Different list for findmarker function

[angkoomin]

Hi there,

I'm trying to do a DEG between groups. I applied these two codes:

Code 1: marker.genes.cluster <- findMarkers(cdScFiltAnnot_Luminal, groups = cdScFiltAnnot_Luminal$Tissue_Zone, pval.type = "all", BPPARAM = bpp)

Code 2: marker.genes.cluster.up <- findMarkers(cdScFiltAnnot_Luminal, groups = cdScFiltAnnot_Luminal$Tissue_Zone, pval.type = "all", lfc = 0.5, direction = "up", BPPARAM = bpp)

I'm getting a very different list of upregulated genes for both as below. This is the gene list in descending order of LFC for code 1:

And code 2:

I was wondering if someone can help to explain why I'm getting 2 very different list as I thought the 'direction' and 'lfc' is to help to filter rather than give a very different list. Please help as it will determine which code to use for further downstream analysis.

Thanks

[PeteHaitch]

It looks like the function is doing exactly what you're telling it to do. With your second function call, you're saying "prioritise genes that have a logFC > 0.5 (lfc = 0.5 and direction = "up") in all pairwise comparisons (pval.type = "all"). There aren't many genes that satisfy those conditions. But eyeballing the first list, those genes include: ZFP36, CEBPD, DUSP1, NR4A1, CYP1B1, ZFP36L2, KLF6, ..., which are exactly the top hits in the second list.