GSEA for multiple groups

[Mohamad Bairakdar]

Hi,

Hope you're doing well. I have 3 groups (control, treated 1, treated 2) for which I obtained log fold changes between the different conditions and would like to run GSEA to compare all of them (so that I can get gene sets that are enriched in each of the conditions). Would I just have to consider them pairwise? More importantly, would I have to correct for multiple hypothesis testing?

Best,

Mohamad

[Anthony Castanza]

Hi Mohamad,

GSEA just has support for two-phenotype analysis by default. You can run the various comparisons pairwise, but we don't have any built in support for adjusting for multiple hypothesis testing. You can visualize multiple GSEA runs simultaneously in EnrichmentMap if you load the GSEA results from the menu within Cytoscape rather than using the built in view in GSEA but that's more just a visualization tool that will help with looking at the results.

I'm not sure if there's a consensus on correction for multiple hypothesis testing for comparing multiple GSEA runs, possibly a Bonferroni correction for the number of pairwise comparisons.

-Anthony

Anthony S. Castanza, PhD

Curator, Molecular Signatures Database

Mesirov Lab, Department of Medicine

University of California, San Diego