Hi, I am trying to assign Gene Ontology (GO) Biological Process terms to proteins so that I can decompose a protein interaction network into GO-derived modules. Each module should contain proteins that are all annotated to the same biological process (overlapping modules are fine). The challenge is, I want to select GO terms (modules) which are not overly "redundant" terms/modules. What is the best way to do this?
I tried using GoTermFinder, which annotates proteins to GO annotations, and then ReviGo, which removes redundant terms based on semantic similarity. But I've found that ReviGo is pretty limited, and we're still getting pretty generic terms.
GoTermFinder: http://go.princeton.edu/cgi-bin/GOTermFinder
ReviGo: http://revigo.irb.hr/
Can anyone suggest another tool that can help with this?
Please note: I don't think enrichment-based tools will help because I want to decompose the entire proteome network into modules, not a specific subset that is active in some condition, etc.
Thanks!