Gene Ontology Tools Widely Used

Question

[alittleboy]

Hi biologists:

May I know the currently widely used tools for GO enrichment analysis? I have some experience in GOseq, topGO, GOstat and Ontologizer, but I am not sure which one would you prefer?

There are many reasons to choose one over another tool, such as adjusting for gene length using GOseq and accounting for parent-child relationship using Ontologizer. It seems that the results from these different approaches are quite different, and no single tool can account for everything.

Thanks for sharing your experience on the selection of GO tools :)

Answer 1

[Khader Shameer]

I prefer Ontologizer 2 for detailed GO term analysis and DAVID for quick analysis. Having said that Ontologizer is having only 69 citations, but DAVID seems to have 2000 citations (mentioned in DAVID website). So if you are looking for popularity in terms of citation, you may run a quick citation analysis using various GO enrichment tools.

[boczniak767]

I also prefer Ontologizer, it has both command-line and webstart versions. I used to use command-line version due to greater set of options but now webstart version is even better. Another plus is that it offers many methods including tree-structure correction.

[Czh3]

Yes, command line version of Ontologizer is really useful for large sample set. how to set the parameter of Ontologizer? It seems getting very different result when use different parameter(especially -m, -c)

[boczniak767]

Depending on setting of these options, the program uses different computation method (-m) and multiple comparison p-value correction (-c). The choice depends on your preference of each or just popularity in literature.

Answer 2

[Tomasz]

ontocat R package - http://www.ontocat.org/wiki/r

Here's an example how it's used for gene enrichment test http://www.ontocat.org/browser/trunk/ontoCAT/src/uk/ac/ebi/ontocat/examples/R/Example1.R

Answer 3

[ALchEmiXt]

Hi you might want to take a look at the R driven bioconductor packages for GO analysis. Most important I remembered from a collegue studying this was the correction for multiple independant vs dependant relationships within GO; I think its roughly explained in this paper of Haisheng Nie. I am definitly not an expert but GOstats has been communicated to us as the way to go.

On aside, we often use the excellent DAVID webservice for our data analysis as well. Not sure what they use for enrichment analysis exactly. But they do it for GO as well as for interpro.

My 2ct.

[alittleboy]

Thank you for your comments! Can I know when you use GOstats, do you consider the dependencies of the GO terms? Or traditional term-by-term tests will suffice?

[ALchEmiXt]

Currently for convenience we mostly use the DAVID webservice since it in addition to GO it also considers other "pathway" linked enrichments. But I think it is crucial to consider the dependencies.

Answer 4

[fransua]

one quite user friendly tool that account for many problems is Blast2GO, enrichment test, annotation redundance, it loads the sequences...: http://www.blast2go.org/

Answer 5

[Iain]

I really like the GOrilla web service.

And you can visualize the output using the ReviGO server.

[madkitty]

How do you visualize the output in revigo? I don't seem to be able to find how to export the result file from gorilla..