How to obtain FDR across all regions investigated in DiffBind

Question

[g.persic1991]

Hi guys, I'm interested to make a Volcano plot and using different color i want to show which genes are UP and which are DOWN. To do this I need to use the logFc and FDR relative to ALL investigated regions and not just of DEGs. While logFC was found in in this way:

dbObj_region <- dba.analyze(dbObj_region, method=DBA_EDGER)
b <- dbObj_region$contrasts
all_site <- as.data.frame(b[[1]][5]$edgeR$LRT$table)

I can't find FDR value of all sites. Can you help me??

Thanks a lot Giuseppe

Answer 1

[Rory Stark]

The "correct" way to retrieve results in DiffBind is via the dba.report() function. Setting th=1 returns all results with FDR <= 1.0, which is all of them.

Answer 2

[seidel]

Could it be as simple as:

p.adjust(all_site$PValue, method="fdr")

? Your code isn't quite clear (to me), but if edgeR is used in the process, and I'm guessing that last line returns a typical edgeR table containing all the things investigated, then a typical way to adjust p-values (i.e. examine FDR) is to use the p.adjust function).

[g.persic1991]

Good solution... Thanks :)

Giuseppe