Hi, I have 12 CEL files and 4 Different groups. 2 untreated and 2 treated. Can anyone tell me what could be wrong with my design matrix. My CEL files are highly variable or something? Any answer would be helpful. I have below error : > results <- decideTests(fit); Error in array(x, c(length(x), 1), if (!is.null(names(x))) list(names(x), : attempt to set an attribute on NULL What I did is: > treatments <- factor(c(1,1,1,2,2,2,3,3,3,4,4,4),labels = c("Group1","Group2","Group3","Group4")); > contrasts(treatments) <- cbind(comp1 = c(1,0,1,0),comp2 = c(1,1,0,0), comp3 = c(0,1,1,0)); > design <- model.matrix(~treatments); > colnames(design) <- c("Intersept","G1_G3","G1_G2","G2_G3"); > design Intersept G1_G3 G1_G2 G2_G3 1 1 1 1 0 2 1 1 1 0 3 1 1 1 0 4 1 0 1 1 5 1 0 1 1 6 1 0 1 1 7 1 1 0 1 8 1 1 0 1 9 1 1 0 1 10 1 0 0 0 11 1 0 0 0 12 1 0 0 0 attr(,"assign") [1] 0 1 1 1 attr(,"contrasts") attr(,"contrasts")$treatments comp1 comp2 comp3 Group1 1 1 0 Group2 0 1 1 Group3 1 0 1 Group4 0 0 0 > fit <- lmFit(myRMA,design); > results <- decideTests(fit); Error in array(x, c(length(x), 1), if (!is.null(names(x))) list(names(x), : attempt to set an attribute on NULL > vennDiagram(results); Error in .class1(object) : Object "results" not found Thank you, Saurin

You must use eBayes() before decideTests(), as in all the documentation examples: fit <- lmFit(myRMA,design) fit <- eBayes(fit) results <- decideTests(fit) Gordon > Date: Tue, 14 Dec 2004 08:20:11 -0800 (PST) > From: Saurin Jani <saurin_jani@yahoo.com> > Subject: [BioC] Limma error > To: Bioconductor Bioconductor <bioconductor@stat.math.ethz.ch> > > Hi, > > I have 12 CEL files and 4 Different groups. 2 > untreated and 2 treated. Can anyone tell me what could > be wrong with my design matrix. My CEL files are > highly variable or something? Any answer would be > helpful. > > I have below error : > >> results <- decideTests(fit); > Error in array(x, c(length(x), 1), if > (!is.null(names(x))) list(names(x), : > attempt to set an attribute on NULL > > What I did is: > >> treatments <- > factor(c(1,1,1,2,2,2,3,3,3,4,4,4),labels = > c("Group1","Group2","Group3","Group4")); >> contrasts(treatments) <- cbind(comp1 = > c(1,0,1,0),comp2 = c(1,1,0,0), comp3 = c(0,1,1,0)); >> design <- model.matrix(~treatments); >> colnames(design) <- > c("Intersept","G1_G3","G1_G2","G2_G3"); >> design > Intersept G1_G3 G1_G2 G2_G3 > 1 1 1 1 0 > 2 1 1 1 0 > 3 1 1 1 0 > 4 1 0 1 1 > 5 1 0 1 1 > 6 1 0 1 1 > 7 1 1 0 1 > 8 1 1 0 1 > 9 1 1 0 1 > 10 1 0 0 0 > 11 1 0 0 0 > 12 1 0 0 0 > attr(,"assign") > [1] 0 1 1 1 > attr(,"contrasts") > attr(,"contrasts")$treatments > comp1 comp2 comp3 > Group1 1 1 0 > Group2 0 1 1 > Group3 1 0 1 > Group4 0 0 0 > >> fit <- lmFit(myRMA,design); >> results <- decideTests(fit); > Error in array(x, c(length(x), 1), if > (!is.null(names(x))) list(names(x), : > attempt to set an attribute on NULL >> vennDiagram(results); > Error in .class1(object) : Object "results" not found > > Thank you, > Saurin