Abstract

SALOMON, N; MIRANDA, R  and  ORTIS, L. Disección de la correlación en variables que miden la aptitud panadera del trigo y modelos predictores de la calidad industrial. BAG, J. basic appl. genet. [online]. 2012, vol.23, n.1, pp.14-24. ISSN 1852-6233.

Baking quality is one out of four objectives in a breeding program for wheat bread. This is a quantitative variable governed by many genes and strongly influenced by environmental conditions. Several authors have examined this complex variable through different statistical methods. In the present study, path analysis was used to build quality models through the dissection of the correlation between two variables and to compare them in four Argentine wheat sub-regions. The model built with 21 variables yielded high coefficients of determination (over 95%) in all sub-regions, except V South subregion (6.47%). To build Model II, those variables showing either high correlation or scarce contribution to Model I were removed so that the coefficient of determination (R2) markedly increased in V South sub-region (from 6.47 to 30%). In the remaining sub-regions this value decreased as follows: II North > II South > IV. This wide variation among R2 values was attributed to the variables that formed each model and to the high environmental influence over them. A high correlation among the same variables was observed in the four sub-regions. As for the contribution of each variable to Model II depending on each sub-region, common variables such as Wet Gluten, Bakery Strength, Index Gluten, Dough Equilibrium and Flour Ash were observed. Other variables were typical of some of the sub-regions. Dissection of each sub-region explained why some variables had low correlations with Loaf Volume while others scarcely contributed to the model. The variables that built Model II in each of the sub-regions could be considered as participants in a breeding index.

Keywords : Path analysis; Bread wheat; Industrial quality.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )