风味物质香味强度的定量构效关系模型

QSAR Model for Prediction of the Aroma Intensities of Volatile Compounds in Chinese Rice Wine

基于分子拓扑理论,计算黄酒中13种风味物质分子的电性距离矢量(Mk^d)。运用最佳变量子集回归方法,建立上述物质的香味强度与Mk^d的定量构效关系(QSAR)模型。该三元(M2~9、M1~2、M9~9)模型的判定系数(R2)为0.957,估计标准误差(s)为0.600。经R2、F、VIF检验,表明该模型具有良好的稳健性及预测能力。根据进入模型的M2~9、M1~2、M9~9参数可知,影响风味物质香味强度的主要因素是分子的疏水性与亲水性。

Electronegativity distance vectors （Mk-d） of 13 volatile aroma compounds in Chinese yellow rice wine were calculated based on the molecular topology theory in this work. By using Leaps-and-Bounds regression, a satisfactory quantitative structure-activity relationship （QSAR） model has been developed between the aroma intensity （lgU） and Mk-d of these compounds. The correlation coefficient （R2） and the standard deviation （s） of the three-variable model （M2-9, M1-2 and M9-9） were 0.957 and 0.600, respectively. The QSAR model had both favorable estimation stability and good prediction capability as demonstrated by R2, F and VIF tests. From the three variables of the model, it is deduced that the dominant influencing factors of aroma intensity are the hydrophobic and hydrophilic properties of volatile aroma compounds.