论文部分内容阅读
吡咯烷与正丁烷类CCR5(化学趋化因子受体5)拮抗剂可通过抑制人类免疫缺陷病毒(HIV-1)包膜蛋白与CCR5的相互作用而阻断病毒进入细胞.本文使用已知拮抗剂结构和活性信息构建了一个三维药效团模型.按照Catalyst/HypoGen模块的要求,选择了25个结构和活性均具备差异性的分子作为药效团产生的训练集.其中训练集分子以IC_(50)值表示的生物活性值跨度为0.06到10000 nmol·L~(-1).最好的药效团模型(Hypo 1)由两个正离子化特征以及三个疏水特征组成,训练集预测相关系数为0.924,均方根偏差为1.068.模型用于预测由74个分子组成的测试集化合物活性,结果表明模型可以提供较好的活性预测结果并用于新的拮抗剂的设计.
Pyrrolidine and n-butane CCR5 (chemoattractant receptor 5) antagonists can block the entry of viruses into cells by inhibiting the interaction of the human immunodeficiency virus (HIV-1) envelope protein with CCR5. Antagonist structure and activity information to construct a three-dimensional pharmacophore model.According to the requirements of the Catalyst / HypoGen module, select 25 molecules with the same structure and activity as the training set generated by the pharmacophore.The training set molecules The value of IC 50 showed that the span of bioactivity was 0.06 to 10000 nmol·L -1 .The best pharmacophore model (Hypo 1) consisted of two positive ionization features and three hydrophobic features, and the training The set-prediction correlation coefficient was 0.924 and the root-mean-square deviation was 1.068. The model was used to predict the activity of the test-set compounds consisting of 74 molecules and the results showed that the model could provide better activity predictions and be used in the design of new antagonists.