Artificial Neural Network for Production of Antioxidant Peptides Derived from Bighead Carp Muscles with Alcalase
Lin Li1,2, Jinshui Wang1, Mouming Zhao1*, Chun Cui1 and Yueming Jiang3
1College of Light Industry and Food Science, South China University of Technology, Guangzhou 510640, PR China
2Zhongshan College, University of Electronic Science and Technology of China, Zhongshan 528402, PR China
3South China Botanical Garden, The Chinese Academy of Science, Guangzhou 510650, PR China
Article history:
Received November 14, 2005
Accepted March 14, 2006
Key words:
antioxidant peptides, artificial neural network (ANN), bighead carp, enzymatic hydrolysis
Summary:
Controlled enzymatic modification proteins are currently being used as good sources of bioactive protein ingredients, and hydrolysates derived from bighead carp muscles may serve as antioxidants through the control of the processing-related parameters. The antioxidant ability was evaluated with regard to the scavenging effect on free radical DPPH·, OH· and O2·–. Due to the robustness, fault tolerance, high computational speed and self-learning ability, artificial neural network (ANN) can be employed to build a predictive model for hydrolysis and optimize the hydrolysis variables: pH, temperature, hydrolysis time, muscle/water ratio and enzyme/substrate ratio (E/S) for the production of antioxidant peptides. Optimum conditions to achieve the maximum antioxidant ability were obtained. The hydrolysates, which scavenged most effectively the DPPH·, OH· and O2·–, were hydrolyzed for 4.8 h with an activity of alcalase of 4.8 AU/kg, for 6 h with 3.84 AU/kg and for 4.3 h with 4.8 AU/kg, at pH=7.5 and 60 °C. Their respective muscle/water ratio was 1:1.9, 1:1.4 and 1:1. The present study confirmed that ANN could be used to simulate the hydrolysis process and predict hydrolysis conditions under which the hydrolysates could show the most effective scavenging ability on DPPH·, OH· and O2·–.
*Corresponding author:
++86 20 87 113 914
++86 20 87 113 914