桂枝指纹图谱的建立及识别模式研究  

Chemical Fingerprint Establishment and Pattern Recognition for Cinnamomi Ramulus

作　　者：王元清[1] 韩彬[2] 陈婷[3] 贺福元[3] 严建业[3] 

机构地区：[1]中南林业科技大学生命科学与技术学院生物技术与工程实验室,湖南长沙410004 [2]广东药学院中药学院,广东广州510006 [3]湖南中医药大学药学院中药现代化重点实验室,湖南长沙410208

出　　处：《中药材》2017年第7期1622-1627,共6页Journal of Chinese Medicinal Materials

基　　金：国家自然科学基金(81173194);湖南省教育厅科研项目(15C1432);中南林业科技大学人才引进项目(2013RJ026);湖南省中药活性物质筛选工程技术研究中心项目资助(湘科计字[2013]83号);湖南省"中药学"重点学科资助(湘教通[2011]76号)

摘　　要：目的:建立桂枝指纹图谱及总量统计矩分析、聚类分析与主成分分析识别模式,为其质量控制提供新方法。方法:建立11批桂枝饮片的HPLC指纹图谱后采用总量统计矩法定量识别各批样品的总量零阶矩(AUCT)、总量响应率(AUCPWT)、总量一阶矩(MCRTT)、总量二阶矩(VCRTT)等总量统计矩参数;分别采用总量统计矩法、夹角余弦法和相关系数法对各批次样品进行相似度评价;采用聚类分析法与主成分分析法对桂枝样品进行分类识别分析。结果:桂枝HPLC指纹图谱中共标定了11个共有峰,指认了香豆素、肉桂醇、肉桂酸、肉桂醛、2-甲氧基桂皮醛5个成分峰;总量统计矩参数AUCT、AUCPWT、MCRTT、VCRTT值分别为:1.60×10~5AU·min、5.38×10~2AU·min·m L/mg、2.95×10~1min、1.78×101min^2;各批次样品的相似度在0.794以上;聚类分析法与主成分分析均将样品分为4类。结论:桂枝HPLC指纹图谱的建立可行,总量统计矩分析、聚类分析与主成分分析可以识别其指纹图谱模式。Objective： To establish chemical fingerprint and pattern recognition of total statistical moment method,cluster analysis and principal component analysis（PCA） for Cinnamomi Ramulus,and provide a method for quality control of Cinnamomi Ramulus.Methods： Chemical fingerprint of eleven batches of Cinnamomi Ramulus were established by HPLC,and total statistical moment parameters,under curve of total quantum（AUCT）,area under curve of total quantum perweight（AUCPWT）,mean chromatographic retention time of total quantum（MCRTT）,variance of chromatographic retention time of total quantum（VCRTT） were also calculated successfully.The similarities were obtained by total statistical moment method,angle cosine method and correlation coefficient method.Cluster analysis and PCA method were used to classify and recognize the samples.Results： Chemical fingerprint was established with eleven common peaks,and coumarin,cinnamic alcohol,cinnamic acid,cinnamic aldehyde,2-methoxy cinnamic aldehyde were identified.AUCT,AUCPWT,MCRTT,and VCRTT were calculated as 1.60 × 10~5 AU·min,5.38 × 10~2 AU·min·m L/mg,2.95 × 10~1 min,1.78 × 10~1 min^2,respectively.The similarities of all batches of samples were over 0.794.The samples were classified in to four categories by cluster analysis and PCA method.Conclusion： It is feasible to establish chemical fingerprint which were recognized by total statistical moment method,cluster analysis,principal component analysis.

关 键 词：桂枝 指纹图谱 总量统计矩 聚类分析 主成分分析 

分 类 号：R284.3[医药卫生—中药学]