解表类中药挥发性成分与药性的相关性研究  

Study on the correlation between the volatile components of TCMs for relieving the exterior syndromes and their medicinal properties

作　　者：覃洁萍[1] 刘进[2] 陈玉萍[1] 梁臣艳[1] 李耀华[1] 李毅然[1] 黄艳[1] 何俏明 

机构地区：[1]广西中医药大学药学院,广西南宁530001 [2]广西师范学院数学科学学院,广西南宁530023

出　　处：《计算机与应用化学》2013年第1期85-88,共4页Computers and Applied Chemistry

基　　金：广西自然科学资金资助项目(桂科自1013054)

摘　　要：用支持向量机方法分析280个解表类中药挥发性成分的GC-MS数据,探讨解表药挥发性成分与药性的相关性。以水蒸汽蒸馏法提取各药材的挥发性成分,利用GC-MS方法对其挥发性成分进行分析:对各药材不同类型化学成分的含量进行分类统计,并以不同类型化学成分的含量统计结果作为药性分类的特征指标;采用交叉验证法,利用支持向量机对不同药性解表药的数据进行交叉训练,建立解表药药性的预测模型;该模型对预测集中的辛温类药的正确识别率为95.0%,对辛寒类药的正确识别率为91.7%,总正确率为93.6%。实验结果表明解表类中药挥发性成分与其寒热药性具有较高的相关性,其中以解表类中药中的脂肪族及脂肪酸类成分、单萜氧化物成分对识别辛凉解表与辛温解表两种药性的贡献率最大。To study the correlation between the volatile components of the traditional Chinese medicines（TCMs） for relieving the exterior syndromes and their medicinal properties. 280 GC-MS data of the volatile components from these herbs were analyzed by Support Vector Machine （SVM） method. The volatile components of these herb medicines were extracted with steam distillation and analyzed by means of GC-MS quantitatively and qualitatively. Data statistics of various types of chemical components and their contents were selected as index, trained by means of cross validation method through Support Vector Machine, so as to establish the prediction model of warm-pungent diaphoretic drugs and cold-pungent diaphoretic drugs. The model recognition accuracy for warm-pungent diaphoretic drugs in the prediction set was 95.0%, and 91.7% for the cold-pungent diaphoretic drugs, which made up an accuracy of 93.6% in average. This research shows that the volatile components of TCMs for treating exterior syndromes and their properties have a high correlation.

关 键 词：解表药 挥发性成分 中药药性 支持向量机 

分 类 号：TQ015.9[化学工程] O6-39[理学—化学]