天然产物研究与开发 ›› 2022, Vol. 34 ›› Issue (9): 1557-1564.doi: 10.16333/j.1001-6880.2022.9.013

• 开发研究 • 上一篇    下一篇

基于化学模式识别和熵权TOPSIS法分析木棉花不同部位的差异

曾昭君1,余欣彤1,邓成程2,黎桃敏1,史龙梅3,张兰兰1,刘燎原1*   

  1. 1广东一方制药有限公司 广东省中药配方颗粒企业重点实验室,佛山 528244;2湖南一方天江药业有限公司,常德 415000;3陇西一方制药有限公司,定西 748000
  • 出版日期:2022-09-28 发布日期:2022-10-09
  • 基金资助:
    国家工业和信息化部2019年产业技术基础公共服务平台项目(2019-00902-1-2)

Study on the differences in different parts of Gossampim Flos based on chemical pattern recognition and entropy TOPSIS method

ZENG Zhao-jun1,YU Xin-tong1,DENG Cheng-cheng2,LI Tao-min1,SHI Long-mei3,ZHANG Lan-lan1,LIU Liao-yuan1*   

  1. 1Guangdong Provincial Key Laboratory of Traditional Chinese Medicine Formula, Guangdong Yifang Pharmaceutical Co.,Ltd.,Foshan 528244,China; 2Hunan Yifang Tianjiang Pharmaceutical Co.,Ltd.,Changde 415000,China; 3Longxi Yifang Pharmaceutical Co.,Ltd.,Dingxi 748000,China
  • Online:2022-09-28 Published:2022-10-09

摘要:

基于HPLC指纹图谱比较木棉花不同部位(花瓣、花萼、雄蕊)化学成分的差异性,并综合评价木棉花不同部位的质量。建立木棉花及其不同部位的HPLC指纹图谱,采用相似度评价、化学模式识别和熵权TOPSIS法对其化学成分进行差异性研究。建立了15批木棉花药材的HPLC指纹图谱,确定了13个共有峰,并指认了其中4个特征峰,部分批次花瓣与雄蕊存在特征峰丢失;与药材对照指纹图谱相比,花瓣、花萼、雄蕊的相似度分别为0.280、0.976、0.364。热图聚类分析显示,花瓣和雄蕊聚为一类,化学成分主要富集在花萼中。主成分分析显示,花瓣与雄蕊的化学成分更为接近。正交偏最小二乘法判别分析筛选出7个化学成分是引起木棉花不同部位差异的主要变量。熵权TOPSIS法分析显示,木棉花花瓣、花萼、雄蕊的最优解欧氏贴近度均值分别为0.416、0.485、0.388。木棉花不同部位HPLC指纹图谱存在显著差异,该方法可反映木棉花不同部位质量差异性,对木棉花药材的质量控制和整体评价具有重要意义。

关键词: 木棉花, 不同部位, 相似度评价, 化学模式识别, 熵权TOPSIS法

Abstract:

To compare the chemical constituents of different parts (petal,calyx and stamen) of Gossampim Flos based on HPLC fingerprint,and to evaluate the overall quality of different parts of Gossampim Flos. The HPLC fingerprints of Gossampim Flos and its different parts were established. The differences chemical components of different parts of Gossampim Flos were analyzed by similarity evaluation,chemical pattern recognition and entropy TOPSIS method. HPLC fingerprints of 15 batches of Gossampim Flos herbs contained 13 common peaks,and 4 characteristic peaks were identified. Some batches of petals and stamens had characteristic peaks missing. Compared with the control fingerprint of herb,the similarity of petal,calyx and stamen was 0.280,0.976 and 0.364. Cluster analysis of heat map showed that petals and stamens clustered into one type,and chemical components were mainly concentrated in calyx. Principal component analysis showed that the chemical composition of petals and stamens was closer. Orthogonal partial least square discriminant analysis showed that seven chemical components were the main variables causing the difference of different parts of Gossampim Flos. The entropy TOPSIS method showed that the average values of Ci of petals,calyx and stamens were 0.416,0.485 and 0.388. There are significant differences in HPLC fingerprints of different parts of Gossampim Flos. This method can reflect the quality difference of different parts of Gossampim Flos,which is of great significance for quality control and overall evaluation of Gossampim Flos.

Key words: Gossampim Flos, different part, similarity evaluation, chemical pattern recognition, entropy TOPSIS method

中图分类号:  R284.1