Aspartate biosynthesis pathway simulation using an improved differential evolution algorithm through parameter estimation

Abstract

Aspartate Biosynthesis Pathway Simulation Using an Improved Differential Evolution Algorithm through Parameter Estimation Chuii Khim Chong1, Mohd Saberi Mohamad1, Safaai Deris1, Shahir Shamsir2, Yee Wen Choon1 and Lian En Chai1 1 Artificial Intelligence and Bioinformatics Research Group, Faculty of Computer Science and Information Systems, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia. 2 Department of Biological Sciences, Faculty of Biosciences and Bioengineering, Universiti Teknologi Malaysia, 81310 UTM Skudai, Johor, Malaysia Corresponding author’s e-mail: saberi@utm.my Keywords: Parameter Estimation, Differential Evolution Algorithm, Kalman Filter, Simulation. Introduction. Metabolic Engineering is a method which allows modifications of the pathways in suitable host cells. It aims at producing a novel or achieving an expected amount of desire compound for medical and industrial use. Recent studies mainly have concentrated on the aim of analysis by altering the computer readable data from the biological process. Thus, with the study of metabolic pathway, scientists can simulate the process inside the cell by mathematical modeling. The main goal of system biology is to develop an accurate pathway model that functions as a biological function simulator. Aspartate biosynthesis pathway is a sequence of events that occur in a cell causing production of amino acid called aspartate. Aspartate is very crucial in the urea cycle for the proper elimination of waste products from dietary protein. Parameter estimation is one of the crucial steps in constructing mathematical model. Regrettably, it faces a number of difficulties, for example high complexity of the model which is caused by the increasing number of unidentified parameters and equations in the model [1], and the existence of noisy data which causes low accuracy [2]. Thus, we proposed IDE which is a hybrid of DE and KF, to solve the existence of noisy data that leads to low accuracy for estimated result and the increasing unidentified parameters which lead to the complexity of the model. Noisy data can occur when the retrieved results differ from each other and this is due to apparatus limitation or human error. The advantages of DE are speed, straightforwardness, efficiency, and ease of use as it contains only few control parameters [3]. Moreover, KF can improve DE’s performance as it uses Kalman gain value which handles noisy data to update the population [2]. Moonchai Sompop et al. [4] and Christophe Chassagnole et al. [5], implemented DE as a parameter estimation approach to enhance the production of aspartate, bacteriocin, beer, and the simulation of the actual process in cell by estimating the control parameters and kinetic parameters.