Hello everyone,
I'm currently analyzing single-cell RNA-seq data across two different conditions, with two replicates each. Typically, for identifying differentially expressed genes between conditions, creating pseudobulks and then employing DESeq2 or edgeR for differential expression analysis is quite standard and supported by various studies.
However, I'm curious about the feasibility of applying the pseudobulk method for detecting marker genes within a single condition. Specifically, could this method be used to identify differentially expressed genes between, say, cell type X and cell type Y within condition A? Although I see no theoretical reasons against it, I haven't come across any studies utilizing this approach. Most seem to useFindMarkers from Seurat, which does not account for pseudoreplication.
I know that we would still have the issue of "double dipping" (first clustering using gene expression and then comparing gene expression between the clusters) with the pseudobulk approach, however it seems a bit more robust than a simple wilcoxon test.
I would greatly appreciate any insights or experiences regarding this!
Thank you!
