A Geometric Programming Algorithm For Steady-state Optimization Of Biochemical Systems

A biochemical system is often described by Michaelis-Menten equations. It is necessary to develop an efficient nonlinear optimization solver to solve the steady-state optimization problems of biochemical systems. However, it is not easy to do this work because the steady-state optimization problems of biochemical systems are usually nonconvex and nonlinear. It is difficult to obtain the global optimal solution of this class of problem. So it is necessary to propose an appropriate optimization algorithm.This dissertation studies the steady-state optimization problems of biochemical systems in the framework of GMA systems. A geometric programming algorithm is proposed to solve this class of problems.The main contents and results obtained of this thesis are as follows:1. An equivalent transformation is first used to transform the steady-state optimization problem of a biochemical system represented by GMA model into a signomial geometric programming. Then convexification strategies are applied to convert the obtained signomial geometric programming into a sequence of standard geometric programming problems. Thus, a geometric programming algorithm is obtained to solve the steady-state optimization problems of biochemical systems.2. In order to illustrate the effectiveness of the algorithm, the proposed algorithm is applied to tryptophan biosynthesis and anaerobic fermentation in Saccharomyces cerevisiae. The computation results show that the algorithm proposed in this dissertation can obtain the true optimum solution. Compared with the existing geometric programming method, the geometric programming algorithm in this dissertation has the advantage of quickly obtaining the optimal solution of a biochemical system.