Identifying enriched pathways that overlap among a range of databases

Question

[Tash]

Hi everyone,

I have performed pathway analyses with the same filtered gene list using different gene sets. The next step is to identify overlapping enriched pathways in the outputs of the databases used, such as IPA, KEGG, Reactome (as I assume they might have different denominations). I'm curious to know if there is any smart way of doing this, or would I have to do this manually?

Thanks!

Answer 1

[Martombo]

You could definitely retrieve the gene lists associated with every geneset and compute the overlap between them, but I'm not sure that can help you much. I think what people normally do is to use comprehensive genesets, like the ones present in MSigDB. An algorithm that I know is taking direct care of the overlap between genesets is SetRank, or also topGO for gene ontology.

[Tash]

Thanks for this. You're right, computing the overlap wouldn't really help me.

Having a quick look at the SetRank algorithm on the other hand tells me that it would probably do what I wish to do. However, I want something that I can use on my already generated outputs (if possible). I will keep SetRank in mind for the future as it seems to be a good resource.