Metabolic models are typically characterised by a large number of parameters. Traditionally, metabolic control analysis has been applied to differential equation-based models to investigate the sensitivity of predictions to parameters. A corresponding theory for constraint-based metabolic models is lacking due to their formulation as optimization problems. In this webinar we will show several applications of differentiating optimal solutions of constraint-based models, and show how it connects to classic metabolic control analysis. Efficient differentiation of constraint-based models can be used to calculate the sensitivities of predicted reaction fluxes and enzyme concentrations to turnover number parameters in an enzyme-constrained metabolic model of Escherichia coli. Further, it unlocks the ability to use gradient information for parameter estimation. We demonstrate this by improving, genome-wide, the state-of-the-art turnover number estimates for E. coli. Finally, this technique can be used to differentiate the optimal solution of a model incorporating both thermodynamic and kinetic rate equations. The predicted growth rate sensitivity to metabolite concentrations was shown to align well against experimentally measured metabolome changes subject to gene knockouts.

About the speaker

St. Elmo Wilken completed his undergraduate degree in Chemical Engineering at the University of Pretoria. His Ph.D. at the University of California, Santa Barbara leveraged both computational and wet lab aspects to investigate and understand the metabolism of anaerobic gut fungi. His current postdoc at the Institute of Quantitative and Theoretical Biology at the Heinrich Heine University in Düsseldorf is focused on using quantitative models to elucidate the contribution of metabolism to the stability and composition of microbial consortia. He partnered with PerMedCoE researchers, including Dr. Miroslav Kratochvil, to develop a way to differentiate constraint-based models to conduct sensitivity analyses efficiently.