论文部分内容阅读
以92个具有大麻素受体Ⅰ(CB1)拮抗活性的化合物为训练集,39个化合物为测试集,采用DiscoveryStudio V2.5(DS)软件中的3D构效关系药效团产生(QSAR Pharmacophore Generation)模块建立药效团模型.获得的最佳药效团模型的构成为一个氢键受体(HBA)、一个疏水基团(HY)和二个芳环中心(RA),采用费用函数(Cost function)评价药效团模型,该模型的Δcost为119.32,相关性为0.921,均方根偏差为0.730,Configuration cost为16.1229,表明模型能较好地预测化合物的活性.同时针对目前已知的近450个化合物的12种结构类型进行了探讨,所得结果为进一步设计CB1拮抗剂提供了理论依据.
Ninety-two compounds with cannabinoid receptor I (CB1) antagonistic activity were used as a training set and 39 compounds as test sets. QSAR Pharmacophore Generation was generated using the 3D structure-activity-related pharmacophore in Discovery Studio V2.5 (DS) (Pharmacophore) model was constructed.The optimal pharmacophore model was composed of a hydrogen bond acceptor (HBA), a hydrophobic group (HY) and two aromatic ring centers (RA) function was used to evaluate the pharmacophore model with Δcost of 119.32, correlation of 0.921, root mean square deviation of 0.730, and configuration cost of 16.1229, indicating that the model can predict the activity of the compound well.At the same time, The structure and structure of 450 compounds were investigated. The results provided a theoretical basis for the further design of CB1 antagonists.