摘要:本文采用SYBYL-X程序计算了53苯并咪唑类衍生物的化学结构活性参数。运用比较分子力场分析方法(CoMFA)和比较分子相似性指数分析法(CoMSIA)和分子对接方法来进行三文定量构效关系和作用机理的理论研究,获得了它们的结构特征、QSAR、作用位点和可能的作用模式。53个苯并咪唑类衍生物被分为两组:其中41个化合物作为训练集,用三文定量构效关系(3D-QSAR) 方法建立了预测模型,采用抽一法(leave one out , LOO) 检验了预测模型的可信性,12个化合物作为测试集,进一步验证预测模型的预测能力, 用交叉验证系数q2和相关系数R2的值来证实3D-QSAR 预测模型是否有良好的预测能力。研究结果表明我们所建立的3D-QSAR模型在统计学上具有明确、良好的预测能力。5489
关键字: 三文定量构效关系(3D-QSAR);HCV聚合酶抑制剂;2-二取代苯并咪唑-5羧酸;CoMFA ;CoMSIA
3D QSAR studies of benzimidazole derivatives inhibit the hepatitis C virus NS5B
Abstract:Using the SYBYL-X program to calculate the chemical structure-activity parameters of 53 kinds of benzimidazole derivatives. Using the comparative molecular field analysis (CoMFA) index analysis method and comparative molecular similarity (CoMSIA) and molecular docking method for three-dimensional quantitative structure-activity relationship and mechanism of theoretical research, obtained their structure characteristics, QSAR, loci and the possible role model. 53 benzimidazole derivatives were pided into two groups: among 41 compounds as a training set, using three-dimensional quantitative structure-activity relationship (3D-QSAR) method to build a prediction model, using the pumping method (leave one out, LOO) to test the credibility of the forecasting model, and then use a test set by 12 compounds further validate the predictive capability of the prediction model, and using cross-validation coefficient of q2 and the value of the correlation coefficient R2 QSAR prediction model has good predictive ability or not. The results show that the 3D-QSAR model which we built has statistically clear, and good predictive ability. The results show that the 3D-QSAR model which we built has statistically clear, and good predictive ability.