~1HNMR指纹图谱技术分类及鉴别红景天的品种

Classification and identification of Rhodiola based on ~1HNMR fingerprinting technology

目的采用^1HNMR指纹图谱技术结合相似度考察、化学模式识别分析手段分类及鉴别红景天的品种。方法采用^1HNMR技术测定长鞭红景天和四裂红景天的特征性全成分信息,并将其转化为数据矩阵,考察两种红景天的相似度,采用化学模式识别方法中的主成分分析(PCA)、偏最小二乘法-判别分析(PLS-DA)和聚类分析(HCA)进行化学模式识别分析。结果采用^1HNMR指纹图谱技术结合PCA和PLS-DA能有效地分类及鉴别长鞭红景天和四裂红景天。结论基于。HNMR指纹图谱技术考察长鞭红景天和四裂红景天的相似度和采用HCA难以区分这两种红景天,而基于^1HNMR指纹图谱技术结合PCA和PLS-DA是一种有效的分类及鉴别红景天的方法。所用方法高效、便捷,可为红景天药材的质量控制和品质评价提供参考。

OBJECTIVE To develop an authentication method based on ~1HNMR fingerprinting combined with similarity investigation,chemical pattern recognition analysis,and identification of Rhodiola species. METHODS ~1HNMR technique was used to obtain all component information of R. fastigiata and R. quadrifida,and converted into a data matrix. This is to investigate the similarity of these two species of Rhodiola,consequently using pattern recognitions,such as principal component analysis( PCA),hierarchical cluster analysis( HCA) and partial least squares discriminant analysis( PLS-DA),and also to analyze the data from the ~1HNMR spectra. RESULTS R. fastigiata and R. quadrifida can be discriminated by ~1HNMR fingerprint techniques combined with PCA and PLS-DA in chemical pattern recognition method. CONCLUSION Based on the ~1HNMR fingerprint techniques,the similarity investigation and HCA of R. fastigiata and R. quadrifida were not sufficient to distinguish the two Rhodiola species,but the identification method based on ~1HNMR fingerprint techniques combined with PCA and PLS-DA in chemical model is an effective identification method for distinguishing Rhodiola. This method is efficient and convenient,providing reference for quality control and quality evaluation of Rhodiola.