Kinetic model parameter estimation using genetic algorithms of the oxidation of phenol in water catalyzed by the laccase enzyme for the design of a biosensor

Abstract

Predicting the dynamic response of key processes that take place in industrial biochemical systems such as the measurement of operating parameters by means of biosensors, before the biosensors themselves are prototyped, is of utmost importance. The advantages of performing mathematical modeling of biosensor systems in their development stages include cost reduction, easier response tuning and faster performance optimization. In most cases, the mathematical models for enzyme kinetics depend on multiple parameters. Finding the numerical values of such parameters usually requires carrying out a vast number of experiments, which is time intensive, expensive and involves the usage of specialized laboratory equipment. This work proposes the utilization of genetic algorithms as an alternative methodology for kinetic model parameter estimation of the oxidation of phenol in water, catalyzed by the laccase enzyme. The corresponding kinetics mathematical model of the oxidation reaction is used as a case study, to compare the results obtained using the genetic algorithms approach proposed with those found in the literature. The algorithm estimated the values of several parameters of the model, such as reaction rate constants, rate constant of transformation of oxygen by the electrode and stoichiometric coefficient, among others. The results found in this investigation by means of genetic algorithms show an agreement of 91%–99% with the data available in the literature. This approach also proved to be more accurate than the basic polynomial regression estimation method, which is commonly used and was implemented for comparison purposes. The proposed technique for parameters estimation in enzyme reaction models enabled the design of a phenol biosensor for concentrations ranging from 5 to 30 ppm. This technique has a high potential of application in the biosensor industry because of its cost savings, high speed and good accuracy.