摘要
建立RPHPLC指纹图谱-化学模式识别评价丹参质量的方法。利用RPHPLC法测定不同产地丹参药材的指纹图谱及丹参酮ⅡA和隐丹参酮的含量;采用主成分分析、系统聚类分析和逐步判别分析对指纹图谱信息进行化学模式识别研究。在主成分分析的基础上,以前4个主成分为聚类分析的指标进行系统聚类分析,取阈值为8时,所有样品可被分为4类;并建立了相应的判别函数,回判准确率100%;以此为依据,初步建立了丹参化学模式识别的评价方法。
High performance liquid chromatography （HPLC） fingerprint and chemical pattern recognition method were developed for the identification and evaluation of Radix Et Rhizoma Salvia Miltiorrhizae. The fingerprints of Radix Et Rhizoma Salvia Mihiorrhizae from nine different sources in Shandong Province and the content of Tanshinone ⅡA and Cryptotanshinone in the herb were determined by HPLC. The characteristic peak data of HPLC were processed with principal component analysis, hierarchical clustering analysis and stepwise discriminant analysis. All the herb samples were divided into 4 classes on the threshold of 8 by hierarchical clustering analysis according to 4 top principal components, meanwhile the recognition function was established for each class by stepwise recognition analysis. The accuracy of classification testing was 100%. The HPLC fingerprint and chemical pattern recognition method was specific and may serve for quality evaluation of Radix Et Rhizoma Salvia Miltiorrhizae.
出处
《化学分析计量》
CAS
2008年第1期24-28,共5页
Chemical Analysis And Meterage
基金
山东省优秀中青年科学家科研奖励基金项目(02B5085)
关键词
丹参
RPHPLC指纹图谱
化学模式识别
主成分分析
系统聚类分析
逐步判别分析
Radix Et Rhizoma Salvia Miltiorrhizae, RPHPLC fingerprint, chemical pattern recognition, hierarchical clustering analysis, principal component analysis, stepwise recognition analysis
