Does GOstats give the possibility to calculate the corrected Pvalue in addition to the enrichment Pvalue ? Sometimes in some Cytoscape plugins we find extra pvalues generally Benferroni correction, is it possible with GOstats ? Regards Radhouane 2011/5/5 Marc Carlson <mcarlson@fhcrc.org> > Hi Radhouane, > > As a starting point, you might want to read the vignette for the GOstats > package called "Hypergeometric Tests Using GOstats": > > > http://www.bioconductor.org/packages/2.8/bioc/vignettes/GOstats/inst /doc/GOstatsHyperG.pdf > > If you are starting with Ensembl gene IDs, you can easily convert those to > entrez gene IDs, and then use that with GOstats to run a set of > hypergeometric tests for each organism. Once you get that done, > interpreting the other comparisons you suggested sound a bit more murky > because different organisms have been annotated differently depending on how > well studied they are. > > > Marc > > > > On 05/05/2011 09:25 AM, Radhouane Aniba wrote: > >> Thanks Tim, >> >> Actually I have a predicted list of miRNA binding sites targetting >> specific >> genes in 4 genomes. >> >> What I am trying to find is the characteristics of these Genes {g} for a >> specific genome {G} in term of GO enrichement and KEGG enrichment. >> >> My starting point is a file with a list of ENSEMBL IDs, just that, no >> annotation and no scores, just the gene names. >> >> I am looking for the right package to do that, topGO for example seems to >> not accept only gene names, annotation is needed as well as other details, >> I >> am acually reading about all these packages that make gene enrichment >> analyses. >> >> Rad >> >> >> 2011/5/5 Tim Triche, Jr.<tim.triche@gmail.com> >> >> Hi Radhouane, >>> >>> You can get more specific answers if you ask more specific questions. >>> The >>> mathematical formulation of the test(s), and therefore the meaning of >>> your >>> results, will depend directly on >>> >>> 1) the logic of the package you use to test for GO enrichment in a gene >>> list or lists >>> 2) the logic of the package you use to test for KEGG enrichment in a gene >>> list or lists >>> >>> A concise and useful description of the logical basis for hypergeometric >>> and binomial tests: >>> >>> >>> >>> http://great.stanford.edu/help/index.php/Statistics#What_is_the_hy pergeometric_test_formally.3F >>> >>> You mention GSEA simultaneously with GO/KEGG enrichment, thus perhaps it >>> would be best if you provide examples making your question more concrete, >>> so >>> that others may benefit. For example, the logic behind the >>> similarly-named >>> GSA and GSEA procedures differs subtly. On that note, you might find the >>> following two discussions helpful: >>> >>> >>> >>> http://www.broadinstitute.org/cancer/software/gsea/wiki/index.php/ FAQ#What_is_the_difference_between_GSEA_and_an_overlap_statistic_.28hy pergeometric.29_analysis_tool.3F >>> >>> http://www-stat.stanford.edu/~tibs/ftp/GSA.pdf >>> >>> If you haven't already, you will want to read the original GSEA paper in >>> PNAS for background. >>> >>> Best regards, >>> >>> --t >>> >>> >>> On Thu, May 5, 2011 at 8:20 AM, Radhouane Aniba<aradwen@gmail.com> >>> wrote: >>> >>> Hello everyone, >>>> >>>> Well I am aware of some packages in Bioconductor that are useful for >>>> measuring the GO or KEGG gene enrichment in a given file for a given >>>> genome, >>>> GOstats, GSEA etc .. >>>> >>>> My question is : I am working with 4 differents genomes, I have gene >>>> lists >>>> for each of them, and I want for a given gene for each list for each >>>> geneome >>>> : >>>> >>>> - Extract the GO with its pvalue (what does the pvalue actually mean >>>> here, >>>> enrichment ok but how is it calculated ? ) >>>> - Extract the KEGG pathway and its pvalue as well >>>> >>>> Thanks >>>> >>>> Rad >>>> >>>> [[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 >>>> >>>> >>> >>> -- >>> If people do not believe that mathematics is simple, it is only because >>> they do not realize how complicated life is. >>> John von Neumann< >>> http://www-groups.dcs.st- and.ac.uk/%7Ehistory/Biographies/Von_Neumann.html >>> > >>> >>> >>> >> > _______________________________________________ > 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 > -- *Radhouane Aniba* *Bioinformatics Postdoctoral Research Scientist* *Institute for Advanced Computer Studies Center for Bioinformatics and Computational Biology* *(CBCB)* *University of Maryland, College Park MD 20742* [[alternative HTML version deleted]]

2011/5/5 Steve Lianoglou <mailinglist.honeypot@gmail.com> > Hi, > > On Thu, May 5, 2011 at 12:25 PM, Radhouane Aniba <aradwen@gmail.com> > wrote: > > Thanks Tim, > > > > Actually I have a predicted list of miRNA binding sites targetting > specific > > genes in 4 genomes. > > > > What I am trying to find is the characteristics of these Genes {g} for a > > specific genome {G} in term of GO enrichement and KEGG enrichment. > > > > My starting point is a file with a list of ENSEMBL IDs, just that, no > > annotation and no scores, just the gene names. > > > > I am looking for the right package to do that, topGO for example seems to > > not accept only gene names, annotation is needed as well as other > details, I > > am acually reading about all these packages that make gene enrichment > > analyses. > > It seems as if you can use GOstats for this purpose, but you'd first > need to convert these IDs to entrez id (which you can do using the > appropriate org.*.eg.db pacakge (org.Hs.eg.db if we're talking about > human, for example). > > In addition to this 'target list' you have, you'll need a appropriate > way to define "the universe" of gene id's to use for testing. > What is the universe of gene ids ? what does is it means ? > > Given those two lists (targets and universe) you'll be able to run > GOstats to get GO enrichment for your targets. Incidentally, you > should also be able to use GOstats for KEGG ... see the vignettes > here: > > http://www.bioconductor.org/packages/release/bioc/html/GOstats.html > > -steve > > -- > Steve Lianoglou > Graduate Student: Computational Systems Biology > | Memorial Sloan-Kettering Cancer Center > | Weill Medical College of Cornell University > Contact Info: http://cbio.mskcc.org/~lianos/contact > -- *Radhouane Aniba* *Bioinformatics Postdoctoral Research Scientist* *Institute for Advanced Computer Studies Center for Bioinformatics and Computational Biology* *(CBCB)* *University of Maryland, College Park MD 20742* [[alternative HTML version deleted]]