HIV-1整合酶链转移抑制剂的QSAR研究与分子设计  

QSAR study on HIV-1 integrase chain transfer inhibitors and their molecular design

作　　者：陈艳[1] 冯惠 周俊[1] 堵锡华[1] Chen Yan;Feng Hui;Zhou Jun;Du Xihua(School of Material and Chemical Engineering,Xuzhou Institute of Technology,Xuzhou 221018,China)

机构地区：[1]徐州工程学院材料与化学工程学院,江苏徐州221018

出　　处：《南京理工大学学报》2021年第6期716-721,共6页Journal of Nanjing University of Science and Technology

基　　金：国家自然科学基金(21472071);江苏省自然科学基金(BK20171169)。

摘　　要：为了研究人类免疫缺陷病毒1型(HIV-1)整合酶链转移抑制剂(INSTIs)抑制活性的定量构效关系(QSAR),从而获得活性更好的抑制剂,基于拓扑理论,计算了32种INSTIs分子的电拓扑状态指数(Ei)和电性距离矢量(Mj)。通过最佳变量子集回归(LBR)方法建立了化合物抑制活性的六元(M57,M14,E7,E13,E21,M36)QSAR模型。以模型中的6个参数为人工神经网络输入层构建反向传播(BP)算法模型,设定6∶2∶1的网络结构,相关系数R2由多元线性回归的0.946提升到0.992。分析模型的6个变量可知,影响INSTIs抑制活性的主要结构片段是>N—、—N—、—C—、—X、—C<、—CH和—OH。通过结构修饰提出3种具有较高抑制活性的新化合物。In order to study the quantitative structure-activity relationship(QSAR)of the inhibitory activity for the human immunodeficiency virus-1(HIV-1)integrase strand transfer inhibitors(INSTIs),the electrotopogical state index(Ei)and electronegativity distance vector(Mj)of 32 INSTIs molecules are calculated by topological method to obtain inhibitors with better activity.The six-parameter(M57,M14,E7,E13,E21,M36)QSAR model for the inhibitory activity of 32 INSTIs compounds is constructed by leaps-and-bounds regression(LBR).The six structural parameters are used as the input neurons of artificial neural network to establish a back propagation(BP)model,and a 6∶2∶1 network architecture is employed.The correlation coefficient R2 increases from 0.946 for multiple linear regression to 0.992 for the model proposed here.By analyzing the six parameters of the model,the dominant influence factors of the inhibitory activity are the molecular structure fragments:>N—,—N—,—C—,—X,—C<,—CH and—OH.According to the results obtained from the structural modifications,three compounds with high inhibitory activity are proposed.

关 键 词：人类免疫缺陷病毒1型 整合酶链转移抑制剂 抑制活性 定量构效关系 分子设计 电拓扑状态指数 电性距离矢量 最佳变量子集回归 

分 类 号：O626.32[理学—有机化学]