Hi Olivier, The information you are asking about is more about the Gene Ontology itself, which is (primarily) in the GO.db package. This package has information about each GO term, which onotology it is from (mf, bp or cc) and also who its parents/children are etc. It is possible to move up or down the DAG from any GO term in this way. ## So for example, suppose you were interested in MAL13P1.11, and you saw that it had the GO term GO:0042777 associated with it by looking at the org.Pf.plasmoGO mapping: library("org.Pf.plasmo.db") get("MAL13P1.11", org.Pf.plasmoGO)[[1]] ## From this alone you already know that it is from the bp ontology, but now you want to know what the parent terms are for this term so you load GO.db and can check the PARENTS mapping for BP terms like this: library(GO.db) get("GO:0042777", GOBPPARENTS) ## Or maybe you want to know more about the term itself? Like the description for it? Term("GO:0042777") ## Or maybe the full definition Definition("GO:0042777") ## Or it's child terms (this term is actually "terminal", so there are no children) get("GO:0042777", GOBPCHILDREN) ## Or maybe I want ALL of the ancestor terms for it? get("GO:0042777", GOBPANCESTOR) Anyhow, all this sort of GO-centered information is in GO.db. Hope this helps, Marc On 09/02/2011 07:15 AM, Olivier Lucas wrote: > Dear Marc, > > I apologize for my confusing post. What I meant is that I wanted to restrict the search to MF and a specific level. Indeed, I do understand that many genes will have multiple GO association in MF, even if restricted to one specific level. I was just wondering if there is any way to know the association type (MF, BP, CC and level) when retrieving data through biomaRt or in the eg.Pf.plasmo.org package. > In biomaRt, for plasmodium, the relevant attribute in listAttributes() I can find is go_id which does not provide a key to which ontology it is from, neither its level. I don't believe the .org package has that information either. > No worries, it will work as is, I was just wondering if I could get a finer mapping. I may try the goProfiles package as I am interested in comparison. > > Thanks! > > Olivier > ------------------------------ > > Message: 16 > Date: Thu, 01 Sep 2011 14:58:35 -0700 > From: Marc Carlson<mcarlson@fhcrc.org> > To: bioconductor@r-project.org > Subject: Re: [BioC] Simple GO pie and GOSim install error > Message-ID:<4E60000B.3030901@fhcrc.org> > Content-Type: text/plain > > Hi Olivier, > > I worry that you might be confused about how GO annotations work. Many > genes will also be associated with multiple different terms, even from > within the same ontology. The only limit to how many terms can be used > to describe a gene is how many of them actually apply, and many genes > can have more than one kind of function or location in a cell. So even > if you restricted down to just one level of the tree, you still could > have more than one GO term associated with a given gene. Also, for the > org packages at least the terms that will be in the standard GO mapping > should be the ones that are the most appropriate for describing that > gene. That is unless you use the "GO2ALL" mapping, you should not get > all of the DAG levels for each gene. > > I hope this helps. Please let me know if I have misunderstood you, > > Marc > > > > > On 08/29/2011 04:26 PM, Olivier Lucas wrote: >> Dear list, >> >> I only need to compare GO terms associated with entrezID and >> I thought GOSim would be the easiest way to obtain one and only one GO >> ID (MF or BP) per entrezID. BiomaRt returns me many levels of GO annotation and I could not figure out how to keep, for example, MF level 2 only (i.e. I get 2442 GO from 642 genes, I want 642 max (some won't have any MF level 2 I think)). >> My organism is Plasmodium. I also tried through the org.Pf.plasmo.db package too, but I don't think there is a way to select a specific GO level there. >> In the end, I only need to generate 3 descriptive pie charts from 3 genelists of entrezID. >> >> I'm trying to install GOSim, without success. I with the R under development version and R2.13.1 but it does not appear to be available in either of those version? Is it still supported and working or, most likely, am I making a mistake somewhere? >> >>> source("http://www.bioconductor.org/biocLite.R") >> BioC_mirror = http://bioconductor.org >> Change using chooseBioCmirror(). >> Warning messages: >> 1: In safeSource() : Redefining âEUR~biocinstallâEUR^(TM) >> 2: In safeSource() : Redefining âEUR~biocinstallPkgGroupsâEUR^(TM) >> 3: In safeSource() : Redefining âEUR~biocinstallReposâEUR^(TM) >>> biocLite("GOSim") >> Using R version 2.13.1, biocinstall version 2.8.4. >> Installing Bioconductor version 2.8 packages: >> [1] "GOSim" >> Please wait... >> >> Warning: unable to access index for repository http://brainarray.mb ni.med.umich.edu/bioc/bin/macosx/leopard/contrib/2.13 >> Warning message: >> In getDependencies(pkgs, dependencies, available, lib) : >>  package âEUR~GOSimâEUR^(TM) is not available (for R version 2.13.1) >> >>> install.packages("GOSim", repos = "http://cran.r-project.org/web/packages/GOSim/", type="source") >> Warning: unable to access index for repository http://cran.r-project.org/web/packages/GOSim/src/contrib >> Warning message: >> In getDependencies(pkgs, dependencies, available, lib) : >>  package âEUR~GOSimâEUR^(TM) is not available (for R version 2.13.1) >>> sessionInfo() >> R version 2.13.1 (2011-07-08) >> Platform: x86_64-apple-darwin9.8.0/x86_64 (64-bit) >> >> locale: >> [1] en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8 >> >> attached base packages: >> [1] stats    graphics grDevices utils    datasets methods  base    >> >> other attached packages: >> [1] Biobase_2.12.2 >> >> loaded via a namespace (and not attached): >> [1] tools_2.13.1 >> Finally, I tried through the shell directly: >> >> MacBook-Pro:~ apple$ R CMD INSTALL GOSim_1.2.5.tar.gz >> * installing to library âEUR~/Library/Frameworks/R.framework/Versio ns/2.13/Resources/libraryâEUR^(TM) >> ERROR: dependencies âEUR~topGOâEUR^(TM), âEUR~flexmixâEUR^(TM) are not available for package âEUR~GOSimâEUR^(TM) >> * removing âEUR~/Library/Frameworks/R.framework/Versions/2.13/Resou rces/library/GOSimâEUR^(TM) >> MacBook-Pro:~ apple$ >> >> Any suggestion? >> Thanks! >> >> Olivier > [[alternative HTML version deleted]] > > > > _______________________________________________ > Bioconductor mailing list > 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]]