Hi, Bioconductor For affy package, is it possible to calculate expression index with part of probe pairs? For example, I want to focus on probe pairs # 5 - 16 instead of #1 -16 and discard the informations from first four probe pairs. Thanks! -SW [[alternate HTML version deleted]]

yes its possible. the easiest way is to create a new cdf environmet that excludes these. some details: 1 - the hash tables in the cdf environmets contain a J x 2 matrix with the PMs locations in the exprs slot matrix in the first column and the MMs in the second. J is the number of probes (typically 16 in the hgu95a chip) 2- the order in which they appear are the given by "probe number". so all you do is take out the first 5 entries. 3- the cdfName slot in AffyBatch tells the affy package where to find the PM and MM information. once you define the new cdf package or environment you stick the name of the environmnet or package in the cdfName slot. 4- normalization and bg correction will only be done on these probes Leslie Cope (cc-ed) and I are actually working on somehting similar. We will soon have software that automatically does this. rafael On Thu, 4 Apr 2002, Sek Won Kong, M.D wrote: > Hi, Bioconductor > For affy package, is it possible to calculate expression index with part of probe pairs? > For example, I want to focus on probe pairs # 5 - 16 instead of #1 -16 and discard the informations from first four probe pairs. > Thanks! > -SW > > [[alternate HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor >