Hypergeometric test in GSEA

Question

[Ron]

Hi,

I want to do GSEA between two categories of tumors.I am interested in using hypergeometric tool in GSEA.

I am not able to find if its present in its GUI version.Please let me know.This is for getting enrichment of pathways.

Also,I want to do enrichment only for down regulated genes with logfoldchange < -2.Can I do GSEA preranked?

http://www.broadinstitute.org/cancer/software/gsea/wiki/index.php/FAQ#What_is_the_difference_between_GSEA_and_an_overlap_statistic_.28hypergeometric.29_analysis_tool.3F

Thanks,
Ron

[Devon Ryan]

Using a hypergeometric test kind of defeats the entire purpose of the GSEA package.

[russhh]

I see a value in being able to compare the GSEA results to hypergeo test results based on precisely the same annotations as used by GSEA. I don't know how to do it though

[maciek.kandula]

If you are using KEGG gene sets then you can try with KEGGprofile R package. It performs the hypergeometric test. Their KEGG database might be a bit outdated so you might have to filter out some of the resulting pathways to end up with the same set as in GSEA.

[Devon Ryan]

Sure, doing a comparison makes sense. The tricky part would be getting all of the gene set information from GSEA, which I've never needed to do so can't comment on.

[Zhilong Jia]

Agree with Devon'. GSEA used k-s test. Maybe you just meant to use the gene sets from Msigdb. and hyper-geometric test for the down regulated genes only. cogena is the one you probably interested in.

[EagleEye]

try this 1) Gene Set Clustering based on Functional annotation (GeneSCF) 2) A: GO enrichment analysis using a Text file with all the genes and GO ids associat