物理化学学报 >> 2007, Vol. 23 >> Issue (08): 1141-1144.doi: 10.1016/S1872-1508(07)60058-8

研究论文 上一篇    下一篇

用支持向量机建立中药有效成分聚集体的预测模型

黄钦; 庄艳; 乔学斌; 徐筱杰   

  1. 北京大学化学与分子工程学院, 北京 100871
  • 收稿日期:2007-02-09 修回日期:2007-03-26 发布日期:2007-08-03
  • 通讯作者: 徐筱杰 E-mail:xiaojxu@pku.edu.cn

Predicted Model of Aggregation of Molecules in Chinese Herbal Drugs by Support Vector Machines

HUANG Qin; ZHUANG Yan; QIAO Xue-Bin; XU Xiao-Jie   

  1. College of Chemistry and Molecular Engineering, Peking University, Beijing 100871, P. R. China
  • Received:2007-02-09 Revised:2007-03-26 Published:2007-08-03
  • Contact: XU Xiao-Jie E-mail:xiaojxu@pku.edu.cn

摘要: 化合物可以形成聚集体, 这种分子聚集体可能对靶点具有混杂抑制活性. 在中药中已经发现这种现象, 为了进一步研究这种现象,使用支持向量机(SVM)方法建立了分子形成聚集体的分类预测模型. 研究表明, 这个模型具有良好的预测能力, 并且具有稳定性. 通过使用现有化合物对该模型进行验证, 发现该模型具有良好的推广能力. 这个模型被用于对中草药有效成分三维结构与性质数据库(CHDD)中的分子的预测.

关键词: 聚集体, 分类, 支持向量机

Abstract: Some molecules can formaggregates in solution that may inhibit several enzymes, known as promiscuous inhibitors. This phenomenon has been found in traditional Chinese medicinal recipes. To study the aggregation, classification model was constructed by the support vector machine (SVM) classifier. The results indicated that this classification model had good stability and good prediction performance. The experimental results had showed that this predicted model had good generalization. The model was also used to predict the molecules in the Chinese herbal drugs database (CHDD) for further research.

Key words: Aggregation, Classification, Support vector machine

MSC2000: 

  • O641