基于指纹图谱及多指标定量结合熵权-TOPSIS模型的不同产地淡豆豉质量评价

曾茂贵; 郑文炜; 张宽; 郑婉晴

doi:10.13748/j.cnki.issn1007-7693.20243456

基于指纹图谱及多指标定量结合熵权-TOPSIS模型的不同产地淡豆豉质量评价

Quality Evaluation of Sojae Semen Praeparatum from Different Origins Based on Fingerprint and Multicomponent Quantification Combined with Entropy Weight-TOPSIS Model

摘要

摘要:
目的基于指纹图谱定性、多指标差异成分定量与化学计量学结合熵权-逼近理想解排序法(TOPSIS)的整合分析方法综合评价不同产地淡豆豉饮片质量。
方法对32批不同产地来源的淡豆豉进行外观性状及薄层色谱鉴别，采用ChromCore 300 C₁₈色谱柱(4.6 mm×250 mm，5 μm)，流动相乙腈-0.1%乙酸水，梯度洗脱，体积流量1.0 mL·min⁻¹，检测波长260 nm，柱温30 ℃，进样量10 μL，建立HPLC指纹图谱并分析相似度；采用聚类分析、主成分分析、正交偏最小二乘法-判别分析筛选差异成分，结合利用熵权-TOPSIS法对淡豆豉质量进行综合评价。
结果各批次淡豆豉均符合2020年版中国药典要求，但饮片质量存在差异。32批淡豆豉的指纹图谱相似度处于0.807~0.990，标定了20个共有峰，化学计量学分析结果一致，可将样品根据不同产地分为6组，依据VIP值筛选出6个差异性指标成分(大豆苷、黄豆黄苷、染料木苷、大豆苷元、黄豆黄素、染料木素)，熵权-TOPSIS法评价表明河北产地S3批次的淡豆豉饮片综合质量最优。
结论建立淡豆豉质量评价方法简便易行，结合化学计量学分析能实现对不同产地淡豆豉的判别归属，通过熵权-TOPSIS模型可评价不同批次淡豆豉综合质量，为其整体质量控制提升提供参考。

Abstract:
OBJECTIVE To comprehensively evaluate the quality of Sojae Semen Praeparatum decoction pieces from different origins using the integrated analytical method based on fingerprint qualitative, multi-index differential component quantification and chemometrics combined with entropy weight TOPSIS method.
METHODS The appearance characteristics and thin-layer chromatography(TLC) identification were carried out for 32 batches of Sojae Semen Praeparatum from different origins. An HPLC fingerprint was established and similarity analysis was performed using ChromCore 300 C₁₈ chromatographic column(4.6 mm×250 mm, 5 μm), acetonitrile-0.1% acetic acid as mobile phase of gradient elution, flow rate of 1.0 mL·min⁻¹, detection wavelength of 260 nm, column temperature of 30 ℃, injection volume of 10 μL. Hierarchical cluster analysis, principal component analysis and orthogonal partial least squares discriminant analysis were used to screen the different components, and entropy weight-TOPSIS was used to evaluate the quality of Sojae Semen Praeparatum.
RESULTS The index results of all batches of Sojae Semen Praeparatum met the requirements of the 2020 edition of Chinese Pharmacopoeia, but there were differences in the quality of decoction pieces. The similarity of HPLC fingerprint of 32 batches of Sojae Semen Praeparatum was 0.807–0.990, 20 common peaks were calibrated. The results of stoichiometric analysis were consistent, the samples could be divided into 6 groups according to different origins, and 6 different index components(daidzin, glycitin, genistin, daidzein, glycitein, genistein) were screened according to VIP value. The evaluation result of entropy weight TOPSIS method showed that the comprehensive quality of S3 batch of Sojae Semen Praeparatum decoction pieces from Hebei origin was the best.
CONCLUSION The established quality evaluation method for Sojae Semen Praeparatum is simple and feasible. Combined with chemometrics analysis, it can achieve the discrimination and attribution of Sojae Semen Praeparatum from different origins. The entropy weight-TOPSIS model can evaluate the comprehensive quality of different batches of Sojae Semen Praeparatum, providing a reference for improving its overall quality control.

HTML全文

参考文献(21)

施引文献

资源附件(0)