论文部分内容阅读
目的建立川产麦冬野生资源的化学模式识别方法。方法采用高效液相色谱法,建立川产麦冬野生资源的HPLC指纹图谱,使用SPSS17.0软件对26批不同来源川产麦冬样品的HPLC指纹图谱进行主成分分析,在主成分得分系数矩阵的基础上,对其进行系统聚类分析;并使用相似度评价软件进行相似度评价验证。结果 26批麦冬样品共提取出4个主成分,被分为4大类;与共有峰直接聚类相比,主成分-聚类分析更符合相似度评价结果。结论利用SPSS软件对麦冬HPLC指纹图谱进行主成分-聚类分析,所建立的模式识别方法,操作简便,统计结果具有可靠性,可对麦冬化学计量学分类及其质量评价提供有效参考。
Objective To establish a chemical pattern recognition method for the wild resources of Ophiopogon japonicus. Methods The HPLC fingerprinting of wild Radix anguillicaudatus was established by HPLC. The principal component analysis was performed on the HPLC fingerprints of 26 samples of Radix Ophiopogon japonicus by SPSS17.0 software. The principal component score coefficient matrix Based on which, the system is clustered and analyzed; and the similarity evaluation software is used to evaluate the similarity. Results A total of 4 principal components were extracted from 26 batches of Radix Ophiopogon japonicus and divided into 4 major categories. Compared with common peak direct clustering, principal component-cluster analysis was more in line with the similarity evaluation results. Conclusion The principal components-cluster analysis of Ophiopogon japonicus HPLC fingerprinting using SPSS software was established. The established pattern recognition method was simple and easy to operate. The statistical results were reliable and could provide an effective reference for the classification and quality evaluation of Ophiopogon japonicus.