Same zero cross value

[Samer Olleik]

Dear GSEA team,

I am encountering a problem with the GSEA run, I have a list of about 56000 genes but with all the gene sets that I have in addition to Hallmarks MM I noticed that I am getting a p value of 0.00 while the zero cross remains constant 15681. I tried to change the permutation number, unfortunately, still getting the same answer. Please find attached my input files including the gene sets. Would you please help me to solve this issue.

Regards,

[Anthony Castanza]

Hi Samer,

The zero cross is a feature of the dataset itself and isn't expected to change for a given dataset at any point. The pValue is derived from the permutation matrix though. What did you change your permutation value to? For a really extreme score it might not be possible for GSEA to calculate a precise value. If you're running in gene set permutation mode you can try 10000 permutations or more but substantially larger numbers could be an infeasibly long runtime, in which case you can simply report the pValue as <1/nperm (or more precisely, <1/same sign nperm, but that number is more difficult to pull out of the software)

-Anthony

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

--
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 on the web visit https://groups.google.com/d/msgid/gsea-help/ddc554fc-e756-42b2-ae90-3563f599d6afn%40googlegroups.com.

[Samer Olleik]

Hello Anthony, 

You cannot imagine how glad I am to receive your answer, 

I tried to change the permutation number till 10000, but still getting the norm p value of 0.00 and the same zero cross value.

Based on your answer, this result does not show an issue, am I right? What I have to do is to report the norm p-value in the way you mentioned, would you please confirm me this.

Many thanks, and best regards,

To view this discussion on the web visit https://groups.google.com/d/msgid/gsea-help/CAGCeyZzRyPsXUfQXKDpeOZfMMEs3WT8hxeV2giv1Ga%3Doo-XPBg%40mail.gmail.com.

[Anthony Castanza]

Hi Samer,

Yes, this sounds in line with normal GSEA behavior 

-Anthony

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

To view this discussion on the web visit https://groups.google.com/d/msgid/gsea-help/CAPWAaYQ6j9afAdgnQM5FnXp%2BVjJdHNnk7vyaBm%3DhJm3T7iwb%3Dw%40mail.gmail.com.

[Samer Olleik]

Dear Anthony,

I appreciate your support and help.

Regards,

To view this discussion on the web visit https://groups.google.com/d/msgid/gsea-help/CAGCeyZxx98HYSj6SMyTnX%3Dguvb8JnVHQH%3D72sct2mLmdznJOtw%40mail.gmail.com.