论文部分内容阅读
本文建立了2个180个含苯基的羧酸类化合物酸碱解离常数(pKa)的定量预测模型。这些化合物分子量在122.12到288.34的范围内,包含H,C,N,O,S,F,Cl,Br及I等元素.使用Cerius~2程序计算236个分子描述符来表述这些化合物,并使用统计学方法从中选择了12个描述符.分别使用多元线性回归分析(MLR)及支持向量机回归(SVM)结合10重交互检验方法来预测pKa数值.多元线性回归模型对pKa的预测结果相关系数为0.90,标准偏差为0.32;支持向量机模型结果较好,相关系数为0.91,标准偏差为0.31.
In this paper, a quantitative prediction model of acid-base dissociation constants (pKa) for two 180 phenyl-containing carboxylic acids has been established. These compounds have molecular weights in the range of 122.12 to 288.34 and contain elements such as H, C, N, O, S, F, Cl, Br, and I. The 236 molecule descriptors were calculated using the Cerius-2 program to characterize these compounds and used Statistical methods were selected from 12 descriptors.Multiple linear regression analysis (MLR) and support vector machine regression (SVM) combined with 10-fold reciprocal test to predict the pKa values.Multivariate linear regression model pKa prediction results correlation coefficient 0.90 with a standard deviation of 0.32. The support vector machine model showed good results with a correlation coefficient of 0.91 and a standard deviation of 0.31.