Computation of elementary flux modes by column generation

Abstract

Identification of elementary flux modes in genome-scale metabolic networks is difficult due to the combinatorial nature of the problem. However, in systems biology, constraint-based modeling utilizes only a few relevant elementary flux modes. So, their computation using optimization is emerging as a recent trend. We proposed an algorithm based on the column generation paradigm for computing elementary flux modes that allow for massive computational-time savings compared to previous approaches. The algorithm proved capable of efficiently generating elementary flux modes for genome-scale metabolic networks with thousands of intracellular reactions.