Using parameter sampling to explore key properties of E. coli's anaerobic metabolism

Abstract

The high uncertainty and incomplete knowledge of kinetic parameters is a major challenge for using kinetic metabolic models. Literature values are scarce and fitting procedures require large experimental data sets and complex computations. In this study we show that Monte-Carlo sampling of kinetic parameters allows the identification of system's properties such as stability and flux control patterns. Applying this computationally simple method to the anaerobic central metabolism of E. coli we determine that only few network parameters are directly correlated with stability. We show that low enzyme concentrations often correspond to positive eigenvalues, suggesting that enzyme-optimized pathways lack in flexibility. Analysis of the distribution of flux control coefficients reveals that the highest control on the network rates is exerted by reactions utilizing ATP. Finally, a comparison with experimental evidence confirms this method's potential to qualitatively analyze networks for which no information on parameters are available.