.71 | 481.00 |
| regulatory T cell | 0.84 | 0.14 | 0.97 | 0.94 | 0.89 | 0.84 | 329.00 |
| hematopoietic precursor cell | 0.74 | 0.15 | 0.95 | 0.91 | 0.36 | 0.28 | 180.00 |
| group 2 innate lymphoid cell, human | 0.63 | 0.36 | 0.14 | 0.60 | 0.29 | 0.32 | 93.00 |
| T-helper 2 cell | 0.79 | 0.40 | 0.15 | 0.69 | 0.74 | 0.65 | 55.00 |
| myeloid lineage restricted progenitor cell | 0.53 | 0.28 | 0.97 | 0.74 | 0.56 | 0.37 | 53.00 |
| megakaryocyte | 0.47 | 0.25 | 0.98 | 0.79 | 0.54 | 0.44 | 53.00 |
| T-helper 17 cell | 0.79 | 0.55 | 0.82 | 0.78 | 0.48 | 0.38 | 13.00 |
</details>
# Model Properties
We provide here key parameters used to setup and train the model.
<details>
<summary><strong>Model Parameters</strong></summary>
These provide the settings to setup the original model:
```json
{
"n_latent": 20,
"gene_dispersion": "gene",
"protein_dispersion": "protein",
"gene_likelihood": "nb",
"latent_distribution": "normal",
"empirical_protein_background_prior": null,
"override_missing_proteins": false
}
```
</details>
<details>
<summary><strong>Setup Data Arguments</strong></summary>
Arguments passed to setup_anndata of the original model:
```json
{
"rna_layer": "counts",
"protein_layer": null,
"batch_key": "donor_id",
"size_factor_key": null,
"categorical_covariate_keys": null,
"continuous_covariate_keys": null,
"modalities": {
"rna_layer": "rna",
"protein_layer": "protein",
"batch_key": "rna"
}
}
```
</details>
<details>
<summary><strong>Data Registry</strong></summary>
Registry elements for AnnData manager:
| Registry Key | scvi-tools Location |
|--------------------------|--------------------------------------|
| X | adata.mod['rna'].layers['counts'] |
| batch | adata.mod['rna'].obs['_scvi_batch'] |
| labels | adata.obs['_scvi_labels'] |
| latent_qzm | adata.obsm['totalvi_latent_qzm'] |
| latent_qzv | adata.obsm['totalvi_latent_qzv'] |
| minify_type | adata.uns['_scvi_adata_minify_type'] |
| observed_lib_size | adata.obs['observed_lib_size'] |
| proteins | adata.mod['protein'].X |
- **Data is Minified**: False
</details>
<details>
<summary><strong>Summary Statistics</strong></summary>
| Summary Stat Key | Value |
|--------------------------|-------|
| n_batch | 120 |
| n_cells | 647366 |
| n_extra_categorical_covs | 0 |
| n_extra_continuous_covs | 0 |
| n_labels | 1 |
| n_latent_qzm | 20 |
| n_latent_qzv | 20 |
| n_proteins | 192 |
| n_vars | 4000 |
</details>
<details>
<summary><strong>Training</strong></summary>
<!-- If your model is not uploaded with any data (e.g., minified data) on the Model Hub, then make
sure to provide this field if you want users to be able to access your training data. See the
scvi-tools documentation for details. -->
**Training data url**: Not provided by uploader
If provided by the original uploader, for those interested in understanding or replicating the
training process, the code is available at the link below.
**Training Code URL**: https://github.com/YosefLab/Thymus_CITE-seq/blob/main/totalVI_AllData/totalVI_thymus111.ipynb
</details>
# References
Steier, Z., Aylard, D.A., McIntyre, L.L. et al. Single-cell multiomic analysis of thymocyte development reveals drivers of CD4+ T cell and CD8+ T cell lineage commitment. Nat Immunol 24, 1579–1590 (2023). https://doi.org/10.1038/s41590-023-01584-0.
|