I have a number of slides from different experiments and I would like to combine them in a model to be able to make contrasts between conditions not compared directly, and I am wondering what would be the best way to do that. It started as a simple comparison, and things were added later, so that's why the design is not as it would have been if we had sat down at the beginning and decided what to do... I have two normal cell lines, call them N1 and N2. I have two cancer cell lines, call then C1 and C2. Then I have a treatment on each cancer line, call them C1T and C2T. The direct experiments performed were: N1 vs C1 N1 vs C2 N2 vs C1 N2 vs C2 C1 vs C1T C2 vs C2T each containing 3-4 slides. I'm thinking that the simplest way would be to make two models, one with C1 as reference, and another with C2 as reference: N1 vs C1 N2 vs C1 C1T vs C1 ... and similarly for C2. Of course, it doesn't compare C1 vs C2, or C1T vs C2T, which may be interesting. I've only analysed very simple designs before, and I'm not quite sure what would be the best way to combine all this in a reasonable way. I can use part of the data to answer all my questions of interest... but since everything is related directly or indirectly, I guess there must be a way to take advantage of all of the data... any pointers much appreciated! Jose -- Dr. Jose I. de las Heras Email: J.delasHeras at ed.ac.uk The Wellcome Trust Centre for Cell Biology Phone: +44 (0)131 6513374 Institute for Cell & Molecular Biology Fax: +44 (0)131 6507360 Swann Building, Mayfield Road University of Edinburgh Edinburgh EH9 3JR UK