本文以对在CHO细胞中表达的人类ETB受体有拮抗作用的28个化合物为训练集，利用Catalyst计算所得ETB受体拮抗剂的最优药效团模型由两个氢键受体、一个芳环基团和一个疏水基团组成。最优模型的Fixed cost、Total cost和△Cost值分别为114.792、136.967 和38.081。训练集化合物活性的实测值与预测值相关系数为0.870，偏差值为1.198，有效活性命中率A%和综合评价指数CAI 值分别为84%和2.389。该模型对测试集化合物也有较好的预测能力，所命中测试集化合物的实际活性值和预测活性值的线性相关系数为0.736。研究结果表明所构建的药效团模型具有一定的可靠性，可以用来进行数据库的搜索，寻找新型的ETB 受体拮抗剂。

A three-dimensional pharmacophore model of ETB inhibitors was generated by Catalyst based on 28 ETB inhibitors of human endothelin B (ETB) receptor expressed in CHO cells. The fitting hypothesis, including two hydrogen-bind acceptors, one ring aromatic and one hydrophobic, had a correlation of 0.870, a root mean square (RMS) deviation of 1.198, a Fixed cost of 114.792, a Total cost of 136.967 and a △Cost of 38.081. The values of effectively active hit A% and comprehensive evaluation index CAI are 84% and 2.389, respectively. The model has been further validated by a test set which shows high correlation of 0.736. The results showed that the pharmacophore is reliable which can be used to screen database and find novel ETB inhibitors.

科学技术部国家支撑计划项目（2008BAI51B01）：中药有效成分群辨识技术研究，负责人：张宏桂；国家自然基金项目（81173522）：点线面三维模拟活血化瘀中药防治冠心病的作用机制研究，负责人：乔延江；北京中医药大学自主课题（2011-CXTD-11）：中药信息工程，负责人：乔延江；北京中医药大学自主课题（2011-JDJS-12）：活血化瘀中药治疗冠心病的作用机制研究，负责人：张燕玲。