How can I do the "fdr" mutiple test correction to the GO enrichment results from "TopGO" ?

Question

[dukecomeback]

It seems that the multiple test for the GO enrichment result is important. However, the TopGO didn't do that, so I decide to do that by myself. The delemma looks like this: fdr correction with p.adjust() in R need the complete GO enrichment result, whereas the GenTable(,topNodes =N) in TopGO only give out the top N result. So how can I do the "fdr" mutiple test correction to the GO enrichment results from "TopGO"?

Sincerely

[Jean-Karim Heriche]

p.adjust only needs a vector of p-values which you can get from the TopGO results. Read the doc: http://www.bioconductor.org/packages/release/bioc/vignettes/topGO/inst/doc/topGO.pdf

Answer 1

[komal.rathi]

Instead of specifying topNodes to some value N, you can get all available GO Terms using:

allGO = usedGO(object = GOdata) 
# use it in GenTable as follows:
GenTable(GOdata, ... ,topNodes = length(allGO))

Then you can go about doing your p-value adjustment.

[dukecomeback]

Thank you for all of your help. Finally, the topNodes = length(allGO) from komal.rathi helped me out.