Automatic synthesis of both the topology and sizing of metabolic pathways using genetic programming
John R. Koza, William Mydlowec, Guido Lanza, Jessen Yu, Martin A. Keane · 2001
The concentrations of substances participating in networks of chemical reactions are modeled by non-linear continuous-time differential equations. Recent work has demonstrated that genetic programming is capable of automatically creating complex networks (e.g., analog electrical circuits, controllers) whose behavior is modeled by linear and nonlinear continuous-time differential equations and whose behavior matches prespecified output values. This paper describes how genetic programming can be used to automatically synthesize (reverse engineer) both the topology of the network of chemical reactions and the rates (sizing) of each reaction of a network such that the behavior of the automatically created network matches the observed timedomain data. Genetic programming has automatically created metabolic pathways that contain noteworthy topological features, such as an internal feedback loop, a bifurcation point where one substance is distributed to two different reactions, and an accumulation point where one substance is accumulated from two sources. 1