I would like to know why some of you bioconductorians are using Limma and some are using Sam. I find some quite convinient and the theory seems straightforward in a statistical sense. Please let me know what your reasons are for using one of these two approaches. Cheers

Well, first SAM is only used to compare two samples (one comparison at a time) and identify differentially genes, while limma can make more than one comparison at a time through modelling fitting, e.g., 3 pairwise comparison out of 3 samples. Second, limma is also able to perform other analysis beside identifying differentially expressed genes, like test the interaction terms in expression profiles, study the effect of a particular factor in gene expression. And the bottom line is: SAM is non- parametric which doesn't have any assumption whie limma does assume a linear model for the expressin. That is what came to my mind immediately, but there should be more. Hopefully this helps. Fangxin > I would like to know why some of you bioconductorians are using Limma and > some are using Sam. I find some quite convinient and the theory seems > straightforward in a statistical sense. Please let me know what your > reasons are for using one of these two approaches. > Cheers > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > > -- Fangxin Hong, Ph.D. Plant Biology Laboratory The Salk Institute 10010 N. Torrey Pines Rd. La Jolla, CA 92037 E-mail: fhong@salk.edu