Computational Strain Design for Improved Productivity, Yield and Robustness

Bioprocess development for biofuels and biochemicals typically requires several rounds of metabolic engineering to meet process targets including product yield, titer and productivity, all of which impact the process economics.  Similar advances in computational modeling techniques have allowed the development of genome-scale models of metabolism in several organisms and their use in rational design of strains. In the first part, a rational approach based on bi-level optimization to enhance bioprocess productivity by forcing co-utilization of substrates will be shown.  Experimental results from the application of this approach to enforce substrate co-utilization in Escherichia coli will be discussed. In the next part of the talk, a novel nested nonlinear optimization method for metabolic engineering resulting in over hundred different strain design strategies for biochemicals production will be presented.  We will also examine the role of the redundant pathways from a design perspective and present computational results on how these pathways is valuable for robust design.