Dear Prasad, You don't need to worry about setting up the contrast matrix, as limma will do this for you. You can use makeContrasts() and tell it directly what comparisons you want to do. For example: cont.matrix <- makeContrasts(inf1-inf2,inf2-inf3,inf1-control1,levels=design) You get the idea. You can ask for any number of comparisons. There are many examples of this in the limma User's Guide. With respect to your first question, see Section 7.3 "Common reference designs". The easiest way to is to created an identifier for each unique treatment/time point combination in your experiments. Use modelMatrix() with ref="ref", and then proceed as above. Best wishes Gordon > Date: Tue, 22 Feb 2011 19:35:46 +0000 > From: Prasad Siddavatam <siddavatam at="" gmail.com=""> > To: <bioconductor at="" stat.math.ethz.ch=""> > Subject: [BioC] LIMMA package differentially expressed genes > Message-ID: <loom.20110222t201744-87 at="" post.gmane.org=""> > Content-Type: text/plain; charset="us-ascii" > > I am kind of new to Gene expression studies. > > I have two questions. > 1. What would be the design matrix for reference design with dye- swap time > course experiment. I have 6 time points and at each time point, I have dye-swaps > and 3 replicates for infection and control samples (A total of 12 samples at > every time point). What I want is the differentially expressed between control > and infection. > > > 2. In other set, I have to make multiple comparisons (infection1, infection2, > infection3). This is also a reference design with two-colors. > sample cy3 cy5 > array1 ref inf.1 > array2 ref inf.1 > array3 ref inf.1 > array4 ref inf.2 > array5 ref inf.2 > array6 ref inf.2 > array7 ref inf.3 > array8 ref inf.3 > array9 ref inf.3 > array10 ref control1 > array11 ref control2 > array12 ref control3 > what I want is differentially expressed genes between every pair of comparisons > (ex. inf1-inf2). > my design looks like this > inf1 inf2 inf3 control > array1 1 0 0 0 > array2 1 0 0 0 > array3 1 0 0 0 > array4 0 1 0 0 > array5 0 1 0 0 > array6 0 1 0 0 > array7 0 0 1 0 > array8 0 0 1 0 > array9 0 0 1 0 > array10 0 0 0 1 > array11 0 0 0 1 > array12 0 0 0 1 > > what would be my contrast matrix would be? > and how to get the genes between inf2 and inf3? > > Prasad siddavatam ______________________________________________________________________ The information in this email is confidential and intend...{{dropped:4}}