应用ICE和PCA方法评价硝基芳烃的综合毒性

应用种间相关估算(ICE)方法预测50种硝基芳烃缺失的生物毒性数据,对通过ICE计算获得的各种生物毒性数据进行主成分分析(PCA),计算综合毒性因子(ITI),进行综合毒性评价.结果表明,除了黄瓜种子发芽抑制毒性以外,其他各种生物毒性都在1%的显著水平上呈显著的线性正相关.硝基芳烃的这些生物毒性机制基本相似,因此应用ICE方法预测其毒性数据是可行的.QSAR分析表明,ITI与分子最低未占轨道能E_lumo有显著的相关性,r=0.869,表明亲电反应性是硝基芳烃的各种生物毒性所综合反映的主要致毒机理.基于ICE和PCA方法计算得到的ITI与基于QSAR和PCA方法计算得到的ITI的大小趋势具有一致性,ICE与PCA方法的联合应用可以成功地评价和预测硝基芳烃的综合毒性.

Evaluation of Integrated Toxicity of Nitroaromatic Compounds by the Combination of ICE and PCA

Interspecies correlation estimation (ICE) method was used to predict missing toxicity data of 50 nitroaromatic compounds. Principle component analysis (PCA) method was applied to calculate their integrated toxicity index (ITI), based on the toxicity data predicted from ICE method. Then the integrated toxicities were evaluated. Significant positive correlation was found among various biological toxicities at 1% significance level, except for phytotoxicity to seed germination of Cucumis sativus. Mechanisms of various biological toxicities of these nitroaromatic compounds are similar, so prediction of missing toxicities is feasible and satisfactory. QSAR analysis shows that ITI is significantly correlated to energy of the lowest unoccupied molecular orbital (E_lumo), and the correlation coefficient is 0.869. Electrophilic reactivity is the main toxicity mechanism that is synthetically reflected by various biological toxicities of nitroaromatic compounds. ITI estimated by ICE and PCA coincides well with ITI predicted by QSAR and PCA. The combination of ICE and PCA can successfully evaluate and predict the integrated toxicity of nitroaromatic compounds.