An embedded and charset-unspecified text was scrubbed... Name: not available Url: https://stat.ethz.ch/pipermail/bioconductor/attachments/20060126/ 947981bf/attachment.pl

Hi John, I think the best would be joint statistical model (eg linear model on log intensities) that incorporates both the usual normalization effect, the species effect, the chiptype effect, and whatever biological effect you might be looking for. If you can't do that, I'd try both: gcRMA to normalize the entire collection of arrays as a single set; apply gc RMA separately to each set of chips. After appropriate subsequent analysis for diff. expressed genes, the results shoulnd't differ much. There are some further thoughts about this here: http://www.bepress.com/bioconductor/paper8/ Best wishes Wolfgang Cornell, John E wrote: > Hi Folks: > > > > We have affy chips from human tissues (hgu133A chip) and tissues from a > mouse model (MOE403A chip). The objective of this experiment is to > compare expression levels across species. My primary question for the > group is, "What is the best strategy for normalizing the arrays?" > Should we use gc RMA to normalize the entire collection of arrays (human > and mouse) as a single set; or, should we apply gc RMA separately to > each set of chips? What do you suggest? > > > > Cheers, > > > > John E. Cornell, Ph.D. > > Associate Professor > > Center for Epidemiology and Biostatistics > > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor -- Best regards Wolfgang ------------------------------------- Wolfgang Huber European Bioinformatics Institute European Molecular Biology Laboratory Cambridge CB10 1SD England Phone: +44 1223 494642 Fax: +44 1223 494486 Http: www.ebi.ac.uk/huber