Source: Dissertation Abstracts International, Volume: 73-07(E), Section: B
Advisers: Bernhard O. Palsson; Jeff Hasty
Thesis (Ph.D.)--University of California, San Diego, 2012
Fundamental physical phenomena are studied with a "cause and effect" approach. This enables understanding and prediction by employing mathematically formulated physical laws. Such approaches are less successful in biological systems, since they are subject to dual causation. That is, both physicochemical laws and evolving genetic constraints govern organisms. Biological systems respond immediately to stimuli (proximal causation) against a constant genetic background; however, these responses depend upon evolving genetic programs. Alterations in genetic programs are manifestations of distal causation, representing changes induced by genetic drift and natural selection. Constraint-based reconstruction and analysis is an emerging modeling approach that can account for both physicochemical constraints in biological systems and some evolutionary selective pressures. Here, constraint-based modeling is deployed to integrate disparate data types with genome-scale metabolic models to gain insight into mechanisms in proximal and distal causation, and conceptual advances are presented with respect to how these data are interpreted using constraint-based models. Specifically, these advances are used to suggest mechanisms determining proximal responses with respect to disease progression in human brain metabolism and the regulation of prokaryotic metabolism in dynamic environments. In addition, methods are presented that use genome-scale models of metabolism to analyze various data types to identify determinants of distal causation. Specifically, these methods are deployed to show that the evolution of enzyme specificity is guided by network context and the need to produce biomass. Moreover, these pressures further tune expression levels of metabolic pathways in laboratory evolved bacteria. Thus, through network reconstruction and data integration, vast amounts of data can be queried and provide detailed insight into proximal and distal causation in complex biological networks
