Dear Heidi and Bioconductor users, I am Jordi Altirriba from the Centre M?dical Universitaire, in Gen?ve, Switzerland. I have been using the R package HTqPCR, which I have found really useful. I have been working with Taq Man Low Density Arrays from Applied Biosystems (two different plates, A and B) to measure the miRNA expression of different samples and I have been looking to the distribution of the samples through the plotCtCategory and plotCtDensity to detect any outlier. Whereas the results with plotCtCategory and plotCtDensity where consistent in the first serie of experiments (all the graphs can be downloaded from the link below), in the second serie it was inconsistent, meaning that in the plotCtCategory we can observe a ?different? behaviour in the samples 39B, 41B and 42B whereas in the plotCtDensity we can observe that samples 40B, 26B, 30B and 12B show a different behaviour. For this reason I redid the density graphs modifying the parameter bandwith in the density function, now observing that samples seems to be more homogeneus than what it seemed with ploCtDensity and more in accordance with what is observed in plotCtCategory. Is this modification correct or am I missng something? (sorry for my small knowledge in statistics). Below is the code used and the session info. Thanks, Jordi Altirriba Link for images: http://dl.dropbox.com/u/7491875/resultados.ppt Code: Plate A library("HTqPCR") files<-read.delim("files.txt")raw<-readCtData(files=files$File,path=ge twd(), n.features = 384, flag = 2, feature = 4, type = 5, position = 1, Ct = 6, header = T, SDS = FALSE, na.value = 40) raw.cat<-setCategory(raw,groups=files$Treatment) plotCtCategoryraw.cat) plotCtDensity(raw, cex=0.6, lwd=1) for (x in 1:30) {plot(density(exprs(raw)[,x],bw=0.1), col=x, lty=x, ylim=c(0,5), xlim=c(0,40)) par(new=T)} legend(0,5,colnames(exprs(raw)), cex=0.7,lty=c(1:30), col=c(1:30)) for (x in 1:30) {plot(density(exprs(raw)[,x],bw=0.1), col=x, lty=x, ylim=c(0,5), xlim=c(39,40)) par(new=T)} legend(39,5,colnames(exprs(raw)), cex=0.7,lty=c(1:30), col=c(1:30)) plotCtHeatmap(raw,gene.names="",dist="euclidean") Plate B library("HTqPCR") files<-read.delim("files.txt")raw<-readCtData(files=files$File,path=ge twd(), n.features = 384, flag = 2, feature = 4, type = 5, position = 1, Ct = 6, header = T, SDS = FALSE, na.value = 40) raw.cat<-setCategory(raw,groups=files$Treatment) plotCtCategoryraw.cat) plotCtDensity(raw, cex=0.6, lwd=1) for (x in 1:28) {plot(density(exprs(raw)[,x],bw=0.1), col=x, lty=x, ylim=c(0,5), xlim=c(0,40)) par(new=T)} legend(0,5,colnames(exprs(raw)), cex=0.8,lty=c(1:28), col=c(1:28)) for (x in 1:28) {plot(density(exprs(raw)[,x],bw=0.1), col=x, lty=x, ylim=c(0,5), xlim=c(39,40)) par(new=T)} legend(39,5,colnames(exprs(raw)), cex=0.8,lty=c(1:28), col=c(1:28)) plotCtHeatmap(raw,gene.names="",dist="euclidean") sessionInfo() R version 2.12.0 (2010-10-15) Platform: i386-pc-mingw32/i386 (32-bit) locale: [1] LC_COLLATE=Spanish_Spain.1252 LC_CTYPE=Spanish_Spain.1252 [3] LC_MONETARY=Spanish_Spain.1252 LC_NUMERIC=C [5] LC_TIME=Spanish_Spain.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base other attached packages: [1] HTqPCR_1.4.0 limma_3.6.9 RColorBrewer_1.0-2 Biobase_2.10.0 loaded via a namespace (and not attached): [1] affy_1.28.0 affyio_1.18.0 gdata_2.8.1 [4] gplots_2.8.0 gtools_2.6.2 preprocessCore_1.12.0