Automatic selection of the most promising enzymatic modulations for metabolic engineering: a multi-objective optimization approach

Single objective optimization methods are unable to capture the underlying complexity of cellular metabolism. Connectivity and regulatory effects make it necessary to resort to multi-objective strategies in order to avoid solutions involving unrealistic values in key processes. In this work we address the multi-objective global optimization of metabolic networks that are described by the generalized mass action (GMA) kinetic formalism. We present a systematic framework for the efficient solutions of these problems based on the combined used of the epsilon constraint method and tailored global optimization strategies. Our approach includes also a set of Pareto filters that allow reducing the number of alternatives resulting from the multi-objective analysis to a small enough set embedding only the most promising enzymatic profiles. The usefulness of our method is demonstrated by means of a case study that addresses the optimization of the ethanol production in Saccharomyces Cerevisiae.