几种金银花的模式识别

目的对金银花的种类进行模式识别 ;寻找与抑菌活性显著相关的化学成分。方法进行高效液相色谱分析 ,获得反映样品整体化学特征的数据 ,利用聚类分析进行分类 ,同时选择体外抑菌指标进行活性测定 ,用线性回归分析揭示化学信息和药理活性之间的相关关系。结果化学特征的系统聚类分析将样品划分为 7类 ,色谱指纹图谱中的 9个成分与抑菌活性相关。结论该方法能较好的识别金银花的种类并可寻找与药理指标相关的化学成分 ,为综合评价金银花质量提供有用的信息

Pattern recognition of several species of Flos Lonicerae

Objective To identify the species of the Flos Lonicerae by the pattern recognition and to investigate the chemical variables related to its bacteriostasis activity.Methods The chemical variables as a whole were obtained by the HPLC analysis.The classification was carried out by using the Hierarchical clustering analysis.The correlation between the variables and the extracorporeal bacteriostasis activity was analysed by the linear regression.Results The 20 samples are identified as 7 categories,according to its chemical features and 9 chemical variables in the HPLC are correlated with the bacteriostasis activity of the samples.Conclusions This method is useful in the species identification of the Flos Lonicerae and in the investigation of the relationship between its chemical variables and pharmacological activity.