Hi, After viewing the various postings on this topic ("technical replicates (again!): a summary"), I couldnt find a solution to a pretty basic problem. I have two technical replicates per sample, and 7 samples per treatment. I have 4 treatments, which I want to contrast with each other (all using 2-color chemistry). I created a design which treated each sample as a duplicate (i..e 1, -1 as dye swaps were involved), and then tried to use this to create contrasts, with subsetting etc. This ended up being very laborious, and there may not be a way to do this easily in limma, but I could be missing something. What I have is this: 4 treatments: A,B,C,D For each treatment, I have 2 technical replicates (dye swaps), I want to average these in limma, and then look at comparisons between groups, and create contrasts between A and B, etc. However, I dont see an easy way of doing this, short of a lot of subsetting, and renaming variables. Any help would be appreciated. thanks Simon.

The simplest thing to do is to average the technical reps afternormalization but before running limma. --Naomi At 04:40 PM 5/2/2004, Simon Melov wrote: >Hi, >After viewing the various postings on this topic ("technical replicates >(again!): a summary"), I couldnt find a solution to a pretty basic >problem. I have two technical replicates per sample, and 7 samples per >treatment. I have 4 treatments, which I want to contrast with each other >(all using 2-color chemistry). > >I created a design which treated each sample as a duplicate (i..e 1, -1 as >dye swaps were involved), and then tried to use this to create contrasts, >with subsetting etc. This ended up being very laborious, and there may not >be a way to do this easily in limma, but I could be missing something. > >What I have is this: 4 treatments: A,B,C,D >For each treatment, I have 2 technical replicates (dye swaps), I want to >average these in limma, and then look at comparisons between groups, and >create contrasts between A and B, etc. However, I dont see an easy way of >doing this, short of a lot of subsetting, and renaming variables. > >Any help would be appreciated. > >thanks > >Simon. > >_______________________________________________ >Bioconductor mailing list >Bioconductor@stat.math.ethz.ch >https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor Naomi S. Altman 814-865-3791 (voice) Associate Professor Bioinformatics Consulting Center Dept. of Statistics 814-863-7114 (fax) Penn State University 814-865-1348 (Statistics) University Park, PA 16802-2111