Hello forum, Outline of my analysis: lumi for pre-processing and filtering, then i take a subset of the normalized group and run this through limma. I've noticed that if i select my subset differently, the topTable generated after limma differs. ie rather than select=c(1,2,3,4,5,6) it is changed to select=c(1,4,3,2,5,6) the final topTable differ although the samples input are actually identical Does this mean that the subset order must be identical to that in the targets file? Or have I missed something in my current limma code (see below)? I believe I'm currently lacking a line which would indicate which sample fits in each spot of the model.matrix, if that makes sense? Any help would be greatly appreciated!! example of my code. subset <- subset(selDataMatrix, select=c(1,2,3,4,5,6)); targets<-read.csv("targets.csv", skip=7); targets Sample_Name Sample_Group Pairs Sentrix_ID Sentrix_Position 1 T 1 2 C 2 3 C 1 4 T 2 5 T 3 6 C 3 clinical <- factor(targets$Sample_Group); individual <- factor(targets$Pairs); design <- model.matrix(~1+clinical+individual) #design headers appear are as followe: (Intercept) clinicalT individual2 individual3 fit <- lmFit(subset, design); ebayes <- eBayes(fit); [[alternative HTML version deleted]]

On Nov 30, 2011 5:22 PM, "Kripa R" <kripa777@hotmail.com> wrote: > > > Hello forum, > Outline of my analysis: lumi for pre-processing and filtering, then i take a subset of the normalized group and run this through limma. > I've noticed that if i select my subset differently, the topTable generated after limma differs. ie rather than select=c(1,2,3,4,5,6) it is changed to select=c(1,4,3,2,5,6) the final topTable differ although the samples input are actually identical > Does this mean that the subset order must be identical to that in the targets file? Or have I missed something in my current limma code (see below)? > I believe I'm currently lacking a line which would indicate which sample fits in each spot of the model.matrix, if that makes sense? > Yes. The targets and sample info need to be in the same order and you are responsible for making sure that is the case. While it is a bit more work, I highly recommend constructing an ExpressionSet to avoid ordering errors and to keep the sample info and data together. Sean > Any help would be greatly appreciated!! > > > example of my code. > subset <- subset(selDataMatrix, select=c(1,2,3,4,5,6)); > > targets<-read.csv("targets.csv", skip=7); > targets > Sample_Name Sample_Group Pairs Sentrix_ID Sentrix_Position > 1 T 1 > 2 C 2 > 3 C 1 > 4 T 2 > 5 T 3 > 6 C 3 > > > clinical <- factor(targets$Sample_Group); > individual <- factor(targets$Pairs); > > design <- model.matrix(~1+clinical+individual) > #design headers appear are as followe: (Intercept) clinicalT individual2 individual3 > > fit <- lmFit(subset, design); > ebayes <- eBayes(fit); > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor [[alternative HTML version deleted]]

Hi Kripa, Yes, the subset order has to be the same with that in your target file. Otherwise you will be comparing wrong samples. BTW, limma has its own normalization function for normalizing Illumina BeadChip data. Have a look at the neqc function in limma. Cheers, Wei On Dec 1, 2011, at 9:20 AM, Kripa R wrote: > > Hello forum, > Outline of my analysis: lumi for pre-processing and filtering, then i take a subset of the normalized group and run this through limma. > I've noticed that if i select my subset differently, the topTable generated after limma differs. ie rather than select=c(1,2,3,4,5,6) it is changed to select=c(1,4,3,2,5,6) the final topTable differ although the samples input are actually identical > Does this mean that the subset order must be identical to that in the targets file? Or have I missed something in my current limma code (see below)? > I believe I'm currently lacking a line which would indicate which sample fits in each spot of the model.matrix, if that makes sense? > > Any help would be greatly appreciated!! > > > example of my code. > subset <- subset(selDataMatrix, select=c(1,2,3,4,5,6)); > > targets<-read.csv("targets.csv", skip=7); > targets > Sample_Name Sample_Group Pairs Sentrix_ID Sentrix_Position > 1 T 1 > 2 C 2 > 3 C 1 > 4 T 2 > 5 T 3 > 6 C 3 > > > clinical <- factor(targets$Sample_Group); > individual <- factor(targets$Pairs); > > design <- model.matrix(~1+clinical+individual) > #design headers appear are as followe: (Intercept) clinicalT individual2 individual3 > > fit <- lmFit(subset, design); > ebayes <- eBayes(fit); > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor ______________________________________________________________________ The information in this email is confidential and intend...{{dropped:6}}