基于HS-SPME-GC-MS代谢组学技术分析不同系列白酒特征性化合物

发布时间：2019-01-29 20:55

【摘要】：本文创新性地将SPME-GC-MS检测手段和代谢组学数据处理技术结合应用于白酒特征化合物鉴定和真假区分中。17个白酒样品经顶空固相微萃取,富集酒中挥发性化合物。所得GC-MS数据集经代谢组学技术降维处理,经聚类分析真酒和假酒样品得到了正确的区分,同时其他酒系列所得分类结果基本符合实际酒样信息。在主成分分析中,提取了PC1和PC2两个主成分,解释了不同酒样品特征变量58.7%的方差信息,其中PC1为44.7%,真假酒霍特林椭圆区域区分明显,可视化效果直观。其他系列酒样也保持了对真实酒样信息的吻合。利用偏最小二乘判别分析法建立酒类相关模型,去除基酒样品以突出真假酒的物质区别,得出变量重要性表并且查库得出相关物质十二种,差异最显著的前三位特征化合物是乙酸丁酯、己酸异戊酯和己酸-1-甲乙酯。
[Abstract]:In this paper, SPME-GC-MS and metabolomics data processing techniques were innovatively applied to the identification and identification of liquor characteristic compounds. Seventeen liquor samples were extracted by headspace solid phase microextraction to enrich volatile compounds in liquor. The GC-MS data sets were treated by metabonomics technique, and the samples of real wine and fake wine were correctly distinguished by cluster analysis, and the classification results of other wine series were basically consistent with the actual wine sample information. In principal component analysis, two principal components, PC1 and PC2, were extracted, and the variance information of characteristic variables of different wine samples was explained by 58.7%, in which the PC1 was 44.7, the elliptical region of real and fake wine was distinguished obviously, and the visualization effect was intuitionistic. Other series of wine samples also maintain a true match for the sample information. Using partial least squares discriminant analysis to establish alcohol correlation model, removing base wine samples to highlight the material difference between genuine and fake wines, obtaining the important list of variables and checking the library to find twelve related substances. The top three characteristic compounds were butyl acetate, isoamyl caproate and 1-methyl ethyl caproate.
【作者单位】：赤峰学院附属医院;上海海洋大学食品学院;上海水产品加工及贮藏工程技术研究中心;中国水产科学研究院东海水产研究所;
【基金】：国家自然科学基金资助项目(81341082) 上海市科委工程中心建设(11DZ2280300)
【分类号】：TS262.3;O657.63

【相似文献】