Biological systems are quintessentially complex.  To understand the genetic variation that leads to disease susceptibility or drug/vaccine response, this complexity must be reflected in the corresponding mathematical and computational models.  We are developing computational and mathematical methods that integrate biological data from different dimensional and temporal scales in order to tackle (1) the identification of gene-gene and gene-environment interactions that influence disease susceptibility in genome-wide data; (2) the discovery of interaction networks from noisy dynamic and static expression and genotypic data; (3) statistical clustering of expression data; (4) information theoretic methods to assess cluster quality with respect to known biological pathways; and (5) computational protein-protein docking.  We are particularly interested in applications of these methods to neural and immunological systems.