Evaluation of Several Mathematical Models for Fitting the Growth and Kinetics of the Catechol-degrading Candida parapsilopsis: Part 2

Abstract

Kinetic equations, which describe the activity of an enzyme or a microorganism on a particular substrate, are crucial in understanding many phenomena in biotechnological processes. They allow the mathematical prediction of growth parameters important for identifying key parameters for controlling growth. We remodelled the published work of Rigo et al. (2010) using several more growth kinetic models such as Monod, Haldane, Teissier, Andrews and Noack, Hinshelwood, Moser, Aiba, Webb (Edward), Yano and Koga, Han and Levenspiel and Luong and evaluated the accuracy of the fitted model using statistical analysis such as Root Mean Square (RMSE), adjusted Coefficient of Determination (R2), corrected Akaike Information Criterion (AICc), Bias Factor and Accuracy Factor The accuracy and statistical analysis of the eleven kinetic models used shows that only the Haldane, Teissier, Moser, Aiba, Webb (Edward), Yano and Koga, and Luong could fit the data with the best model was Haldane with low values for RMSE and AICc, highest adjusted R2 values, and with Bias Factor and Accuracy Factor nearest to unity (1.0). The calculated value for the Haldane constants maximal degradation rate, half saturation constant and half inhibition constant symbolized by μmax, Ks and Ki, were 0.462 hr-1, 112.21 mg/L and 211.34 mg/L, respectively. The true umax value occurs where the gradient for the slope is zero and in this case the value was approximately 0.188 h-1 at 155 mg/L catechol. The results indicate that the exhaustive use of mathematical models on available published results could gleam new optimal models that can provide new knowledge on the way toxic
substance inhibit growth rate in microbes.