Predicting the experimental data of the substrate specificity of Aspergillus niger inulinase using mathematical models, estimating kinetic constants in the Michaelis-Menten equation, and sensitivity analysis

Germec M., TURHAN İ.

BIOMASS CONVERSION AND BIOREFINERY, vol.13, no.12, pp.10641-10652, 2023 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 13 Issue: 12
  • Publication Date: 2023
  • Doi Number: 10.1007/s13399-021-01830-1
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Compendex, INSPEC
  • Page Numbers: pp.10641-10652
  • Keywords: Aspergillus niger, Submerged fermentation, Enzyme, Substrate specificity, Mathematical modeling, Sensitivity analysis, ETHANOL-PRODUCTION, FERMENTATION, PURIFICATION, REACTOR
  • Akdeniz University Affiliated: Yes


This study aimed to develop a model to estimate experimental substrate specificity (inulin and sucrose) data for Aspergillus niger inulinase. For this purpose, ten mathematical models were used. Root-mean-square-error (RMSE), mean-absolute-error (MAE), mean standard deviation (MSD), determination coefficient (R-2), slope (m), Akaike information criterion (AIC), bias factor (B-F), and accuracy factor (A(F)) were used to be able to evaluate the models. Besides, the experimental enzyme kinetics were also compared with those calculated from the models. Also, the statistical significance of the models was determined by f-testing. Furthermore, the sensitivity analysis of the selected model was performed by making a 5% increase and decrease in each model parameter. Based on the results, the von Bertalanffy model was the best (RMSE = 28.48 and 27.34 U/mL, MAE = 21.53 and 25.14 U/mL, MSD = 15.23 and 17.77 U/mL, R-2 = 0.9908 and 0.9962, m = 1.02 and 1.00, AIC = 81.93 and 84.79, B-F = 1.00 and 1.00, and A(F) = 1.04 and 1.06 for inulinase and invertase-type activity, respectively). The predicted inhibition constant and maximum enzyme velocity for inulin and sucrose, which were highly good agreement with their experimental values, were calculated as 17.43 mg/mL and 1101 U/mL and 51 mg/mL and 1625 U/mL, respectively. According to the f-testing results, only the Mitscherlich model was significant for invertase-type activity while other all models except for the Stannard model were statistically significant for inulinase activity (p < 0.05). In the case of a deviation of 5% in the experimental values, extremely satisfactory results were obtained by the sensitivity analysis. Consequently, this is the first study to develop a model for predicting the kinetic parameters of the inulinase enzyme. The suggested model can be used in the industry to readily calculate the parameters regarding enzyme kinetics.