Dear list, What are the differences between these two packages (mogene10sttranscriptcluster.db and mogene10stprobeset.db)? As far as I know, the first annotation package is used when ST arrays are normalized at the "core" level and the second one when data is normalized at the "probeset" level. Is that right? Another question is related to the pd.mogene.1.0.st.v1 package, which is the annotation package used in conjunction with the oligo package. What are the differences between this package and the others (mogene10sttranscriptcluster.db and mogene10stprobeset.db)? Thanks, Javier

Hi Javier, On 9/27/2012 7:36 AM, Javier P?rez Florido wrote: > Dear list, > What are the differences between these two packages > (mogene10sttranscriptcluster.db and mogene10stprobeset.db)? > > As far as I know, the first annotation package is used when ST arrays > are normalized at the "core" level and the second one when data is > normalized at the "probeset" level. Is that right? If by core and probeset you are referring to arguments for rma() in oligo, then yes. Although to be pedantic I should correct you and say that you should replace normalize with summarize. Regardless of the summarization level, all data are normalized at the probe level. It is only at the summarization step that there are differences between the core and probeset arguments to rma(). In addition, a more satisfying description IMO is transcript and probeset, where probeset is roughly equivalent to exon (although some probesets span multiple exons). > > Another question is related to the pd.mogene.1.0.st.v1 package, which > is the annotation package used in conjunction with the oligo package. > What are the differences between this package and the others > (mogene10sttranscriptcluster.db and mogene10stprobeset.db)? Benilton has already answered this question. I would only add that the mogene* packages are equivalent to the 'regular' annotation packages that exist for the 3'-biased arrays, which map the Affy IDs to various other annotation sources, whereas the pd.mogene package is primarily intended for internal use by oligo (although I find it quite useful for doing things like removing the control probesets which have an annoying propensity for appearing in the list of differentially expressed genes). Best, Jim > > Thanks, > Javier > > _______________________________________________ > 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 -- James W. MacDonald, M.S. Biostatistician University of Washington Environmental and Occupational Health Sciences 4225 Roosevelt Way NE, # 100 Seattle WA 98105-6099

Re: your question about the pd.* packages... They are build from the official PGF / CLF / mps / NetAffx Annotation CSV files from Affymetrix. They contain information about the geometry and physical location locations of every single probe on the chip, in addition to the probeset information (both for exon and transcript levels) and biological annotation (as defined by Affymetrix). b On 27 September 2012 12:36, Javier P?rez Florido <jpflorido at="" gmail.com=""> wrote: > Dear list, > What are the differences between these two packages > (mogene10sttranscriptcluster.db and mogene10stprobeset.db)? > > As far as I know, the first annotation package is used when ST arrays are > normalized at the "core" level and the second one when data is normalized at > the "probeset" level. Is that right? > > Another question is related to the pd.mogene.1.0.st.v1 package, which is the > annotation package used in conjunction with the oligo package. What are the > differences between this package and the others > (mogene10sttranscriptcluster.db and mogene10stprobeset.db)? > > Thanks, > Javier > > _______________________________________________ > 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