采用SPME-GC-MS结合GC-O技术对不同产区的15个羊肉样品中挥发性成分进行分析鉴定，以系统聚类分析法优选了14个羊肉样品，共筛选出34种检测频率大于50%的共有成分作为羊肉特征风味化合物构建羊肉特征风味指纹图谱。为了考察测定方法的准确性，通过精密度实验、重复性实验和样品的稳定性实验对所建立的指纹分析方法进行了验证，同时进行了相似度分析，14个羊肉样品的相似度均大于0.80，结果表明不同批次的羊肉样品较为一致，该方法符合指纹图谱建立的要求，可以应用于羊肉指纹图谱的构建。依据所建立的指纹图谱信息，采用PCA分析法对其它不同肉类样品和掺假样品进行对照研究，显示羊肉样品和其它肉类样品和掺假样品均存在明显差异，表明该风味指纹图谱可用于羊肉样品质量的鉴定和监控。

Fifteen mutton samples from different districts were analyzed by gas chromatography-mass spectroscopy (GC-MS) and gas chromatography-olfactometry (GC-O), among which 14 samples were selected to establish a flavour fingerprint according to the results of clustering analysis. Thirty-four compounds were identified as odor-active compounds of characteristic information of mutton flavour fingerprint. The validation of method for fingerprint establishment was evaluated by the test of precision, repeatability and stability, and the similarities of the 14 mutton samples were more than 0.80, indicating that the samples from different batches were relatively consistent. These results also indicated that the method was reliable and applicable to establish chromatographic fingerprint of mutton flavor. Based on the fingerprint, it could be easily to differentiate mutton from other meats a0nd fraud samples by principal component analysis (PCA).

国家自然科学基金（31071602），兵团科技攻关项目（2012BA006），兵团兴边富民科技专项（2012BA068）