用QSTR模型预测苯酚类化合物大鼠经口的急性毒性△  

The Oral Toxicity Prediction of Phenols in Rat via QSTR

作　　者：孙也之[1] 闫心丽[2] 李佐静[2] 孟繁浩[1] 

机构地区：[1]中国医科大学药学院,沈阳110001 [2]沈阳药科大学基础学院

出　　处：《数理医药学杂志》2008年第5期513-516,共4页Journal of Mathematical Medicine

基　　金：国家自然科学基金资助项目(30571591)

摘　　要：对39种苯酚类化合物大鼠经口急性毒性(LD50)的定量结构-毒性效应关系(Quantitative Structure-Toxicity Relation-ships,QSTR)进行了研究。采用HyperChem7.5中MOPAC-AMI量子化学法计算了39种苯酚类化合物的13种量子化学结构参数。其中取22个化合物作为样本集对LD50进行多元逐步回归分析,得到最佳方程,该方程的相关系数R2=0.9749。应用所建立的QSTR模型验证了苯酚类化合物的LD50值。经过17个预测样本对该模型进行验证,结果表明该模型具有很好的预测能力。同时分析了苯酚类化合物的毒性机理。Quantitative structure-toxicity relationship （QSTR） were developed to predict the toxicity of phenols in rat via oral LD50 with 13 molecular descriptors. With the HyperChem7.5 software, 13 quantum chemistry parameters for 39 phenols are calculated. The QSTR of 22 phenols is set up which combined with forward regress. The coefficient correlation which is 0. 9749 is discussed with the relationships of LDs0 and their structure parameters. It is shown that the QSTR model has a good stability. The calculated values with the QSTR model have a good consistent with the experimental values and it is tested that the model has a good forecasting ability with other 17 compounds. And we also analyse the toxic mechanism of the phenol compounds.

关 键 词：苯酚类化合物 急性毒性 定量结构-毒性效应关系 

分 类 号：R927.2[医药卫生—药学]