Multiple testing correction with GOstats in R

Question

[YanO]

I've carried out a GO analysis in R using GOstats, and have a list of p-values. However, as I understand it they have not been corrected for multiple testing, and I cannot see a way to do it in GOstats.

Did I miss something is there a way to do it in GOstats?

I understand why a multiple testing correction may not be appropriate, but all of the web-browser GO programs offer it, and I'd like to see what my results look like with it. I can't seem to find a consensus over whether GO p-values should be corrected or not.

Is there an alternative R package I should use that does offer correction, or is there a way to correct the p-values I have as an output of GOstats?

(I realise there has been many many posts on GO but none have given me exactly what I'm looking for.)

Thanks!

Answer 1

[raunakms]

If you already have the p-values then you can simply use p.adjust command from the R base package

http://stat.ethz.ch/R-manual/R-patched/library/stats/html/p.adjust.html

[YanO]

Very helpful, thanks! But what is my n here? The length of values that have p<0.05?

Answer 2

[EagleEye]

This might be useful to you. Gene Set Clustering based on Functional annotation (GeneSCF)