Investigation of fermentation-induced changes in the volatile compounds of Trachinotus ovatus (meixiangyu) based on molecular sensory and interpretable machine-learning techniques: Comparison of different fermentation stages

Abstract

Fermented golden pomfret (Trachinotus ovatus) is appreciated by local consumers owing to its distinct flavor. Electronic nose (E-nose) and gas chromatography-ion mobility spectrometry (GC-IMS) technologies were used to analyze the changes in volatile compounds responsible for evolution of the golden pomfret odor profile during fermentation. Forty-five ion peaks were detected using GC-IMS. Although aldehydes represented the major initial volatile compound group, their levels decreased as fermentation proceeded. Between 3 and 15 days, increased levels of esters contributed to a stable volatile organic compounds profile. After 18 days, high levels of indole and pyrazines were detected. Eleven key volatile compounds were screened based on partial least squares discriminant analysis (PLS-DA). Back propagation artificial neural network (BP-ANN) predicted the fermentation stage enabling the development of better strategies to regulate golden pomfret fermentation. This study provided a theoretical basis for real-time monitoring and quality control of Chinese fermented golden pomfet.

Keywords: Back propagation artificial neural network; Electronic nose; Fermented Trachinotus ovatus; Gas chromatography-ion mobility spectrometry; Quality control; Volatile organic compounds.