Quantification of uncertainty in metabolic networks through the sampling of kinetic data space

Study of a metabolic network with respect to the sensitivity of the metabolite concentrations and the metabolic fluxes to genetic, biochemical, and environmental variations requires knowledge of the kinetic properties of the enzymes in the network. Experimentally observed data about the in vivo enzyme kinetics are subject to large variations of underlying physiochemical parameters. In this contribution, we present a computational framework that allows us to model the uncertainty in the kinetic data and to predict quantitatively the steady-state responses of metabolic networks in the presence of genetic and environmental perturbations. The proposed methodology relies on formalism from process control and metabolic control analysis and uses a Monte Carlo sampling method. In addition, it accounts explicitly for mechanistic properties of enzymes and physicochemical and thermodynamic constraints. The method employs a novel, efficient sampling procedure that allows us to simulate all possible meaningful states of a metabolic network. We demonstrate the properties of the proposed framework through an example of a prototypical module of biosynthetic pathways - the branched pathway.