Screening method for baking quality assessment in wheat flour using multivariate techniques associated with protein profile of wheat cultivars

Abstract

This work aimed to establish a relationship between the granulometric distribution of wheat proteins by chromatography and electrophoretic profiling and the bread-making quality of the flour using clustering and canonical correlations analysis (CC). Twenty-seven wheat line’s electrophoretic, protein profile, and baking characteristics were determined. The practical application of the CC model was validated using lineage data available in the literature. The cluster analysis promoted the separation into four groups: two with lines with superior technological potential for breadmaking and two with inferior technological potential. CC demonstrated a high correlation between flour protein fraction data and baking quality. The first correlation showed a 0.93 correlation, which makes it possible to classify flours by baking potential with protein fraction data. According to the generated protein profile index, lines with a positive index have a high potential for baking and should remain in the breeding programs. On the other hand, lines with a negative index have a low baking potential and should be removed. 60% of the cultivars analyzed in the external validation were correctly classified, and the poorly classified ones presented characteristics that deviate from normality regarding the ratio between gliadin and glutenin. This suggests that the model can be successfully applied to the selection of lines in breeding programs, provided that they present a ratio between gliadins and glutenins close to 1.2. Thus, this index can be a screening tool for the inclusion of technological quality in breadmaking as a parameter in the selection of new wheat lines.