Wednesday, July 15, 2009

Gene ontology tool suggestions

I have used a few GO tools in the past, but after looking at the massive list of tools on the gene ontology page I'm hoping someone can give me a good suggestion for my problem.

Basically, I have several lists of GO terms (~4-15 terms per list) and I would like to see if at a "higher" branch they share a common molecular function. Ideally, a tool that could be run from the command line and outputs significance scores would be great, but a GUI tool would also work since I have about 70 lists that I would need to run.

Note, that this is slightly different than the usual over-representation analysis which usually takes a list of genes as input. In my problem I am starting with GO terms.

Any suggestions would really be welcome!

3 comments:

Leon French said...

Hey, I'm not sure about how'd you go about it but I'm biased towards ErmineJ - http://www.chibi.ubc.ca/ermineJ/ . I've never used it, but usually the input is genes or probes.

Hope that helps,

Morgan Langille said...

After looking at lots of tools, I realized that what I was looking for was not really the norm. Basically, I want to check the semantic similarity of any two GO terms. The one (and only) tool that I came across is G-SESAME. Unfortunately, the only way to run multiple queries is to send http requests (ick)! I hate using closed-source bioinformatic tools, but I don't have many options.

Benjamin Good said...

Personally I'd much rather run multiple http requests (assuming the server i'm talking to is reliable) then go through the hassle of installing some else's code and using my own CPU/hard drive space..

But, to answer your question, you will probably find the following review helpful.
http://www.ploscompbiol.org/article/info%3Adoi%2F10.1371%2Fjournal.pcbi.1000443