球形映照方法用于三唑类抗真菌药物活性类别的预报(英文)

Using Sphere-Linear map for the prediction of classes of antifungal activi- ties of Triazole-Derivatives

在进行分子力学和量子化学计算的基础上,采用球形映照(SLM)方法对三唑类衍生物进行了构效关系的研究,建立了三唑类衍生物抗真菌活性类别预报的数学模型,其留一法交叉检验(LOOCV)的预报正确率为88.2%。将该结果与主成分分析(PCA),Fisher,最近邻(KNN)的预报结果进行比较,所得SLM的预报正确率高于PCA,Fisher,KNN的结果。因此,SLM方法有望成为研究药物构效关系的有用工具。

Sphere-linear Map（ SLM）, as a novel technique for data mining, was proposed to build the qualitative structure-activity relationship （qSAR） model for a series of 17 triazole-derivatives. The molecular descriptors were calculated by using the semi-empirical method PM3 and molecular mechanics method MM ＋ supplied by the HyperChem^TM package. By feature selection based on vote method, the feature subset including dipole, density, diameter, dihedral and distance was used for the mathematical modeling. By using the above parameters, SLM method could be used to discriminate between the active and non-active compounds reasonably well, with the accuracy of classification 94.1%. The accuracies of prediction of SLM, principal component analysis （PCA）, Fisher and K-Nearest-Neighbor （KNN） methods were 88.2% , 82.0% , 76. 5% and 58% respectively. The results indicated that the predictive capability of SLM model was better than those of PCA, Fisher and KNN models for this real world data set. Therefore, the SLM model should be useful for the prediction of classes of new triazole-derivatives.