My experimental design is as follows: I have 5 disease state patients (Y) and 5 controls (N). For each patient I have samples from 4 x 2 (biological replicates) tissues (1,2,3,4). So I can set up a target file such as Sample;Disease;Individial;Tissue X1;Y;1;1 X2;Y;1;1 X3;Y;1;2 X4;Y;1;2 . . . X80;N;5;4 In this case I am interested in comparing the differential gene expression between tissues 1,2,3 combined in disease and control. So (Y.1.1+Y.1.2+Y.1.3+Y.2.1+...Y.5.3)- (N.1.1+ N.1.2+ N.1.3+ N.2.1+...N.5.3). Is this the correct way to set up the contrast matrix? An alternative approach is setting up the targetfile like below (this one is missing the "individual" column, so that all disease samples are treated as 1 and all controls as another): Sample;Disease;Tissue X1;Y;1 X2;Y;1 X3;Y;2 X4;Y;2 . . . X80;N;4 Shouldn't I get the same answer if I set up the constrast matrix using the above targetfile and now I would be interested in (Y.1+Y.2+Y.3)-(N.1+N.2+N.3)? I don't get the same answer so I'm a little confused on how to set up contrast matrices. Can anyone help? Thanks! [[alternative HTML version deleted]]