Modern technology in computational biology enables us to quantitatively model integrated physiological systems of ever increasing size and complexity. For example, constraint-based approaches to modeling biochemical systems provide powerful computational tools for predicting metabolic capabilities and mapping gene product network interactions in whole-genome cell systems. One key to increasing the amount of information available from computer models and improving the accuracy of constraint-based modeling approaches is the identification of further physicochemical constraints under which biological systems operate. Toward that aim, we have introduced thermodynamic-based constraints on biochemical fluxes and ? concentration to augment the flux-balance analysis (FBA) approach. Our methodology allows us to make predictions not available from FBA alonc namely, predictions of reactant concentrations, reaction potentials, and enzyme activities. These new capabilities will allow us to introduce a wide range of new applications of constraint-based modeling, including: making predictions that are more physically realistic and biologically meaningful than are provided by FBA-based methods; making direct comparisons of computation predictions with experimental results; and predicting the regulatory and control mechanisms operating in cells. Our efforts in the development and evaluation of these computational tools will target energy metabolism in cardiac and skeletal muscle and in hepatocytes. A major goal of this proposed research program are to build and validate metabolic models of health and disease. Those models will be used for a number of applications including: (1) computational profiling of metabolic function in healthy and diseased tissues; (2) identification and testing of putative regulatory mechanisms; (3) determination of the mechanism of action of certain therapeutic agents; and (4) target identification and lead optimization. In addition, the computational tools that are developed by this research program will be disseminated through a partnership with a software company. ? ? ?
Showing the most recent 10 out of 20 publications
