Unsupervised exploratory analyses as a raw material selection tool to develop innovative food products
Artigo de Conferência
Visão geral
Pesquisas
Informação adicional documento
Ver Todos
Visão geral
resumo
The LOCALNUTLEG project aims to valorise local Mediterranean nuts and legumes by developing
innovative plant-based food products. Thus, one of its priorities is to obtain a catalogue with the
complete nutritional and biochemical characterisation of the chosen raw materials, which will
facilitate a better selection. In this sense, the work presented at this congress aimed to determine
the nutritional characterisation and the individual composition of fatty acids and sugars of 6
species of legumes and 19 different varieties. Their centesimal composition was obtained by
AOAC official procedures, while soluble sugars and fatty acids were determined by HPLC-RI and
GC-FID, respectively. A comparative study of the nutritional profiles of each variety was carried out
through two unsupervised exploratory analyses, principal component analysis (PCA) and
hierarchical cluster analysis (HCA). The PCA revealed that within all species of the Leguminosae
family, all chickpea varieties studied (Cicer ariteinum) were associated with high amounts of fibre,
energy, and fat. Carob (Ceratonia siliqua) was correlated with high concentration of carbohydrates,
and all identified free sugars, except for raffinose. All other species and varieties studied were
associated with high protein concentrations, a characteristic feature of Fabaceae. Similarly, to the
PCA, HCA analysis using the variables from nutritional composition classified the species studied
into three large groups. Chickpea varieties were compiled into one large group, the carob into
another, and the remaining species and varieties were relegated to the third group. Therefore,
depending on the type of new food to develop, one could choose carob, to produce food products
rich in carbohydrates and sugars, chickpea to obtain a high calorie, fat, and fibre product, or any
other species for a high protein product. In conclusion, these statistical techniques can be
successfully used to assist in the identification of the best raw material to create new plant-based
food products.