Hi, has anyone had any problems with the annotation or codelink data once exported from the Codelink App, i.e. changing accession IDs. I apologise in advance for not having the codelink version number to hand but can't access my laptop at the moment. I am running the Codelink app to normalise my Human whole genome codelink arrays. In the first week in June I updated R, Bioconductor, and Codelink to get the latest versions on my Mac OSX10.5. One month in I have discovered that some very important genes are missing from my "writecodelink" export file. More accurataly, it appears that the latest annotaion database has re-assigned over 8000 of the IDs. Some are an improvement, from NAs to a database number but many have re- assigned a NM_ id to an est ID such as BG... or AK.... There is no wrong gene asignments but the est IDs are not recognised by my downstream analysis software. Is there a reason for this or is it a database glitch Matt

Hi Matt Your issue is probably not related to the codelink package but to the annotation packages for Codelink arrays. There is a recent thread with a similar question that has been very well answered already: http://article.gmane.org/gmane.science.biology.informatics.conductor/2 4007 I can add to the answer that the manufacturer's assignments for Codelink arrays had not changed in the last two years (at least) and therefore your problem is likely an effect of updated annotations at NCBI. I guess that in the output of writeCodelink() you obtain the same results in both analyses (i.e. in terms of probes differentially expressed) but with different annotations. In any case, a more detailed description of the issue with some reproducible example and the output of sessionInfo() is always welcomed. Diego On Mon, Jul 6, 2009 at 8:39 PM, Matthew Neville<matthew.neville at="" oxlip.ox.ac.uk=""> wrote: > Hi, > > has anyone had any problems with the annotation or codelink data once > exported from the Codelink App, i.e. changing accession IDs. > > I apologise in advance for not having the codelink version number to hand > but can't access my laptop at the moment. > I am running the Codelink app to normalise my Human whole genome codelink > arrays. In the first week in June I updated R, Bioconductor, and Codelink > to get the latest versions on my Mac OSX10.5. One month in I have > discovered that some very important genes are missing from my > "writecodelink" export file. More accurataly, it appears that the latest > annotaion database has re-assigned over 8000 of the IDs. Some are an > improvement, from NAs to a database number but many have re-assigned a NM_ > id to an est ID such as BG... or AK.... There is no wrong gene asignments > but the est IDs are not recognised by my downstream analysis software. Is > there a reason for this or is it a database glitch > > Matt > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > -- Diego Diez, Ph Bioinformatics center, Institute for Chemical Research, Kyoto University. Gokasho, Uji, Kyoto 611-0011 JAPAN diez at kuicr.kyoto-u.ac.jp