The prediction of bread wheat quality: joint use of the phenotypic information brought by technological tests and the genetic information brought by HMW and LMW glutenin subunits

Attempts have been made to predict Chopin alveograph or French bread-making tests, using tree-based models and PLS regressions. Data came from three sets of trials, involving 130, 214 and 103 different genotypes, which were described for HMW-GS, LMW-GS and small-scale tests currently used in breeding programs. Segmentation trees and PLS regressions indicated that HMW-GS and LMW-GS were not sufficient to explain alone the variability of bread wheat quality. This could be partly due to “allele × environment” and “locus × locus” interactions. For HMW-GS, Glu-B1 was the predominant locus for alveograph and French bread-baking, and some differences in the alleles hierarchy were demonstrated according to the end-use parameter considered. For LMW-GS, Glu-B3 seemed to be preponderant, with alleles b′, c and g being favourable and allele c′ unfavourable. Joint use of the information brought by glutenin subunits and technological tests did not enable to predict satisfactorily, neither the different parameters of French bread-baking, nor the extensibility L of alveograph. Only the prediction of the strength W proved reliable, and robust PLS equations were proposed for this alveograph parameter. These prediction equations could be of interest to select for high values of W in the mid generations of breeding.