Hello, since recently there's the affy package offers the mas5calls fucntion to get the P, A or M calls for a summarized probeset. The usual processing protocol that I use is: 1. read cel file 2. bg correct 3. normalize 4. pm correct 5. summarize Where would I stick in the mas5calls? I assume this would be between steps 2 and 3 (i.e. before normalization), but it could also be after normalization or even after pm correct. I'd be interested in your comments. I'm currently processing some chips from different experimental batches that have (depending on the batch) a shift in intensities with one batch also having more present genes (as determined by the affy .DAT files) than the other. For this situation I could imagine to use the mas5calls *after* normalization to get a "faire" call (i.e. what's called 'P' on one chip should ideally also be called 'P' on the other chip even though the intensities are quite a lot lower). kind regards + thanks for your comments, Arne -- Arne Muller, Ph.D. Aventis Pharma, Drug Safety Evaluation 13 quai Jules Guesde 94403 Vitry-sur-Seine CEDEX, FRANCE phone: +33-(0)1589-32896 fax : +33-(0)1589-38197 email: arne.muller@aventis.com

if you want to reproduce affymetrix's calls you should do it after 1. results may improve if you do it after 3 but you certainly want to do it before 4. On Mon, 27 Oct 2003 Arne.Muller@aventis.com wrote: > Hello, > > since recently there's the affy package offers the mas5calls fucntion to get > the P, A or M calls for a summarized probeset. > > The usual processing protocol that I use is: > > 1. read cel file > 2. bg correct > 3. normalize > 4. pm correct > 5. summarize > > Where would I stick in the mas5calls? I assume this would be between steps 2 > and 3 (i.e. before normalization), but it could also be after normalization > or even after pm correct. I'd be interested in your comments. > > I'm currently processing some chips from different experimental batches that > have (depending on the batch) a shift in intensities with one batch also > having more present genes (as determined by the affy .DAT files) than the > other. For this situation I could imagine to use the mas5calls *after* > normalization to get a "faire" call (i.e. what's called 'P' on one chip > should ideally also be called 'P' on the other chip even though the > intensities are quite a lot lower). > > kind regards + thanks for your comments, > > Arne > > -- > Arne Muller, Ph.D. > Aventis Pharma, Drug Safety Evaluation > 13 quai Jules Guesde > 94403 Vitry-sur-Seine CEDEX, FRANCE > phone: +33-(0)1589-32896 > fax : +33-(0)1589-38197 > email: arne.muller@aventis.com > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor > --