cell type specific enrichment of background in cellbender results #431
Description
I observed something odd in a dataset I have been working on:
specifically, background enrichment seems biased towards t cells and macrophages. Is this likely to be a false positive signal? Is there some metric I can use to figure this out? Or is cellbender just not a good fit for the data? Umaps colored by background fraction and cell type attatched:
