Emma Dann

Exact same problem here running `fastMNN` within Jupyter notebook. Things I tried that didn't work: - saving the SingleCellExperiment object as RDS and reloading (works within R) - Running `library(SingleCellExperiment)`...

Hopefully the new tutorial covers the use of contrasts to specify comparisons. I could add an example on subsetting to a certain lineage or something.

In principle I agree with adding the diagnostic plots to the package, although in my head that's more material for extended tutorials since it needs some guidance for interpretation. But...

Hi @adamgayoso, thanks a lot for these suggestions! I will regenerate the datasets for the website and notify here as soon as they are up. The jax speed up sounds...

The batch key used is a concatenation of `method` (10X protocol, 3' or 5') and `donor` (see under "Add batch key " [here](https://github.com/Teichlab/Pan_fetal_immune/blob/master/src/02_integration_scRNA/integration_scRNA_scVI_fulldata.ipynb))

The .h5ad objects for download should now be updated. Could I ask you to try again to check if the plotting problem persists?

Notes from troubleshooting attempts: Part of the problem could be the NaNs (https://github.com/theislab/scanpy/issues/2133): I found the maternal contaminants were not flagged correctly in this object, these are cells with `adata.obs['celltype_annotation']`...

sounds good to try, although I am not sure how comparable results would be, and it would need custom implementation for some edgeR innards e.g. TMM normalization

I get a very similar error running `rsc.pp.neighbors`. I just installed `rapids-singlecell` in a fresh conda environment, following the [installation instructions](https://rapids-singlecell.readthedocs.io/en/latest/Installation.html). Then I ran (in jupyter environment): ```python adata =...

Hi @Zethson Apologies for the late react on this and thanks a lot for the feedback. I am a bit slumped with some urgent work at the moment so I...