Enrichment maps-GSEA

[Kevin David Calupiña]

Hi, I'm Kevin. I'm working with human RNA-seq data and performed a GSEA using the software. When I try to export the results as an enrichment map to Cytoscape, it asks me to set two cutoffs: one for the p-value and one for the FDR. But isn’t the FDR calculated from the p-value? If so, wouldn’t applying both filters be redundant?

[Anthony Castanza]

Hi Kevin,

Sorry for the delay getting back to you. 

The GSEA's FDR is not just a corrected pValue. The nom pValue is a self-test of a single gene set vs the random null distribution generated for that set, whereas the FDR is a global competitive test among all the sets/distributions. You end up on basically the same place - you're not generally going to get a set with a significant FDR but an insignificant pValue, but because one test isn't a strict dependency of the other it can be appropriate to filter by both values.

Hope this helps,

-Anthony

Anthony S. Castanza, PhD
Curator, Molecular Signatures Database
Mesirov Lab, Department of Medicine
University of California, San Diego

On Tue, May 13, 2025, 7:32 AM Kevin David Calupiña <davidca...@gmail.com> wrote:

Hi, I'm Kevin. I'm working with human RNA-seq data and performed a GSEA using the software. When I try to export the results as an enrichment map to Cytoscape, it asks me to set two cutoffs: one for the p-value and one for the FDR. But isn’t the FDR calculated from the p-value? If so, wouldn’t applying both filters be redundant?

--
You received this message because you are subscribed to the Google Groups "gsea-help" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gsea-help+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/gsea-help/90a4fbf9-fdd8-4a9d-9a45-6bfd9cb6d637n%40googlegroups.com.