The ultimate goal of genetics, and therefore of genomics, is to understand the relationship between genotype (sequence) and phenotype (function). Our primary aim is to gain the ability to predict (and mathematically model) the phenotypic outcome of mutations (polymorphisms). Achieving this aim will depend critically on our ability to understand the interactions among the nucleic acids, proteins, and other metabolites that comprise genetic regulatory networks. To this end we are applying both experimental and computational approaches to unravel the rules that govern the interactions among sets of genes that contribute to the same phenotype. Currently we are focusing our efforts in two areas.

1) Complex Traits. Using yeast as a model system we study the genetic basis of naturally occurring phenotypic variation. We have assembled a collection of natural isolates of S. cerevisiae that show many phenotypic differences. Using a combination of modern functional genomics and classical quantitative genetics we identify the specific DNA variants that underlie these differences. In parallel we build mathematical models designed to explain how these specific variants impact the flow of information through genetic regulatory networks. We then experimentally test these models and use the resulting data to refine the models. Our aims are to gain the ability to precisely predict the phenotype of an individual based on their genotype, and to understand the molecular mechanisms that underlie differences between individuals.

2) Engineering Gene Expression. We aim to gain the ability to predict the expression pattern of a gene based on the sequence of its promoter. To this end we are implementing a "synthetic biology" approach by developing new technologies to assay large numbers of engineered promoters with different combinations of cis-regulatory sites. Simultaneously we are developing a quantitative framework to use this data to predict the expression patterns both of promoters in the genome and of novel, engineered promoters.