Hi, I am working with the Bioconductor annotation package for the Arabidopsis thaliana GeneChip ATH1 and have found a strange difference between version 2.7 and versions 2.8 or 2.9. In 2.7: source("http://bioconductor.org/biocLite.R") biocLite("ath1121501.db") library("ath1121501.db") length(as.list(toggleProbes(ath1121501ACCNUM,"all"))) produces the following output: [1] 22810 In 2.8 or 2.9 the same code produces the following output: [1] 22809 Comparing the lists of probe_ids shows that the probe_id 245651_s_at is present in 2.7 and not present in 2.8 or 2.9. I know that the annotations for a given probe_id can change with updates but why would a probe_id be removed? This probe_id is still present in the latest annotation files provided by Affymetrix. Thanks for any idea. Stefan Bleuler

Hi Stefan, I have done some detective work and tracked this down for you. The basic problem here seems to be that for this one probe there is no data whatsoever coming from tair for ANY of the genes that map to it (probeset "245651_s_at"). According the sources I have, affy matches this probeset matches 5 loci IDs and TAIR does not provide data for any of them. I agree with you though that the probe should still be present and will look into a correction for this. Marc On 02/13/2012 12:23 AM, Stefan Bleuler wrote: > Hi, > > I am working with the Bioconductor annotation package for the > Arabidopsis thaliana GeneChip ATH1 and have found a strange difference > between version 2.7 and versions 2.8 or 2.9. > > In 2.7: > source("http://bioconductor.org/biocLite.R") > biocLite("ath1121501.db") > library("ath1121501.db") > length(as.list(toggleProbes(ath1121501ACCNUM,"all"))) > > produces the following output: > [1] 22810 > > In 2.8 or 2.9 the same code produces the following output: > [1] 22809 > > Comparing the lists of probe_ids shows that the probe_id 245651_s_at > is present in 2.7 and not present in 2.8 or 2.9. I know that the > annotations for a given probe_id can change with updates but why would > a probe_id be removed? This probe_id is still present in the latest > annotation files provided by Affymetrix. > > Thanks for any idea. > > Stefan Bleuler > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor

Hi Marc Thank your very much for your research. A fix would be great. In the meantime I will just treat this probe_id as having no annotation in any category. Stefan > Hi Stefan, > > I have done some detective work and tracked this down for you. The > basic problem here seems to be that for this one probe there is no data > whatsoever coming from tair for ANY of the genes that map to it > (probeset "245651_s_at"). According the sources I have, affy matches > this probeset matches 5 loci IDs and TAIR does not provide data for any > of them. I agree with you though that the probe should still be present > and will look into a correction for this. > > Marc > > On 02/13/2012 12:23 AM, Stefan Bleuler wrote: >> Hi, >> >> I am working with the Bioconductor annotation package for the >> Arabidopsis thaliana GeneChip ATH1 and have found a strange difference >> between version 2.7 and versions 2.8 or 2.9. >> >> In 2.7: >> source("http://bioconductor.org/biocLite.R") >> biocLite("ath1121501.db") >> library("ath1121501.db") >> length(as.list(toggleProbes(ath1121501ACCNUM,"all"))) >> >> produces the following output: >> [1] 22810 >> >> In 2.8 or 2.9 the same code produces the following output: >> [1] 22809 >> >> Comparing the lists of probe_ids shows that the probe_id 245651_s_at >> is present in 2.7 and not present in 2.8 or 2.9. I know that the >> annotations for a given probe_id can change with updates but why would >> a probe_id be removed? This probe_id is still present in the latest >> annotation files provided by Affymetrix. >> >> Thanks for any idea. >> >> Stefan Bleuler >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor at ... >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: >> http://news.gmane.org/gmane.science.biology.informatics.conductor > > _______________________________________________ > Bioconductor mailing list > Bioconductor at ... > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor