DESeq2 mentioned in Vignettes

[roanvanscheppingen]

DESeq2 doesn't have native support anymore according to #2938

Although it is still mentioned in the Vignette for Psuedobulking, in which it still can be used. I think it would be good to put a clarification that DESeq2 isn't of good use for sparse data and 'non pseudobulked' data. This is now not clear from the vignetteds https://satijalab.org/seurat/articles/de_vignette#perform-de-analysis-after-pseudobulking

[nigiord]

Hi @roanvanscheppingen , I read https://github.com/satijalab/seurat/issues/2938 but I'm not sure I understand. Does that mean that even after pseudobulking (AggregateExpression) the DESeq2 approach is not recommended?

The issue you reference is from 2020 and from what I see the DESeq2 option is still around in FindMarkers though. From what I understand, the pseudobulking should in principle fix the sparsity of the data (if you have enough cells), so I do not see why using DESeq2 (or any bulk approach) would be a problem in that case.