Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, China West Normal University, Nanchong 637002, China.
Eur J Med Chem. 2010 Sep;45(9):3970-80. doi: 10.1016/j.ejmech.2010.05.052. Epub 2010 Jun 1.
QSAR studies have been carried out on carboxylic acid derivatives as HIV-1 Integrase inhibitors using 3D-MoRSE (3D-Molecular Representation of Structure based on Electron diffraction) descriptors. The stepwise multiple linear regression (stepwise-MLR) and replacement method (RM) methods are used to select descriptors which are responsible for the inhibitory activity of these compounds. Mathematical models are obtained by support vector machine (SVM), back-propagation neural networks (BPNN) and multiple linear regression (MLR). Leave-one-out, Leave-many-out (7% and 18%) cross-validation and external validation are carried out with the aim of evaluating the predictive ability of the models. The values of their respective squared correlations coefficients are 0.731, 0.664, 0.523 and 0.766, respectively. Our best QSAR model reveals the polarizability, mass as the most influencing atomic properties in the structures of the carboxylic acid derivatives.
QSAR 研究已经在 HIV-1 整合酶抑制剂的羧酸衍生物上进行，使用了 3D-MoRSE（基于电子衍射的结构的 3D 分子表示）描述符。逐步多元线性回归（stepwise-MLR）和替换方法（RM）用于选择负责这些化合物抑制活性的描述符。数学模型通过支持向量机（SVM）、反向传播神经网络（BPNN）和多元线性回归（MLR）获得。采用留一法、留多法（7%和 18%）交叉验证和外部验证，以评估模型的预测能力。它们各自的平方相关系数值分别为 0.731、0.664、0.523 和 0.766。我们的最佳 QSAR 模型揭示了在羧酸衍生物结构中，极化率和质量是最具影响力的原子性质。
Zotero 插件