基于超快速气相电子鼻对不同产地枸杞子快速识别及气味差异物质研究

  目的  通过气味测定研究, 快速识别不同产地枸杞子及气味差异物质。  方法  收集甘肃、内蒙古、宁夏、新疆及青海等5个产地枸杞子, 利用超快速气相电子鼻对样品进行检测, 建立气味指纹图谱, 并利用AroChemBase数据库对13个共有气味峰进行指认。运用判别因子分析(Discriminant factor analysis, DFA)、卷积神经网络(Convolutional neural networks, CNN)判别模型、主成分分析(Principal component analysis, PCA)及偏最小二乘判别分析(Partial least squares discriminant analysis, PLS-DA)对所得数据进行处理, 并通过变量权重重要性排序(Variable importance in projection, VIP)值作图筛选差异成分。  结果  不同产地枸杞子能区分开, 气味色谱峰峰8(己醛)、峰9(糠醛)为5个产地枸杞子样品间的主要差异成分, 己醛含量排序为: 内蒙古>甘肃>宁夏>新疆>青海; 糠醛含量排序为: 甘肃>新疆>宁夏>青海>内蒙古。  结论  超快速气相电子鼻可较好分析枸杞子气味特征, 并快速、准确识别差异物质, 为不同产地枸杞子的用药选择和综合利用提供参考。 

Rapid Identification and Odor Differential Markers of Distinct Origin-Derived Lycium barbarum L. Based on Ultra-Fast Gas Phase Electronic Nose

  OBJECTIVE  To identify origin-derived Lycium barbarum L. and its odor differential markers via detecting odors.  METHODS  The Lycium barbarum L. selected from five distinct origins (Gansu, Neimenggu, Ningxia, Xinjiang and Qinghai) was detected based on ultra-fast gas phase electronic nose, and the odor fingerprint was subsequently built. Thirteen shared odor peaks were identified through AroChemBase. The discriminant models of discriminant facor analysis (DFA), convolutional neural networks (CNN), principal component analysis (PCA) combined with partial least squares discriminant analysis (PLS-DA) were applied to analyze data. The differential components of distinct origin-derived Lycium barbarum L. were analyzed based on variable importance in projection (VIP).  RESULTS  The 56 batches of Lycium barbarum L. from distinct origins were distinguished. Besides, peak 8 (hexanal) and peak 9 (furfural) were found as the main differential components of the five distinct origin-derived Lycium barbarum L. through variable importance in projection (VIP). The content of hexanal was Neimenggu>Gansu>Ningxia>Xinjiang>Qinghai, and the content of furfural was Gansu>Xinjiang>Ningxia>Qinghai>Neimenggu.  CONCLUSION  This study found that the odor characteristic and differential markers of Lycium barbarum L. could be analyzed through Heracles NEO ultra-fast gas phase electronic nose, providing references for choosing and utilizing Lycium barbarum L. from different origins comprehensively.