Computer-Aided Optimization and Mechanistic Elucidation of the Peanut-Derived Peptides Preparation Process

Peanuts (Arachis hypogaea L.) are one of the important oil crops in China, rich in oil and protein. However, compared to soybean protein, peanut protein has not been widely utilized in the food industry. To improve the comprehensive utilization of peanut protein resources and streamline production processes, this study employed computer-aided response surface methodology to optimize enzymatic hydrolysis parameters using peanut meal as the raw material. Alkaline protease and papain were utilized for stepwise hydrolysis to investigate the preparation of peanut peptides. Machine learning models (random forest regression) were integrated to analyze single-factor experimental data, identifying critical process variables. The optimal conditions for alkaline protease hydrolysis were determined as pH 9.0, 53°C, enzyme dosage of 2083 U/g, and 4-hour hydrolysis at 5% substrate concentration, while papain hydrolysis achieved maximum efficiency at pH 7.0, 55°C, 400 U/g enzyme dosage, and 4-hour reaction. Under these computationally optimized conditions, the peptide content and yield reached 79.88% and 88.83%, respectively, with a 32% reduction in experimental iterations compared to conventional approaches. This methodology provides an efficient strategy for valorizing peanut protein resources in functional food applications.