Ben Bolstad has done some work on this issue. http://stat-www.berkeley.edu/users/bolstad/mixtureCDF/MixtureCDF.html HTH, Jim James W. MacDonald Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623 >>> Willy Wynant <willy.wynant@curie.fr> 06/30/04 05:39AM >>> Hello, I would like to normalize data from fifty chips with the RMA method. The problem is that I have both U95A and U95AV2 chips which seem to differ from around only fifteen probesets. Has anyone already encountered this problem? Should I have to remove the probesets (on the CEL files) which are not the same in the chips, and normalize only the others? Thanks a lot for your help, Regards, Willy _______________________________________________ Bioconductor mailing list Bioconductor@stat.math.ethz.ch https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor