Optimization of process parameters for improved chitinase activity from Thermomyces sp. by using artificial neural...

Area of Science:

• Enzymology
• Industrial Biotechnology
• Microbial Fermentation

Background:

• Chitinase enzymes break down chitin into N-acetyl-glucosamine units.
• Chitinases have significant applications in waste management and pest control.
• High yield and thermal stability are crucial for industrial chitinase applications.

Purpose of the Study:

• To optimize chitinase production from Thermomyces lanuginosus MTCC 9331.
• To compare the efficacy of Response Surface Methodology (RSM), Artificial Neural Network (ANN), and Genetic Algorithm (GA) for process optimization.

Main Methods:

• Utilized central composite design with RSM for initial parameter optimization.
• Employed ANN (multilayer feed-forward algorithms: Levenberg-Marquardt, Bayesian Regularization, Scaled Conjugate Gradient) for prediction.
• Validated results using GA and ANN, comparing predicted values with experimental data.

Main Results:

• ANN predicted a higher chitinase activity (102.24 U/L) compared to RSM (88.38 U/L).
• ANN predictions closely matched experimental observations.
• ANN demonstrated superior optimization performance over RSM and GA.

Conclusions:

• Artificial Neural Network (ANN) is a more effective method for optimizing chitinase production parameters.
• Optimized conditions using ANN lead to significantly higher chitinase yields.
• This study provides a pathway for enhanced industrial chitinase production.