Monte Carlo sampling of metabolic fluxes under thermodynamic constraints

A new constraint-based analysis method for metabolic networks has been developed. The possible steady-states in metabolic network were treated as a thermodynamic ensemble and a potential energy function enforcing additional constraints and virtual biomass was defined. The sampling in the stead-state flux space of the central metabolic network of Escherichia coli can avoid irrational fluxes violating thermodynamic constraints and mass balance and the results were consistent with the experimental data. The proposed method is more efficient than those reported, and the flux samples have better distribution than the random sampling method. Besides, other samples can be obtained, such as ethanol optimizing, via modifying the network and the potential function, which can be helpful to metabolic engineering.