Dear Dr. Smyth, I have "two groups Affymetrix" experiment with 5 biological replicates in one group and 2 technical replicates in other. In other words one group has all biological replicates and the other has all technical replicates. I would really be thankful to know how to use LIMMA (to identify differentially expressed genes) for such kind of design. After reading LIIMMA user's guide and your post on [BioC] on "defining and handling replicates", the code appears to be written in following way: design <- cbind(disease=c(1,1,1,1,1,0,0),control=c(0,0,0,0,0,1,1)) corfit <- duplicateCorrelation(eset, design, ndups=?, block=c(???)) fit <- lmFit(eset, design, block=c(???), correlation=corfit$consensus) But how do I write block and ndups? Your help is really appreciated. Thanks and Regards, -- ~Nikhil Garge 3040 Cornwallis RD Durham NC 27709 Research Triangle Institute [RTI} Email: ngarge@rti.org [[alternative HTML version deleted]]