Characterizing complex phenotypes in metabolism : an "omics"-driven systems approach

The advent of high-throughput technologies has resulted in an explosion of molecular data. A major challenge is found in interpreting and understanding these different types of data sets at a phenotypic level. Systems biology has capitalized on these technologies by consolidating various types of biological information into structured networks for their analysis and computation. The bottom-up systems biology approach, in particular, has been crucial in providing mechanistic foundations for systems-level modeling in microorganisms, and its extension to eukaryotic metabolism has made it possible to elucidate complex phenotypes in a systematic manner. The work presented in this dissertation describes the integrative use of high-throughput data and genome-scale network reconstructions to characterize complex phenotypes of eukaryotic metabolism. First, the genome-scale reconstructions of yeast and human metabolism are discussed, which provide the contextual basis in which "omics" data is analyzed. Previously developed constraint- based modeling approaches were refined to analyze "omics" data sets, in particular for transcriptomic and metabolomic data. Finally, example applications are presented in the evaluation of physiological and perturbed metabolic states of yeast and human cellular systems. The studies discussed herein are: (1) analyzing drug response phenotypes of human metabolism; (2) evaluating genetic and environmentally perturbed processes in yeast ammonium assimilation; and (3) characterizing the pluripotent phenotype of embryonic stem cell metabolism. The work described in these studies represents advancement towards integrating bottom-up and data-driven approaches to understanding broader "omics"-to-phenotype relationship