Dear Dr Smyth, I have experiments on 1 normal tissue(the 1 normal tissue pooled from 8 individual tissues ) and 12 individual tumor tissues using Affymetrix HG-U133A genechips. no replicates. I plan to pre-process the .cel files using RMA and do data analysis with moderated t statistics of Limma package. My steps as follows, pls give me any suggestions and advice.if i want to get boxplot or some graphs, pls tell me how and where to add the related commands. thanks much for your help in advance. > dir() [1] "hgu133acdf" "Normal.CEL" "TG_05.CEL" "TG_10.CEL" "TG_12.CEL" [6] "TG_15.CEL" "TG_19.CEL" "TG_9.CEL" "TH_04.CEL" "TH_05.CEL" [11] "TH_07.CEL" "TH_10.CEL" "TH_11.CEL" "TH_14.CEL" > library(limma) > library(affy) Loading required package: Biobase Loading required package: tools Welcome to Bioconductor Vignettes contain introductory material. To view, simply type: openVignette() For details on reading vignettes, see the openVignette help page. Loading required package: reposTools > Data <- ReadAffy() > eset <- rma(Data) Background correcting Normalizing Calculating Expression > pData(eset) sample Normal.CEL 1 TG_05.CEL 2 TG_10.CEL 3 TG_12.CEL 4 TG_15.CEL 5 TG_19.CEL 6 TG_9.CEL 7 TH_04.CEL 8 TH_05.CEL 9 TH_07.CEL 10 TH_10.CEL 11 TH_11.CEL 12 TH_14.CEL 13 > tissue <- c("n","t","t","t","t","t","t","t","t","t","t","t","t") > design <- model.matrix(~factor(tissue)) > colnames(design) <- c("n","tvsn") > design n tvsn 1 1 0 2 1 1 3 1 1 4 1 1 5 1 1 6 1 1 7 1 1 8 1 1 9 1 1 10 1 1 11 1 1 12 1 1 13 1 1 attr(,"assign") [1] 0 1 attr(,"contrasts") attr(,"contrasts")$"factor(tissue)" [1] "contr.treatment" > fit <- lmFit(eset, design) > fit <- eBayes(fit) > options(digits=2) > topTable(fit, coef=2, n=100, adjust="fdr") ID M A t P.Value B 4556 205029_s_at -6.16 2.9 -37.9 1.4e-10 9.770 4557 205030_at -7.22 3.3 -22.2 7.8e-08 8.874 16787 217422_s_at -1.80 2.9 -8.4 8.4e-03 4.246 568 201040_at -0.97 6.4 -6.7 7.9e-02 2.594 21918 38521_at -1.93 5.5 -6.4 8.7e-02 2.292 5497 205970_at -1.76 4.3 -6.3 8.7e-02 2.238 10456 211010_s_at -1.21 4.4 -6.2 9.6e-02 2.061 18277 218913_s_at -0.98 5.3 -6.0 1.1e-01 1.885 the above mentioned methods suitable for my experiments? in the topTable,how to select the all significantly differentially expressed genes? if based on p value cutoff, how to select the all signifcantly diffierntially expressed genes? how to discriminate between the up-regulated and down-regulated genes? positive M means upregulated,negative for repressed? if i want to get the fold change, where and how to get?can get the fold change by M2 ? Best Regards Weinong Han --------------------------------- Stay connected, organized, and protected. Take the tour [[alternative HTML version deleted]]

Hi > I have experiments on 1 normal tissue(the 1 normal tissue pooled from > 8 individual tissues ) and 12 individual tumor tissues using > Affymetrix HG-U133A genechips. no replicates. I plan to pre-process the .cel > files using RMA and do data analysis with moderated t statistics of Limma > package. Can I ask what question you are asking biologically/clinically? Do you want genes that vary between normal and tumour samples, or genes that vary amongst your tumour samples (and might therefore be a useful starting point for some sort of classification study)? If you are going to do the first then you have a big problem - with only one 'normal' chip you have no idea just how variable any particular gene will be amongst a population of normal individuals. Regards Professor David Kipling Department of Pathology School of Medicine Cardiff University Heath Park Cardiff CF14 4XN Tel: +44 29 2074 4847 Fax: +44 29 2074 4276 Email: KiplingD@cardiff.ac.uk