I would like to use DESeq2 to analyze differential gene expression between two clusters. Basically, during the clustering process with my data, it assigns 12 clusters. Then, based on the presence of certain conserved markers, each of those 12 clusters is assigned a cell type, or unknown. Previously I have used this to create a DESeq2 matrix based on each cell type, and then calculate DGE between the treated and control samples within that cell type. However, we have multiple clusters that assign to each cell type, and appear slightly separated on a UMAP. I would like to use DESeq2 to analyze DGE between two different clusters that map to the same cell type, but I am unsure if this is possible with the package or how to set up a matrix that allows for this. Does anyone know if this is possible, or how to do this?

analyze DGE between two different clusters

DGE between clusters identified with the expression data violates the model, see recent work by Witten lab at UW.