Hi Paul, That makes a lot more sense. To do that you will probably want to use the UNIPROT mappings that can be found in the organism packages (like org.Hs.eg.db) to modify your gene universe etc. Thus enabling you to, as needed, replace entrez gene IDs with uniprot IDs. There are instructions for that 1st part in the AnnotationDbi package and it should be straightforward but if it is not obvious I can help. But then you will have a situation that is not unlike that faced by people who use less popular model organisms. Ie, you will have things mapped to uniprot ID space and that will be *as if* you had an organism that we don't have pre-made annotation packages for. And traditionally, packages like GOstats and Category have expected you to have a package. But fortunately for you, I think we still have a way to help you since we worked out how to help people in that situation once before (and wrote a vignette about it). So you will want to have a look at the vignette that we wrote for the GOstats package called "Hypergeometric tests for less common model organisms". You can find it here: http://www.bioconductor.org/help/bioc- views/release/bioc/html/GOstats.html I think that if you do what is prescribed in that vignette, it should basically get the job done for you. You will need to follow the example and just pretend that you are working on a less popular organism once you have uniprot. But let me know if there are further difficulties or if I have overlooked something. Marc On 02/10/2011 11:08 AM, Paul Rigor wrote: > Hi Marc, > > Thank you for the response. Sorry for the confusion, but what I > actually wanted to do was perform enrichment analysis using uniprot > accessions instead of entrez id's. There's another package, topGO, > which allows one to specify a universe of id's, the mapping between > id's and their GO terms and a list of interesting genes (with their > associated uniprot id's). > > I believe that package also uses GOdb to resolve the topology for each > ontology. However, I would like to use the hypergeometric test instead > of their available suite of tests. > > Hope that clears things up! > > Thanks, > Paul > > On Thu, Feb 10, 2011 at 11:00 AM, Marc Carlson <mcarlson@fhcrc.org> <mailto:mcarlson@fhcrc.org>> wrote: > > Hi Paul, > > We don't presently have that capability built into our Category > testing > functionality. We can test, GO, KEGG and I believe even PFAM, but > there > is nothing presently there for Uniprot. But I must also admit > that your > request surprised me a bit because Uniprots are accessions and do not > (as I understand them) represent "families" or "categories" that genes > could be grouped into. So I am a little bit perplexed about how they > would be used to represent categories of genes? Is there possibly > some > other information inside of a more complete uniprot record that can be > used in this way and which I am overlooking? If so, then that > represents another challenge that will have to be overcome in > order for > what you ask to become possible. > > > Marc > > > > On 02/08/2011 04:24 PM, Paul Rigor wrote: > > Hello, > > > > How would I configure a hypergeom object to use uniprot ids for > both the > > universe of genes and the list of interesting genes? > > > > Thanks! > > Paul > > > > [[alternative HTML version deleted]] > > > > _______________________________________________ > > Bioconductor mailing list > > Bioconductor@r-project.org <mailto:bioconductor@r-project.org> > > https://stat.ethz.ch/mailman/listinfo/bioconductor > > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > > > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org <mailto:bioconductor@r-project.org> > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > > [[alternative HTML version deleted]]