论文部分内容阅读
采用分子电性距离矢量(molecular electronegativity distance vector,MEDV)表征三嗪类化合物的分子结构和用多元线性回归(multiple linear regression,MLR)建立三嗪类化合物结构与大鼠口服半数致死量(-lgLD_(50))的定量结构-毒性关系(quantitative structure-toxicity relationship,QSTR)模型,同时采用内外双重验证法分析和验证所建模型,建模计算值和留一法交互检验预测值的相关系数R、R_(LOO)分别为0.909和0.813。并用该模型预测三聚氰胺和西玛津的毒性,预测值/实验值分别为-3.52/-3.51和-3.20/-3.70。表明用MEDV表征该类分子的结构信息较好,所建QSTR模型稳定性和预测能力良好。为该类化合物的分子结构设计、毒性研究、食品安全性预测等提供可靠的理论依据。
The molecular structure of triazine compounds was characterized by molecular electronegativity distance vector (MEDV) and the molecular structure of triazine compounds was established with multiple linear regression (MLR) (50)), the quantitative structure-toxicity relationship (QSTR) model was used to analyze and validate the correlation between the calculated model and the one-way reciprocal test predicted value , R_ (LOO) were 0.909 and 0.813 respectively. The model was used to predict the toxicity of melamine and simazine. The predicted values / experimental values were -3.52 / -3.51 and -3.20 / -3.70, respectively. It is indicated that the structure information of these molecules is better with MEDV and the stability and prediction ability of the constructed QSTR model are good. For these compounds molecular structure design, toxicity studies, food safety prediction provide a reliable theoretical basis.