QSAR技术对高关注化学物质生态环境毒理风险预测

将国外先进的QSAR技术应用于高关注化学物质(SVHC)的生态环境毒理风险预测中，以促进QSAR技术在生态环境毒理学领域的发展和应用. 采用国外先进的经科学验证的ECOSAR预测模型，对欧洲化学品管理署(ECHA)于2008年10月28日正式公布的15种SVHC的生态环境毒理风险进行了预测. 结果显示，对于蒽，4,4′-二氨基二苯甲烷和邻苯二甲酸二丁酯等9种化学物质，ECOSAR模型的预测结果与试验结果较为接近. 危害性分级标准预测结果显示，这9种化学物质为极高风险化学物质，这与ECHA将它们列为SVHC的结论一致. 对于另外6种化学物质(二水合重铬酸钠、三乙基砷酸酯、二氯化钴、五氧化二砷、三氧化二砷和砷酸氢铅)，ECOSAR模型预测结果可信度不高. 对于15种SVHC和ECOSAR的预测结果准确率达到60%. ECOSAR预测模型在一定程度上能够提供与试验结果接近的化学物质生态环境毒理风险评价结果，作为一种重要的非试验科学技术手段，能够在一定程度上满足化学品管理的需要. 

Eco-Environmental Toxicity Risk Prediction for Substances of Very High Concern with QSAR Approach

The advanced Quantitative Structure Activity Relationship (QSAR) approach was used to predict the eco-environmental toxicity of Substances of Very High Concern (SVHC), to promote the development and application of the technique inthe area of eco-environmental toxicology. The eco-environmental toxicity of fifteen kinds of SVHCs included in the list issued by European Chemicals Agency (ECHA) on 28th October 2008 was investigated by validated Ecological Structure Activity Relationship (ECOSAR) models. The results indicated that the predicated eco-environmental toxicity data for nine kinds of SVHCs, including anthracene, 4,4′-diaminodiphenyl methane, dibutyl phthalate, etc.,favorably complied with the experimental data. Furthermore, their toxicity concern levels were assigned as very high risk, which is consistent with the fact that ECHA classified them as SVHC. For the other six kinds of SVHC, including dihydrate sodium dichromate, triethylarsenate, cobalt dichloride, diarsenic pentaoxide, diarsenic trioxide and lead hydrogen arsenate, the confidence of the ECOSAR prediction results was low. Therefore, the prediction accuracy for the eco-environmental toxicity assessment ofthe fifteen kinds of SVHC was 60%. The advanced ECOSAR models could to some extent provide prediction results close to the experimental data, which will fulfillthe need for chemicals management in some non-experimental aspects.