release: v1.1 models

README.md

@@ -6,85 +6,69 @@ tags:
 - genomics
 - single-cell
 - model_cls_name:SCANVI
-- scvi_version:1.0.0
-- anndata_version:0.9.1
 - modality:rna
 - annotated:True
 ---
 
 # Description
 
-Human scANVI reference model
+Human preimplantation development model spanning early stages of development. The
+model was trained utilizing single‐cell ANnotation using Variational Inference
+(scANVI, [Xu et al., 2021]) implemented in [scvi-tools]. In short, scANVI raw
+single-cell RNA sequencing (scRNA-seq) count matrix - cell by gene, where values
+represent gene expression measured by counting number of transcribed RNA.
 
-# Model properties
+# Model Training
 
-Many model properties are in the model tags. Some more are listed below.
+- [raw dataset](https://zenodo.org/records/13749348/files/32_human_adata.h5ad)
+- [notebook analysis](https://github.com/brickmanlab/proks-salehin-et-al/blob/master/notebooks/15_human_scANVI_fix.ipynb)
+
+# Metrics
+
+Cell type (`ct`) prediction
+
+| Metric            |  Score              |
+|-------------------|---------------------|
+| Accuracy score    |  0.7968144640551011 |
+| Balanced accuracy |  0.8502734650790613 |
+| F1 (micro)        |  0.7968144640551011 |
+| F1 (macro)        |  0.8150578255414443 |
+
+# Model parameters
+
+Below we provide settings for scANVI setup
+
+`lvae.init_params_["non_kwargs"]`
 
-**model_init_params**:
 ```json
 {
-    "n_hidden": 128,
-    "n_latent": 10,
-    "n_layers": 2,
-    "dropout_rate": 0.1,
-    "dispersion": "gene",
-    "gene_likelihood": "nb",
-    "linear_classifier": false,
-    "latent_distribution": "normal"
+    "n_hidden": 128, 
+    "n_latent": 10, 
+    "n_layers": 2, 
+    "dropout_rate": 0.1, 
+    "dispersion": "gene", 
+    "gene_likelihood": "nb", 
+    "linear_classifier": false
 }
 ```
 
-**model_setup_anndata_args**:
+`lvae.adata_manager.registry['setup_args']`
+
 ```json
 {
-    "labels_key": "ct_fine",
+    "labels_key": "ct",
     "unlabeled_category": "Unknown",
     "layer": "counts",
     "batch_key": "batch",
-    "size_factor_key": null,
-    "categorical_covariate_keys": null,
+    "size_factor_key": null, 
+    "categorical_covariate_keys": null, 
     "continuous_covariate_keys": null
 }
 ```
 
-**model_summary_stats**:
-|     Summary Stat Key     | Value |
-|--------------------------|-------|
-|         n_batch          |   6   |
-|         n_cells          | 2323  |
-| n_extra_categorical_covs |   0   |
-| n_extra_continuous_covs  |   0   |
-|         n_labels         |  15   |
-|          n_vars          | 3000  |
-
-**model_data_registry**:
-| Registry Key |    scvi-tools Location    |
-|--------------|---------------------------|
-|      X       |  adata.layers['counts']   |
-|    batch     | adata.obs['_scvi_batch']  |
-|    labels    | adata.obs['_scvi_labels'] |
-
-**model_parent_module**: https://zenodo.org/records/10669600/files/32_human_adata.h5ad?download=1
-
-**data_is_minified**: False
-
-# Training data
-
-This is an optional link to where the training data is stored if it is too large
-to host on the huggingface Model hub.
-
-<!-- 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: https://github.com/brickmanlab/proks-salehin-et-al
-
-# Training code
-
-This is an optional link to the code used to train the model.
-
-Training code url: N/A
-
 # References
 
-Proks, Salehin et al., biorXiv
+Proks, M., Salehin, N. & Brickman, J.M. Deep learning-based models for preimplantation mouse and human embryos based on single-cell RNA sequencing. Nat Methods 22, 207–216 (2025). [https://doi.org/10.1038/s41592-024-02511-3](https://doi.org/10.1038/s41592-024-02511-3)
+
+[Xu et al., 2021]: https://www.embopress.org/doi/full/10.15252/msb.20209620
+[scvi-tools]: http://scvi-tools.org

_scvi_required_metadata.json

@@ -1,7 +1,7 @@
 {
-    "scvi_version": "1.0.0",
-    "anndata_version": "0.9.1",
+    "scvi_version": "1.1.5",
+    "anndata_version": "0.10.8",
     "model_cls_name": "SCANVI",
-    "training_data_url": null,
+    "training_data_url": "https://zenodo.org/records/13749348/files/32_human_adata.h5ad",
     "model_parent_module": "scvi.model"
 }

adata.h5ad

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a9b4fd24825a7fd5d280486d7295fa3cdce0e9e8d3991b9ddb24dc54e38f719e
-size 431824299
+oid sha256:77de1425327ba6493b33032d362877a4473868f5921fe7c3efe40fa2ef474b65
+size 416287216

model.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:93d0a4886a5759cf597ce78be733863f74a72845d4ae3e0101a0d0f2f31c82b0
-size 8390561
+oid sha256:39dc7404640d99ae5c896b900009976bf80c2ff52a44b792768e0d7f5c0e6055
+size 8393294