氯代苯及酚类化合物对生物毒性的定量研究

Quantitative research of bio-toxicity of chlorobenzene and substituted phenols compounds

根据分子中原子的价层电子结构特征及参与成键时的特性,新定义了分子隐氢结构图中顶点原子i价点价iδY的计算公式:iYδ=(mi(Zi+hi)+∑(pi+hi)(pj+hj))/(ni2(pi+hi)),以价连接矩阵为基础构建了一个新的能表征含多重键、杂原子分子结构信息的价连接性指数mYt:mYt=∑(iYδδjY)-0.5.mYt(0Y1、Y3、Yc)对这些化合物具有良好的结构选择性,且物理意义明确,计算简单.并运用逐步回归分析方法和最佳变量子集算法对变量进行压缩,通过多元回归方法计算分析,研究了氯代芳烃及取代酚(取代基为F,Cl,Br,I,NH2,NO2,OH,COOH和CH3等)类化合物对浮萍、黑曲酶、细胞色素P-450、发光菌、酵母菌、大型蚤、鱼、蝌蚪和Guppy等动植物的毒性,获得了比较好的相关模型,其计算值与实验值比较吻合.

The valence delta δiY based on the structural characteristic bonding atom i and its chemical surroundings of molecule is defined. Based on the adjacency index, a novel molecular structure information connectivity index mYi is proposed：mYi=∑（δiYδjY）-0.5.It is found that mYi（0Y、1Y、3Yc）have a good structure selectivity to these compounds and a correlation to the bioactivity（bio-toxicity） e.g. Asperillus niger, Duckweed, Cytochrome P-450, Saccharomyces cereuisiae, Golden orfe, Microtox, Pimephales,Tadpok and Guppy et. For this new models, calculation is convenient and the results is accurate and has clear physical significance. The topological indices have been used in QSAR study of chlorobenzene and substituted phenols （containing substitutes such as F, Cl, Br, I, NH2, NO2, OH, COOH and alkyl, et.） algorithms of ＂Stepwise regression＂and＂Leaps and bounds regresslon＂are performed for the selection of variables. The results obtained by using the multi-regression are better.