'ATH1121501.db' Dear all, Can anyone explain me why in the Arabidopsis annotation package 'ath1121501.db' there are no Entrez.Gene IDs, while in the other similar they are present. It creates me the problem as I need to realize an analysis of the results from a microarray chip using GOstats, e.g. hypergeometric test. For this I need to create a 'GOHyperGParams' object, which has as the argument the name of the annotation package, from which it takes the data regarding Entrez.Gene.Ids and correspondent GO annotation. Once Entrez is not present it returns me the following error ïÛÉÂËÁ × getUniverseHelper(probes, datPkg, entrezIds) : After filtering, there are no valid IDs that can be used as the Gene universe. Check input values to confirm they are the same type as the central ID used by your annotation package. For chip packages, this will still mean the central GENE identifier used by the package (NOT the probe IDs). for creating the 'universeGeneIds' I used Entrez Ids from the whole chip, but I got them from NetAffx, because I couldnot from 'ath1121501.db' . Also, I consulted the "Bioconductor Case Studies" book, where all the examples are for 'hgu95av2.db', which contains the posibility to extract ENTREZIDs. Please, let me know anyone if I am wrong in something, or what can I do. If there is another possibility to use an alternative of GOstats for the aim of performing hypergeometric test. Regards, Alex from Moldova -- Laboratorul Bioinformatica Centrul Biologe Moleculara Universitatea Academiei de Stiinte din Moldova [[alternative HTML version deleted]]

Hi Alex, The reason that entrez gene IDs are absent is because this package is based primarily on resources from Tair instead of NCBI. This was a preference that was voice from the community of scientists that work on Arabidopsis. However, you should be able to make good use of GOstats anyways. You will just want to specfify your gene list and gene universe using appropriate identifiers (Tair identifiers). Here is an example of how that might look: library(GOstats) library(ath1121501.db) ## For the sake of demonstration, lets just make up a universe from the annotations: universe = mappedRkeys(ath1121501ACCNUM); IDs = universe[1:400] params <- new("GOHyperGParams", geneIds = IDs, universeGeneIds = universe, annotation = "ath1121501", ontology="BP", pvalueCutoff=0.05, conditional=FALSE, testDirection = "over") ## then run hyperGTest etc. Over <- hyperGTest(params) summary(Over) Now if you *really* want entrez gene IDs for arabidopsis, you can get those too! You just have to look for them in the org.At.tair.db package (you already have this as it is loaded when you load ath1121501.db). So for example, if you do the following you can peak at some entrez gene IDs from this package like this: toTable(org.At.tairENTREZID[1:6]) Let me know if you have further questions, Marc On 04/11/2011 07:51 AM, UnASM Lab_bi wrote: > 'ATH1121501.db' > Dear all, > Can anyone explain me why in the Arabidopsis annotation package > 'ath1121501.db' there are no Entrez.Gene IDs, while in the other similar > they are present. > It creates me the problem as I need to realize an analysis of the results > from a microarray chip using GOstats, e.g. hypergeometric test. For this I > need to create a 'GOHyperGParams' object, which has as the argument the name > of the annotation package, from which it takes the data regarding > Entrez.Gene.Ids and correspondent GO annotation. Once Entrez is not present > it returns me the following error > > ïÛÉÂËÁ × getUniverseHelper(probes, datPkg, entrezIds) : > After filtering, there are no valid IDs that can be used as the Gene > universe. > Check input values to confirm they are the same type as the central ID used > by your annotation package. > For chip packages, this will still mean the central GENE identifier used by > the package (NOT the probe IDs). > > for creating the 'universeGeneIds' I used Entrez Ids from the whole chip, > but I got them from NetAffx, because I couldnot from 'ath1121501.db' . > > Also, I consulted the "Bioconductor Case Studies" book, where all the > examples are for 'hgu95av2.db', which contains the posibility to extract > ENTREZIDs. > > Please, let me know anyone if I am wrong in something, or what can I do. If > there is another possibility to use an alternative of GOstats for the aim of > performing hypergeometric test. > > Regards, > Alex from Moldova > > > > _______________________________________________ > 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]]