Publicação

Caracterização e avaliação bioquímica de produtos marinhos para determinação de origem geográfica e qualidade nutricional

Ver documento

Detalhes bibliográficos
Resumo:Seafood fraud, particularly the mislabelling of geographic origin, undermines consumer trust and sustainable fisheries management. This study developed an integrated framework combining consumer insights, nutritional profiling, and amino acid (AA)-based chemometric analysis to authenticate the origin of Trachurus trachurus from the Portuguese coast. A consumer survey (n=102) revealed high awareness of seafood mislabelling (68%) but low selfefficacy in detecting origin (56%), underscoring a clear demand for scientific verification. Biochemically, gravimetric analysis of n-hexane extracts revealed that neutral lipid content varied significantly by region, ranging from 13.5% (Peniche) to 18.2% (Sines) on a dry weight basis. The crude protein content, determined by the Kjeldahl method, was consistently high (76.2–80.2%), with Olhão, Peniche, and Sines forming a distinct high-protein cluster. From 150 individual fish, High-Performance Liquid Chromatography (HPLC) with methanesulfonic acid hydrolysis, optimised to preserve labile AAs, quantified 17 proteinogenic AAs via pre-column derivatization (OPA/FMOC) and fluorescence detection, with pronounced geographical variations observed in glutamic acid, aspartic acid, and proline. Multivariate analysis confirmed distinct location-based clustering in the AA profiles. To build a predictive authentication tool, machine learning models were developed and compared. A supervised Linear Discriminant Analysis (LDA) model achieved 82.7 ± 4.4% cross-validation accuracy, outperforming a Random Forest (RF) classifier (76.0 ± 7.8%), and demonstrated 66.7% accuracy on an independent test set. The top discriminative AAs for origin were glutamic acid, aspartic acid, tyrosine, tryptophan, and isoleucine. These results confirm that AA profiles serve as a robust biochemical fingerprint for geographical authentication. This research provides a practical, science-based framework to enhance traceability systems, thereby supporting sustainable fisheries and restoring consumer confidence in seafood supply chains.
Autores principais:AZEREDO, BEATRIZ JORGE MAGALHÃES DE
Assunto:Trachurus trachurus seafood authentication amino acid profiling machine learning linear discriminant analysis Autenticaçãode pescado Perfil de aminoácidos Análise discriminante linear
Ano:2025
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso embargado
Instituição associada:Instituto Politécnico do Porto
Idioma:inglês
Origem:Repositório Científico do Instituto Politécnico do Porto

Registos relacionados

Registos relacionados