Single-cell RNA-seq FindMarker and EnhancedVolcano Plot: most genes are low -log10P value

Question

[yau]

Dear all,

I am recently running FindMarker and EnhancedVolcano in Seurat for single-cell RNA-seq data. However, the value of -log10p value of most of the genes are very low, only 2 genes are "red". Is there any error? How are I fix it? thanks!!

[yau]

Answer 1

[Mensur Dlakic]

You seem to have too stringent of a p-value cutoff (1e-5). Hopefully you get the concept of how the interplay of -log10P-value and log2FC works. Not many people would quibble with you if instead you took a p-value=0.01 as a cutoff, in which case you would have another 6-8 significantly down-regulated genes.

It is unusual to have only 2 genes with significantly different expression. Whether something is wrong or you simply have 2 similar conditions is tough to say from the (lack of) information we have.

[yau]

Dear Mensur,

Thanks so much for your reply!!

In my case, I got 1 control(healthy) and 2 disease mouse Single cell RNA-sew data, so I expect that there will be more DEG.

I think it’s still not enough even if I get 8 more DEG.

I really hope to find the solution. Is there any info I could provide to solve this? Thanks a lot!