Supervised chemical pattern recognition in almond (Prunus dulcis) Portuguese PDO cultivars: PCA- and LDA-based triennial study uri icon


  • Almonds harvested in three years in Trás-os-Montes (Portugal) were characterized to find differences among Protected Designation of Origin (PDO) Amêndoa Douro and commercial non-PDO cultivars. Nutritional parameters, fiber (neutral and acid detergent fibers, acid detergent lignin and cellulose), fatty acids, triacylglycerols (TAG) and tocopherols were evaluated. Fat was the major component, followed by carbohydrates, protein and moisture. Fatty acids were mostly detected as monounsaturated and polyunsaturated forms, with relevance of oleic and linoleic acids. Accordingly, 1,2,3-trioleoylglycerol and 1,2-dioleoyl-3-linoleoylglycerol were the major TAG. α-Tocopherol was the leading tocopherol. To verify statistical differences among PDO and non-PDO cultivars independently of the harvesting year, data were analyzed through an analysis of variance, a principal components analysis and a linear discriminant analysis (LDA). These differences identified classification parameters, providing an important tool for authenticity purposes. The best results were achieved with TAG analysis coupled with LDA, that proved its effectiveness to discriminate almond cultivars.

publication date

  • January 1, 2012