Hello, I am processing some CEL files using MAS5. I had processed them using earlier version of R(R1.6.1) and Bioconductor. The results I am getting for the same data with R 1.7 and new version of Bioconductor are very different. The new version of bioconductor is running on Linux while the old version of bioconductor is running on Windows. It would be nice if somebody could inform me which results to trust and whether this is a bug or a planned change. Can the different OS be a reason behind different results. Regards, Tapan Mehta

there was a bug in earlier versions of the affy package. if you want details, there were various reports on this list which you can see in the mail archive. the current version reproduces Affymetrix's algorithm closely: http://stat- www.berkeley.edu/users/bolstad/MAS5diff/Mas5difference.html On Wed, 9 Jul 2003, Tapan Mehta wrote: > Hello, > > I am processing some CEL files using MAS5. I had > processed them using earlier version of R(R1.6.1) and > Bioconductor. The results I am getting for the same > data with R 1.7 and new version of Bioconductor are > very different. The new version of bioconductor is > running on Linux while the old version of bioconductor > is running on Windows. It would be nice if somebody > could inform me which results to trust and whether > this is a bug or a planned change. Can the different > OS be a reason behind different results. > > Regards, > > Tapan Mehta > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor >