--- 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 | 3.08 | 3.02 | | Pearson Correlation | 0.70 | 0.72 | | Spearman Correlation | 0.72 | 0.73 | | R² (R-Squared) | 0.25 | 0.28 | 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 | 25.94 | | Pearson Correlation | 0.67 | | Spearman Correlation | 0.71 | | R² (R-Squared) | -54.23 |
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 | | --- | --- | --- | --- | --- | --- | --- | --- | | neutrophil | 0.90 | 4.02 | 0.13 | 0.39 | 0.32 | 0.87 | 2911.00 | | CD4-positive, alpha-beta T cell | 0.87 | 3.14 | 0.21 | 0.49 | 0.45 | 0.82 | 2025.00 | | monocyte | 0.84 | 2.84 | 0.22 | 0.49 | 0.44 | 0.77 | 1389.00 | | CD8-positive, alpha-beta T cell | 0.75 | 4.40 | 0.12 | 0.42 | 0.42 | 0.76 | 1147.00 | | granulocyte | 0.69 | 3.71 | 0.24 | 0.53 | 0.55 | 0.86 | 853.00 | | plasma cell | 0.75 | 4.56 | 0.14 | 0.26 | 0.34 | 0.88 | 825.00 | | erythroid progenitor cell | 0.74 | 4.12 | 0.29 | 0.61 | 0.57 | 0.91 | 757.00 | | mature NK T cell | 0.59 | 5.74 | 0.14 | 0.27 | 0.44 | 0.72 | 678.00 | | hematopoietic stem cell | 0.73 | 3.70 | 0.17 | 0.41 | 0.54 | 0.83 | 617.00 | | memory B cell | 0.52 | 6.95 | 0.22 | 0.20 | 0.43 | 0.72 | 310.00 | | common myeloid progenitor | 0.83 | 4.78 | 0.23 | 0.53 | 0.62 | 0.90 | 287.00 | | macrophage | 0.69 | 7.61 | 0.34 | 0.37 | 0.44 | 0.81 | 265.00 | | naive B cell | 0.46 | 8.94 | 0.22 | 0.16 | 0.45 | 0.73 | 142.00 | | erythrocyte | 0.50 | 10.99 | 0.25 | 0.13 | 0.45 | 0.93 | 87.00 |
# 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 | 12293 | | n_labels | 14 | | 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