GSEA and genes ranked by q value

[afalvarez]

Hi guys,

Do you know if it's correct to use GSEA preranked tool with a .rnk file composed of genes ranked by q value? (rather than fold increase, for instance). If so, are there any parameters that need to be fixed?

Thanks a lot!

[afalvarez]

From GSEA manual

"For instance, you might have used your favorite tTest-like statistic to produce a ranked ordered gene list from your dataset which you now want to test for enrichment."

The question is, what should we include as numeric value for the second column (first corresponds to identifiers) in the .rnk file?

[Eric Katagirya]

Same question here. Do you use the q-values as the weight (class difference metric?). I use the q-value but I am not sure if I can trust the results. I will appreciate any clarification.