Bigeye tuna (BET, Thunnus obesus) is one of the most nutritious and luxurious cosmopolitan fish. The cooked BET products are capturing the interests of consumers by enhancing flavor and ensuring microbiological safety; however, the lipidomic fingerprints during daily cooking processes have not been investigated. In this work, lipid phenotypic data variation in BET during air-frying, roasting, and boiling was studied comprehensively using iKnife rapid evaporative ionization mass spectrometry (REIMS). The outstanding lipid ions mainly including fatty acids (FAs) and phospholipids (PLs) were identified structurally. It was demonstrated that the rates of heat transfer and lipid oxidation in air-fried BET were slower than those in roasted and boiled BET by elucidating the lipid oxidation and PL hydrolysis mechanism. Furthermore, multivariate REIMS data analysis (e.g., discriminant analysis, support vector machine, neutral network, and machine learning models) was used to characterize the lipid profile change in different cooked BET samples, among which FAC22:6, PL18:3/22:6, PL18:1/22:6, and others were the salient contributing features for determining the cooked BET samples. These results may provide a potential strategy for a healthy diet by controlling and improving functional food quality in daily cooking.
Keywords: daily cooking process; iKnife-REIMS; lipidomic fingerprint; multivariate data analysis.