In this talk, we consider the problem of assessing differential expression of entire gene sets in complex biological experiments. We first propose a latent variable model that directly incorporates the underlying regulatory network. We then exploit the theory of mixed linear models (MLM), to develop a general inference framework for analysis of subnetworks, which also accounts for changes in the network structure. We briefly discuss computational issues and apply the proposed method to analyze data from yeast experiments. UCLA Department of Statistics 2009-08-06T17:26:00-07:00 2009-10-13 Statistics true 4:00 PM 416 4660 Geology Bldg. University of Michigan false true UCLA Department of Statistics Seminar George Michailidis 3:00 PM 2009-10-07T07:05:01-07:00