Can Ergen

Results 12 issues of Can Ergen

Uses proportions computed in other deconvolution method to infer cell activation state of those cells. Application activation for e.g. Cell2Location results. L1 loss changed. Still in discussion. Optional obsm field.

In models/_condscvi.py line 139, we don't check for log_variational (default is true in CondSCVI). The input for the decoder at inference is therefore different from the training input. Effect is...

bug

- [ ] Closes #xxxx (Replace xxxx with the GitHub issue number) - [X] Tests added and passed if fixing a bug or adding a new feature - [X] All...

Adds minified mode to CondSCVI. Simplifies minification for new models. Adds add_latent_posterior mode which keeps count data. Fixes glitches in DE function. - [X] Tests added and passed if fixing...

P0

Disable adversarial training when using a single batch.

If defining protein_names_uns_key the DE function fails. This is unexpected behaviour. ``` def test_totalvi_de_error(save_path): adata = synthetic_iid() TOTALVI.setup_anndata( adata, batch_key="batch", protein_expression_obsm_key="protein_expression", ) model = TOTALVI(adata) model.train(1, train_size=0.5) model.differential_expression(groupby="labels", group1="label_1") adata...

bug

CustomDataloaders currently don't support advanced capabilities like scArches or celltype prediction in scANVI. We have to create a registry without setup_anndata that contains the same elements (see below). https://github.com/chanzuckerberg/cellxgene-census/blob/222efddf2ce82f93f76329aa353962c1dc2400ac/api/python/notebooks/experimental/pytorch_loader_scvi.ipynb is...

enhancement
P0

### What kind of feature would you like to request? New plotting function: A kind of plot you would like to seein `sc.pl`? ### Please describe your wishes As far...

Area - Plotting 🌺

Hi, on the webpage the CELLXGENE UMAP links are all broken and likely require an update.