Hello, I was trying to reproduce just an example of differential expressions from FDR (false discovery rates) in ALL data (see the book Bioinformatics and Computational Biology solutions using R and Bioconductor of Gentleman et al., p. 232-234). But I am not sure to understand what it makes. It is an application of multtest package to control the FDR at 0.05 level. In brief, one can write data(ALL) pdat<- pData(ALL) subset <- intersect (grep("^B", as.character(pdat$BT)), which(pdat$mol %in% c("BCR/ABL", "NEG"))) eset <- ALL[,subset] library(genefilter) f1 <- pOverA(0.25,log2(100)) f2 <- function(x) (IQR(x)>0.5) ff <- filterfun(f1,f2) selected <- genefilter(eset,ff) sum(selected) esetSub <- eset[selected,] cl <- as.numeric(esetSub$mol == "BCR/ABL") resT <- mt.maxT(exprs(esetSub), classlabel = cl, B=10000) ord <- order(resT$index) rawp <- resT$rawp[ord] res <- mt.rawp2adjp(rawp, proc="BH") sum(res$adjp[,"BH"]<0.05) > 102 That leaves 102 significant genes. I would like to extract just the expressions of these 102 genes under the two conditions. Therefore I would expect something like several columns corresponding to different persons (37 and 42?) in 102 rows with the values of expressions and the names of these genes (as dimnames?). But I do not know how to handle with this task :-( Any help? Thank you very much jm~ _______________________________ Juan Miguel Marin http://www.est.uc3m.es/jmmarin Dep. of Statistics University Carlos III of Madrid Spain (E.U.)