Answers are grounded in this paper and cite their sources.
Fatty acid composition of seafood can provide insights into its nutritional value and tools for traceability analysis. In this study, the fillet fatty acid compositions of wild-caught gilthead sea bream and specimens produced under cage-reared and pond-reared aquaculture regimes in Northeast Atlantic waters were evaluated to (i) assess nutritional quality using lipid quality indices and (ii) test the traceability discrimination potential of fatty acids using machine-learning classification (Random Forest) and Shapley additive explanations (SHAP) analysis. Distinct fatty acid profiles and total lipid contents were observed between all groups. Although wild-caught fish exhibited higher relative abundance of nutritionally important marine fatty acids EPA and DHA, the greater total lipid content of farmed specimens yielded comparable or higher absolute levels of these omega-3 fatty acids. Mid-Atlantic cage-reared fish contained the highest total fatty acid and EPA+DHA contents, whereas coastal cage-reared specimens displayed favorable cardiovascular health indices. The traceability model achieved high accuracy during the training (100%) and validation phases (99.1%), with higher classification accuracy for wild-caught and pond-reared specimens, followed by coastal and Mid-Atlantic cage-reared specimens. Notably, this methodological approach identified key fatty acids, driving discrimination between groups, providing biologically meaningful markers likely linked to feeding regimes and production practices. Overall, fatty acid profiling coupled with machine learning provides a highly accurate framework for discriminating the production groups examined while delivering consumer-relevant nutritional information and supporting future seafood traceability applications.
No takes yet. Share an insight, caveat, or question.
Feijão et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: