Dear group members, I working with marrayXXX packages on genepix file. I wonder which channel is used as a reference, Green or Red, when the maNorm() function generates the "maM" matrix. Does M is log(R/G) or vise versa. I'm asking that because I analyze a dye swap experiment and I realize that I have to swap my self the reference for further analysis. I there a dye swap normalization function that does it automatically? Thanks in advance, Ron Ron Ophir, Ph.D. Bioinformatician, Biological Services Weizmann Institute of Science POB 26 Rehovot 76100 Israel e-mail: Ron.Ophir@weizmann.ac.il Phone: 972-8-9342614 Fax:972-8-9344113

Hi Ron, By default, M values are calculated as log(R/G) base 2. If you have a simple A vs B comparisons, here is a couple of options: E.g. Assming you have 5 replicated experiment of A vs B. Slides 1,2,4 were dye-swapped. Option A) M <- maM(data) ## Extract M values ## Flip the M according to dye swap experiments newM <- sweep(M, 2, c(-1,-1,1, 1, -1), FUN="*") ## The vector of -1 and 1 indicate the experiments that were dye swaped. Option B) Uses lm.series for a linear model analysis. Input the dye swap information as your design matrix library(limma) fit <- lm.series(maM(data), design=c(-1,-1,1,1,-1)) Cheers Jean On Thu, 26 Jun 2003, Ron Ophir wrote: > Dear group members, I working with marrayXXX packages on genepix file. > I wonder which channel is used as a reference, Green or Red, when the > maNorm() function generates the "maM" matrix. Does M is log(R/G) or > vise versa. I'm asking that because I analyze a dye swap experiment > and I realize that I have to swap my self the reference for further > analysis. I there a dye swap normalization function that does it > automatically? Thanks in advance, Ron > > Ron Ophir, Ph.D. > Bioinformatician, > Biological Services > Weizmann Institute of Science > POB 26 > Rehovot 76100 > Israel > e-mail: Ron.Ophir@weizmann.ac.il > Phone: 972-8-9342614 > Fax:972-8-9344113 > > _______________________________________________ > Bioconductor mailing list > Bioconductor@stat.math.ethz.ch > https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor >