目的 综合评价乌拉尔甘草Glycyrrhiza uralensis和光果甘草G. glabra的质量，通过化学计量学寻找质量差异标志物并建立快速判别乌拉尔甘草和光果甘草模型。方法 建立甘草药材液相指纹图谱，确立共有峰，共有峰数据结合模糊物元模型与偏最小二乘法判别分析（partial least squares discriminant analysis，PLS-DA）化学计量学方法进行质量综合评价及质量差异标志物筛选。基于近红外光谱建立不同的快速判别模型，通过对比筛选出最佳的快速判别模型。结果 模糊物元分析表明乌拉尔甘草与光果甘草存在显著差异；经PLS-DA，结果表明甘草酸铵、甘草素、甘草苷为乌拉尔甘草和光果甘草的差异标志物；光谱经SG预处理，iPLS波段筛选所建立的决策树判别模型，精确率为0.88、准确率为0.88、F1（精确率和准确率的调和平均值）为0.88。结论 乌拉尔甘草与光果甘草之间存在显著差异。通过近红外光谱技术建立的判别模型，为快速区分这2种基原甘草提供了有效手段，有助于提升甘草药材质量控制水平。

Objective To comprehensively evaluate the quality of Glycyrrhiza uralensis Fisch. and Glycyrrhiza glabra L. and to search for quality difference markers through chemometrics and to establish a rapid discriminant model for G. uralensis and G. glabra. Methods HPLC fingerprint of Gancao (Glycyrrhizae Radix et Rhizoma) was established, and the common peak data were combined with fuzzy matter element model and PLS-DA stoichiometric method to evaluate the quality and screen the markers of quality difference. Based on the near infrared spectrum, different rapid discrimination models were established, and the best rapid discrimination models were selected by comparison. Results The fuzzy matter element analysis showed that there were significant differences between G. uralensis and G. glabra. The results of PLS-DA analysis showed that ammonium glycyrrhetate, liquiritigenin and liquiritin were differential markers of G. uralensis and G. glabra. The decision tree discriminant model was established by SG pre-processing and iPLS band screening, and the accuracy rate was 0.88, the accuracy was 0.88, and the F1 was 0.88. Conclusion There is a significant difference between G. uralensis and G. glabra. The discriminant model established by the near-infrared spectroscopy provided an effective method for quickly distinguishing the two kinds of Glycyrrhizae Radix et Rhizoma, and helped to improve the quality control of Glycyrrhizae Radix et Rhizoma.

R286.2

连云港市揭榜挂帅项目（CGJBGS2101）