清热类中药微量元素的主成分分析和聚类分析研究

Study on the Trace elements in Heat-clearing Chinese herbal medicines by Principal Component Analysis and Clustering Analysis

[目的]说明微量元素含量与清热类中药疗效之间的相关性。[方法]以微量元素含量为指标,应用主成分分析法,结合SPSS软件对10种清热类中药中的微量元素进行聚类分析。[结果]相关分析结果表明,前4个主因子含有清热类中药微量元素含量83.898%的信息。通过主成分分析建立的主因子模型分别为:F1=-0.051X1-0.249X2-0.700X3+0.594X4+0.881X5-0.331X6+0.835X7+0.796X8+0.196X9+0.094X10+0.551X11,F2=0.651X1-0.793X2-0.212X3+0.656X4-0.066X5+0.126X6-0.314X7-0.124X8+0.774X9+0.084X10+0.688X11,F3=0.344X1+0.418X2-0.343X3+0.433X4+0.029X5-0.208X6-0.354X7-0.392X8+0.081X9+0.830X10+0.079X11和F4=0.331X1-0.119X2+0.296X3+0.127X4+0.124X5+0.809X6+0.191X7+0.334X8+0.083X9+0.444X10-0.375X11。通过分析主因子模型可知,第1主因子与变量Mg、Ca、Cu、Zn、Mn的相关性较高,第2主因子与变量K、Na、Mg、Cr、Ni的相关性较高,第3主因子与变量Mo的相关性较高,第4主因子与变量Fe的相关性较高。在层次聚类分析中,样本聚成4类,样本3为一类,样本9为一类,样本6、7、8为一类,样本1、2、4、5、10为一类。[结论]该研究为清热类中草药的开发利用提供了科学依据和理论基础。

[Objective] The purpose of the study was to illustrate the correlation between microelement contents in heat-clearing Chinese herbal medicines and their curative effects.[Method] With microelement content as index,the microelement contents in the 10 heat-clearing Chinese herbal medicines were analyzed by principal component analysis,clustering analysis and software SPSS.[Result] The results from correlation analysis indicated that the frontal 4 principal factors contained 83.898% information about microelement contents in heat-clearing Chinese herbal medicines.The principal factor models established through principal component analysis were as follows：F1=-0.051X1-0.249X2-0.700X3＋0.594X4＋0.881X5-0.331X6＋0.835X7＋0.796X8＋ 0.196X9＋0.094X10＋0.551X11,F2=0.651X1-0.793X2-0.212X3＋0.656 X4-0.066X5＋0.126X6-0.314X7-0.124X8 ＋0.774X9＋0.084X10＋0.688X11,F3=0.344X1＋0.418X2-0.343X3＋0.433X4＋0.029X5-0.208X6-0.354X7-0.392X8＋0.081X9＋0.830X10＋0.079X11 and F4=0.331X1-0.119X2＋0.296X3＋0.127X4＋0.124X5＋0.809X6＋0.191X7 ＋0.334X8＋0.083X9＋0.444X10-0.375X11.It could be concluded through analyzing principal factor models that the 1st principal factor had better correlation with variables Mg,Ca,Cu,Zn and Mn,the 2nd principal factor had better correlation with variables K,Na,Mg,Cr and Ni,the 3rd principal factor had better correlation with variable Mo and the 4th principal factor had better correlation with variable Fe.In hierarchical clustering analysis,the samples were classified into 4 categories.The samples 3 and 9 were classified into 2 different categories,the samples 6,7 and 8 were classified into a same category and the samples 1,2,4,5 and 10 were classified into a same category.[Conclusion] This study provided scientific basis and theoretical foundation for the exploitation of heat-clearing Chinese herbal medicines.