CHEMOMETRIC CLASSIFICATION OF AMARANTH BY USING FATTY ACID COMPOSITION

ELVA G. AGUILAR; MIGUEL A. CANTARELLI; EDUARDO J. MARCHEVSKY; NORA L. ESCUDERO; JOSE M. CAMIÑA

BIOCELL

INST HISTOL EMBRIOL-CONICET

Lugar: Mendoza; Año: 2012 vol. 36 p. 140 - 141

0327-9545

In this work, the composition of fatty acids was analyzed in order to obtain a chemometric model for the classification of amaranth varieties seeds: A. cruentus var. candil (Acc), A. hypochondriacus var. dorado (Ahd), A. hypochondriacus line H17a (AH17a) and A. cruentus line G6/17a (AcG6/17a). Chemometric model was performed by linear discriminant analysis (LDA). The results shows that the discriminant model was able to classify correctly all the studied varieties, with 100% of correct classification in all cases, for training and prediction test set. On the other hand, the discriminant functions plot (function 1 vs function 2) showed similarities between A. cruentus var. candil and A. cruentus line G6/17a, while A. hypochondriacus var. dorado and A. hypochondriacus line H17a showed the most important differences with the others ones. As conclusion, LDA in conjunction to fatty acid composition, were able to distinguish between varieties of amaranth, which can be useful as a non expensive method to classify amaranth seeds.