Multilemental Analysis and Classification of Amaranth Seeds According to Their Botanical Origin

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

JOURNAL OF AGRICULTURAL AND FOOD CHEMISTRY

AMER CHEMICAL SOC

Año: 2011 vol. 59 p. 9059 - 9064

0021-8561

The characterization of amaranth seeds (Amaranthus spp.) was developed for A. hypochondriacus, A. cruentus and A. dubius. The elemental concentrations were determined by inductively coupled plasma optic atomic spectroscopy (ICP-OES). Pattern recognition methods were used for the characterization of seed samples: non-supervised methods included Principal Components Analysis (PCA) and Cluster Analysis (CA); supervised methods were Linear Discriminant Analysis (LDA) and Partial Least Squares Discriminant Analysis (PLS-DA). The concentrations of the following elements are informed: Ag, Al, Ba, Ca, Co, Cr, Cu, Fe, K, La, Li, Mg, Mn, Mo, Na, Ni, P, S, Sr, V, Zn and Zr. The lowest mineral content was found in A. hypochondriacus and the highest one in A. dubius. For the classification, selected variables for all multivariate methods were: Ba, Cr, Li, Mn, Ni, S and Sr. Non-supervised methods allowed distinguishing between the three species of amaranth; however PLS-DA supervised methods showed the best prediction ability.