---
library_name: scvi-tools
license: cc-by-4.0
tags:
- biology
- genomics
- single-cell
- model_cls_name:CondSCVI
- scvi_version:1.2.0
- anndata_version:0.11.1
- modality:rna
- tissue:various
- annotated:True
---
CondSCVI is a variational inference model for single-cell RNA-seq data that can learn an underlying
latent space. The predictions of the model are meant to be afterward
used for deconvolution of a second spatial transcriptomics dataset in DestVI. DestVI predicts the
cell-type proportions as well as cell type-specific activation state
in the spatial data.
CondSCVI takes as input a scRNA-seq gene expression matrix with cells and genes as well as a
cell-type annotation for all cells.
We provide an extensive [user guide](https://docs.scvi-tools.org/en/1.2.0/user_guide/models/destvi.html)
for DestVI including a description of CondSCVI.
- See our original manuscript for further details of the model:
[DestVI manuscript](https://www.nature.com/articles/s41587-022-01272-8).
- See our manuscript on [scvi-hub](https://www.biorxiv.org/content/10.1101/2024.03.01.582887v2)
how to leverage pre-trained models.
# Model Description
Tabula Sapiens is a benchmark, first-draft human cell atlas of nearly 500,000 cells from 24 organs of 15 normal human subjects.
# Metrics
We provide here key performance metrics for the uploaded model, if provided by the data uploader.
Coefficient of variation
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**:
| Metric | Training Value | Validation Value |
|-------------------------|----------------|------------------|
| Mean Absolute Error | 1.92 | 1.93 |
| Pearson Correlation | 0.76 | 0.75 |
| Spearman Correlation | 0.76 | 0.74 |
| R² (R-Squared) | 0.46 | 0.44 |
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**:
| Metric | Training Value |
|-------------------------|----------------|
| Mean Absolute Error | 23.52 |
| Pearson Correlation | 0.25 |
| Spearman Correlation | 0.31 |
| R² (R-Squared) | -9.26 |
Differential expression metric
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**:
| Index | gene_f1 | lfc_mae | lfc_pearson | lfc_spearman | roc_auc | pr_auc | n_cells |
| --- | --- | --- | --- | --- | --- | --- | --- |
| conjunctival epithelial cell | 0.86 | 2.12 | 0.41 | 0.73 | 0.36 | 0.92 | 4587.00 |
| corneal epithelial cell | 0.85 | 3.42 | 0.07 | 0.43 | 0.48 | 0.92 | 1488.00 |
| eye photoreceptor cell | 0.70 | 4.86 | 0.26 | 0.44 | 0.42 | 0.86 | 891.00 |
| keratocyte | 0.79 | 3.80 | 0.39 | 0.52 | 0.51 | 0.84 | 595.00 |
| retinal blood vessel endothelial cell | 0.50 | 4.36 | 0.35 | 0.41 | 0.54 | 0.86 | 466.00 |
| Mueller cell | 0.46 | 5.18 | 0.33 | 0.47 | 0.48 | 0.81 | 360.00 |
| stromal cell | 0.59 | 1.91 | 0.42 | 0.60 | 0.37 | 0.69 | 325.00 |
| T cell | 0.26 | 7.06 | 0.16 | 0.13 | 0.40 | 0.80 | 237.00 |
| microglial cell | 0.27 | 6.03 | 0.16 | 0.19 | 0.46 | 0.76 | 209.00 |
| radial glial cell | 0.62 | 4.05 | 0.31 | 0.47 | 0.53 | 0.81 | 195.00 |
| dendritic cell | 0.47 | 7.61 | 0.25 | 0.16 | 0.45 | 0.78 | 182.00 |
| melanocyte | 0.30 | 5.48 | 0.41 | 0.25 | 0.46 | 0.73 | 144.00 |
| stem cell | 0.44 | 4.51 | 0.40 | 0.27 | 0.48 | 0.63 | 144.00 |
| macrophage | 0.28 | 6.21 | 0.29 | 0.24 | 0.48 | 0.74 | 108.00 |
| endothelial cell | 0.46 | 6.09 | 0.40 | 0.36 | 0.42 | 0.69 | 105.00 |
| B cell | 0.39 | 9.19 | 0.19 | 0.16 | 0.44 | 0.78 | 102.00 |
| fibroblast | 0.46 | 5.48 | 0.43 | 0.40 | 0.43 | 0.66 | 95.00 |
| surface ectodermal cell | 0.41 | 6.47 | 0.45 | 0.31 | 0.45 | 0.70 | 54.00 |
| epithelial cell of lacrimal sac | 0.28 | 6.41 | 0.43 | 0.16 | 0.46 | 0.68 | 52.00 |
| retinal pigment epithelial cell | 0.13 | 11.50 | 0.05 | 0.07 | 0.44 | 0.76 | 49.00 |
| plasma cell | 0.10 | 7.87 | 0.33 | 0.16 | 0.49 | 0.86 | 46.00 |
| monocyte | 0.25 | 7.86 | 0.34 | 0.17 | 0.43 | 0.69 | 44.00 |
| fat cell | 0.29 | 6.64 | 0.33 | 0.16 | 0.48 | 0.70 | 44.00 |
| retinal bipolar neuron | 0.20 | 7.49 | 0.32 | 0.28 | 0.40 | 0.64 | 34.00 |
| erythroid lineage cell | 0.11 | 15.97 | 0.01 | 0.01 | 0.51 | 0.96 | 30.00 |
| CD4-positive, alpha-beta T cell | 0.22 | 7.92 | 0.34 | 0.18 | 0.40 | 0.58 | 23.00 |
| pigmented ciliary epithelial cell | 0.28 | 6.96 | 0.44 | 0.23 | 0.38 | 0.63 | 17.00 |
# Model Properties
We provide here key parameters used to setup and train the model.
Model Parameters
These provide the settings to setup the original model:
```json
{
"n_hidden": 128,
"n_latent": 5,
"n_layers": 2,
"weight_obs": false,
"dropout_rate": 0.05
}
```
Setup Data Arguments
Arguments passed to setup_anndata of the original model:
```json
{
"labels_key": "cell_ontology_class",
"layer": null,
"batch_key": null
}
```
Data Registry
Registry elements for AnnData manager:
| Registry Key | scvi-tools Location |
|--------------|---------------------------|
| X | adata.X |
| labels | adata.obs['_scvi_labels'] |
- **Data is Minified**: False
Summary Statistics
| Summary Stat Key | Value |
|------------------|-------|
| n_cells | 10626 |
| n_labels | 27 |
| n_vars | 3000 |
Training
**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/scvi-hub-models/blob/main/src/scvi_hub_models/TS_train_all_tissues.ipynb
# References
The Tabula Sapiens Consortium. The Tabula Sapiens: A multiple-organ, single-cell transcriptomic atlas of humans. Science, May 2022. doi:10.1126/science.abl4896