基于人工神经网络的牡蛎蛋白酶解动力模型的构建
Modeling of the Alcalase hydrolysis of oyster proteins using an artificial neural network
-
摘要: 以牡蛎分离蛋白为底物,采用碱性蛋白酶(Alcalase)进行酶解,在获得一定实验数据基础上利用训练后的人工神经网络(ANN)模型对牡蛎分离蛋白酶解过程进行预测。结果表明:训练后的ANN模型决定系数R2达到0.9998,人工神经网络预测水解度DH值和实验DH值之间具有很强的相关性,相关系数r值达到0.9957。并且在一定的酶浓度及底物浓度范围内,采用ANN预测数据,双倒数作图所得回归方程决定系数R2值达到0.9740,计算得米氏常数Km和最大反应速度V max分别为26.1g/L,8.6g/(L·min)。研究结果为人工神经网络模型在食品蛋白酶促反应动力学方面的应用提供参考。Abstract: The oyster protein isolated as substrate was hydrolyzed using alkaline protease in this paper. And the experimentally determined values for the degree of hydrolysis ( DH) were used with an artificial neural network ( ANN) model to predict the Alcalase hydrolysis of the oyster protein isolated.The results showed:the coefficient of determination R2value of artificial neural network model reached 0.9998 after training. Very strong correlations were found between the calculated DH% values obtained from the ANN and those experimentally determined at the temperature, the correlation coefficient R value reached 0.9957. And in a certain range of substrate and enzyme concentration, using the ANN forecast datas, the coefficient of determination R2value of the regression equation obtained reciprocal plot reached 0.9740, the calculated kinetic constants Km and Vmax were 26.1g /L and 8.6g / ( L·min) , respectively.The results provided a reference for the artificial neural network model application in the food protease kinetic.