The artificial intelligence-based chemometrical characterization of genotype/chemotype of Lupinus albus and Lupinus angustifolius permits their identification and potentially their traceability

A chemotyping and genotyping comprehensive approach may be useful for the analytical traceability of food ingredients. The interest for lupin (Lupinus spp.) is developing owing to the high protein percentage as well as the positive technological and nutraceutical properties. The objective was the development of innovative models for discerning between Lupinus albus and Lupinus angustifolius, the most used in human nutrition, by applying multivariate statistical analysis (Principal Component Analysis, PCA) and artificial intelligence (Self Organising Maps, SUMS) onto chemical parameters (proximate composition, alkaloids, tocopherols) or Random Polymorphic DNA fingerprints. The application of PCA to either chemical or genetic data permitted the effective discrimination between the two species, whereas the application of the SUM approach to both data-sets enabled clustering some cultivars. The possibility of discriminating L albus from L angustifolius is relevant for lupin traceability: the foreseen fields of application are refined flours or ingredients, where morphological analysis is not applicable.