GO enrichment analysis with prob IDs

Question

[zizigolu]

Hey friends,

I did GO analysis with a list of probesets, may you tell me please if the result is trustable or I need the genes matched with these probes to perform GO? I converted affy ids to gene, protein, etc. but for each prob there is more than one entry. Anyway can I trust on my result with prob IDs please?

Answer 1

[andrew]

How did you do the GO analysis? Directly from the Probeset ID? How did you translate the probeset to gene IDs?

It's difficult to say if you can rely on your results. Yes, probes often map to one or more genes - and this can change over time. It is important to us the manufacturers latest annotations for this reason.

In our software, iPathwayGuide, we will take the raw affy file, QC, normalize, then get DEGs, and perform GO analysis and more. When a probeset ID is annotated to multiple genes, we discard it because we cannot assess which gene it represents. However, there are other approaches. That just happens to be what we use.

[zizigolu]

By AGrigo I did GO analysis and tried to get the GO terms, what is your suggestion? iPathway doesn't work in plants