基于启发式方法和支持向量机方法预测药物与人血浆蛋白结合率

应用启发式方法和支持向量机方法建立了70种药物与血浆蛋白结合率的定量构效关系模型, 研究了分子结构对药物与血浆蛋白结合率的影响. 两种方法均得到了较好的结果, 交互检验的相关系数平方分别为0.80和0.82; 通过对模型的稳定性和预测能力比较, 支持向量机建立的QSAR模型能够更好地预测药物与血浆蛋白结合率.

Prediction of Binding Rate of Drug to Human Plasma Protein Based on Heuristic Method and Support Vector Machine

The binding rate to human plasma protein for 70 diverse drugs was modeled using the descrip- tors calculated from the molecular structure along with a quantitative structure-activity relationship (QSAR) technique. The heuristic method (HM) and support vector machine (SVM) were utilized to construct the lin- ear and nonlinear prediction models, leading to a good cross-validation correlation coefficient Rc 2v of 0.80 and 0.82, respectively. By comparison the stability with prediction ability of the models, it was found that support vector machine was a good method for predicting the binding rate of drug to human plasma protein.