A question that spans biology is how metabolic pathways combine to generate the robust and efficient physiology of a living cell. In this project, investigators will probe the physiology of a cell by treating the cell as a complex system, and using tools that include generation of mathematical models of the metabolic system in question, testing of these models with genetic/biochemical analysis, and quantify the metabolites that make up the system. Together, results from each of these areas will inform the other and provide new understanding of the complex interplay of metabolic pathways that lead to the robustness of living systems. Student and postdoctoral personnel involved in this project will be trained at the interface of biology, chemistry, biotechnology, mathematics, and computation. Thus in addition to scientific insights, the work proposed will contribute to the development of critically needed human resources. The PI is further committed to human resource development with a strategy to increase the participation of Puerto Rican students in the research enterprise at the University of Georgia.
The proposed work is a collaboration between a microbial physiologist and and a systems biologist with expertise in metabolic modeling; and an expert in metabolomics and synthetic chemistry. This interdisciplinary project builds the initial collaboration to address a common question: How should one optimally represent the metabolic network of a cell in a succinct, quantitative way that is reliable and predictive, and that can be experimentally confirmed? The specific experimental system is are the purine/histidine/thiamine biosynthetic pathways, which are amenable to precise manipulations and can be used to test a wide array of predictions generated by metabolomic and modeling approaches. This feature of the project will allow the team to establish a "proof of principle" that it is possible to capture all pertinent features of a metabolic system with sufficient granularity to claim that the system is "understood". A convincing proof of this nature has not been provided thus far, but is critically needed in the field of metabolomics. From the modeling perspective, the chosen bacterial system provides a means to generate extensive data from a single system that can be used to refine and train a sequence of models in ways that are not possible with large-scale networks, which depend on data from diverse sources and are intrinsically of variable quality. This multidisciplinary approach will generate new knowledge and advance the approaches needed to quantitatively measure and model metabolism from sets of experimental data. It will also advance the development of tools and techniques that are transferrable to studies of metabolism in all kingdoms of life.
