Background: The advances in NMR and mass spectrometry metabolomics allows a comprehensive profiling of foods, potentially covering geographical origin, authenticity, quality and integrity issues. However, mining specific effects within the corresponding datasets is challenging due to the presence of a set of interacting factors that finally determine metabolomics signatures. Scope and approach: This review provides an overview of the different metabolomics approaches used in food quality and authenticity, then focusing on different chemometric approaches for data interpretation. In particular, data interpretation is hierarchically presented, starting from unsupervised (PCA, hierarchical clusters) to supervised multivariate statistics like OPLS and AMOPLS multiblock ANOVA discriminant approaches. Finally, machine learning approaches like Artificial Neural Networks are discussed as the novel and emerging tool to support food integrity issues. Key findings and conclusions: Tailored data mining approaches are advisable, rather than unique solutions, with unsupervised statistics that naively provide qualitative recognition of patterns, and supervised modeling that support markers identification. Nonetheless, machine learning approaches are emerging as a novel approach able to interpretate complex metabolomics signatures.
Garcia-Perez, P., Becchi, P. P., Zhang, L., Rocchetti, G., Lucini, L., Metabolomics and chemometrics: The next-generation analytical toolkit for the evaluation of food quality and authenticity, <<TRENDS IN FOOD SCIENCE & TECHNOLOGY>>, 2024; 147 (N/A): N/A-N/A. [doi:10.1016/j.tifs.2024.104481] [https://hdl.handle.net/10807/307320]
Metabolomics and chemometrics: The next-generation analytical toolkit for the evaluation of food quality and authenticity
Becchi, Pier Paolo;Zhang, Leilei;Rocchetti, Gabriele;Lucini, Luigi
2024
Abstract
Background: The advances in NMR and mass spectrometry metabolomics allows a comprehensive profiling of foods, potentially covering geographical origin, authenticity, quality and integrity issues. However, mining specific effects within the corresponding datasets is challenging due to the presence of a set of interacting factors that finally determine metabolomics signatures. Scope and approach: This review provides an overview of the different metabolomics approaches used in food quality and authenticity, then focusing on different chemometric approaches for data interpretation. In particular, data interpretation is hierarchically presented, starting from unsupervised (PCA, hierarchical clusters) to supervised multivariate statistics like OPLS and AMOPLS multiblock ANOVA discriminant approaches. Finally, machine learning approaches like Artificial Neural Networks are discussed as the novel and emerging tool to support food integrity issues. Key findings and conclusions: Tailored data mining approaches are advisable, rather than unique solutions, with unsupervised statistics that naively provide qualitative recognition of patterns, and supervised modeling that support markers identification. Nonetheless, machine learning approaches are emerging as a novel approach able to interpretate complex metabolomics signatures.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.