论文部分内容阅读
采用分子表面随机采样分析对26个香豆素类衍生物抗艾滋病药物进行定量构效关系研究。运用多元线性回归(multiple linear regression,MLR)建模,同时采用内部及外部双重验证的办法对所得模型稳定性能进行深入分析和检验。MLR建模的复相关系数(_(cum)~2)、留一法(1eave-one-out,LOO)交互校验(cross-validation,CV)复相关系数(Q_(CV)~2)和外部样本校验复相关系数r~2(test)分别为0.936、0.894、0.772。结果表明,RASMS能较好表征香豆素类衍生物抗艾滋病药物分子的结构信息,且所建模型具有良好稳定性和预测能力,可用于研发新型HIV整合酶抑制剂。
The quantitative structure-activity relationship of 26 coumarin derivatives anti-AIDS drugs was studied by random sampling on molecular surface. Using multiple linear regression (MLR) modeling, the stability of the obtained model is analyzed and verified by both internal and external double verification methods. The correlation coefficient (_ (cum) ~ 2) and the complex correlation coefficient (CV_C) of 1eave-one-out (LOO) cross-validation (CV) The external sample calibration test correlation coefficient r ~ 2 (test) were 0.936,0.894,0.772. The results showed that RASMS could better characterize the structural information of coumarin derivatives anti-AIDS drug molecules. The established model has good stability and predictive ability and can be used to develop novel HIV integrase inhibitors.