Comparative studies on modeling and optimization of fermentation process conditions for fungal asparaginase production using artificial intelligence and machine learning techniques

Abstract

The L-asparaginase is commercial enzyme used as chemotherapeutic agent in cancer treatment and food processing agent in backed and fried food industries. In the present research work, the artificial intelligence and machine learning techniques were employed for modeling and optimization of fermentation process conditions for enhanced production of L-asparaginase by submerged fermentation of Aspergillus terreus. The experimental L-asparaginase activity obtained using central composite experiment design was used for optimization. The Random Forest algorithms machine learning techniques was found best based on the analysis of regression coefficient of ANN model and metric score values of machine learning algorithms. The experimental L-asparaginase activity of 41.58 IU/mL was obtained at the Random Forest algorithm predicted fermentation process conditions of temperature 31 °C, initial pH 6.3, inoculum size 2% (v/v), agitation rate 150 rpm and fermentation time 66 h.