小虾
芳香
化学
对虾
食品科学
感官分析
相似性(几何)
感觉系统
人工智能
计算机科学
生物
渔业
图像(数学)
神经科学
作者
Di Zhang,Hongwu Ji,Shucheng Liu,Jing Gao
标识
DOI:10.1016/j.foodres.2020.109517
摘要
The objective of the study was to determine the similarities of aroma attributes in hot-air-dried whole shrimp (HDWS) and its different parts using sensory analysis and gas chromatography–mass spectrometry (GC–MS). In this study, eight samples of different parts of shrimp were compared with its whole. Characteristic volatile compounds based on odor activity value (OAV) were selected, and a cluster analysis (CA) method was used to determine the similarities of aroma profiles between the whole and parts. The potential correlations among sensory attributes and volatile compounds were analyzed by analysis of partial-least-squares regression (PLSR). A total of 78 volatile compounds were identified, and 25 volatile compounds with OAV greater than 1 were selected as the aroma-active compounds (AACs) contributing to samples’ integral aroma profile. Sixteen AACs in HDWS were identified, and four of them made important contributions, namely, 3-ethyl-2,5-dimethylpyrazine, 2,5-dimethyl pyrazide, trimethylamine, and 3-(methylthio)propionaldehyde. Fourteen AACs were the common constituents of hot-air-dried epidermis (HDE) and the whole, and their contents were quite close. CA result showed that HDWS and HDE had the highest similarity and were the first classified in a cluster. AACs summations and sensory scores of hot-air-dried head, shell, and meat decreased to various degrees after removal of shrimp epidermis, the reduction of which was more than 95% in the shell. The PLSR results showed a good correlation between most variables of sensory attributes and volatile compounds. It was concluded that shrimp epidermis made an important contribution to the aroma profile of dried shrimp and was the main source site of shrimp aroma. These results have important theoretical value for reconstructing and producing characteristic flavor substances of shrimp.
科研通智能强力驱动
Strongly Powered by AbleSci AI