Prediction of Flux Distribution of Mutants by eNZYME Control Fluxes with Maximum Entropy Principle

Abstract:

The integration of omics data in metabolic flux analysis is critical to explore the cellular mechanisms in the post-genomic era. Enzyme activity data can be integrated into elementary modes (EMs) by Enzyme Control Flux (ECF) for the prediction of flux distributions of mutants. A new ECF algorithm, named as ECFMEP, is proposed for a moderate scale of metabolic networks. To efficiently estimate a large number of EM coefficients (EMCs), the Lagrange multiplier is applied to the optimization problem under maximum entropy principle (MEP), thereby remarkably reducing the number of the variables to be explored. ECFMEP is employed to predict the flux distribution of four mutants of Escherichia coli under anaerobic conditions. These cells consist of 102 reactions and 28,555 EMs. We demonstrate that ECFMEP effectively makes use of enzyme activity data for enhanced prediction accuracy in comparison with that by Flux Balance Analysis (FBA) and Minimization of Metabolic Adjustment (MOMA).