Find DE in single cell RNA sequencing

Question

[kayah]

https://satijalab.org/seurat/articles/de_vignette

I follow the protocol which made by Satija lab. but it didnt' work when I'm doing this part(**)

  pseudo_harmony <- AggregateExpression(obj_Harmony, assays = "RNA", return.seurat = T, group.by = c("type","predicted.celltype.l2"))
    pseudo_harmony$celltype.WAT <- paste(pseudo_harmony$predicted.celltype.l2, pseudo_harmony$type, sep = "_")
    Idents(pseudo_harmony) <- "orig.ident"
    View(GetAssayData(pseudo_harmony, slot = "counts", assay = "RNA"))
    **bulk.mac.de <- FindMarkers(object = pseudo_harmony, 
                               ident.1 = "Old_hASPC2", 
                               ident.2 = "Young_hASPC2",
                               test.use = "DESeq2")

It says Error in ValidateCellGroups(object = object, cells.1 = cells.1, cells.2 = cells.2, : Cell group 1 has fewer than 3 cells

I think they consider one cell type as a one cell. but don't know which process go wrong.
I want to compare old sample and young sample by cell type. Thank you!!

[biofalconch]

Hi, the problem here is that you are using AggregateExpression first, converting your data into a pseudo-bulk. FindMarkers requires a normal single cell dataset (not pseudo-bulk), try running it on the obj_Harmony object instead.

[kayah]

I want to try both pseudobulk and regular Seurat data to compare the markers identified from each method.