Study on the structure-activity relationship of new anti-HIV nucleoside derivatives based on the support vector machine method

Jie Wang, Huanxiang Liu, Shen Qin, Xiaojun Yao, Mancang Liu, Zhide Hu, Botao Fan

研究成果: Article同行評審

7 引文 斯高帕斯(Scopus)

摘要

Support Vector Machine (SVM) was used for the classification of the activity of the new anti-HIV nucleosides derivatives for a large and diverse data set using the twelve descriptors that were calculated from the molecular structure alone. The molecular descriptors were selected by the stepwise Linear Discriminant Analysis (LDA) method implemented in SPSS. The correlation between all the descriptors was lower than 0.85. At the same time, in order to build a nonlinear model to classify the new anti-HIV drugs according to their activities, the data set was divided into two subgroups: the training set and the testing set. The nonlinear model gives satisfactory results, which can classify correctly 91.5% of the compounds in the training set and 91.4% of the compounds in the testing set. In addition, this paper provides a new and effective method for classifying the new anti-HIV nucleoside derivatives from their structures according to their activities and gives some insight into structural features related to the activity of the drugs.

原文English
頁(從 - 到)161-172
頁數12
期刊QSAR and Combinatorial Science
26
發行號2
DOIs
出版狀態Published - 2月 2007
對外發佈

指紋

深入研究「Study on the structure-activity relationship of new anti-HIV nucleoside derivatives based on the support vector machine method」主題。共同形成了獨特的指紋。

引用此