Estimation of genetic parameters and prediction of additive genetic value for wheat by mixed models

The objective of this work was to estimate the genetic parameters and to predict the genotypic value of populations and individuals from wheat segregating populations, using the methodology of mixed models (restricted maximum likelihood/best linear unbiased prediction, REML/BLUP). Thirty-six wheat segregating populations and four controls were evaluated in the F₃ generation, in a randomized complete block design, with individual information taken from within the plots. The evaluated traits were: grain yield, harvest index, number of tillers, and plant height. Genetic variability between populations was observed for all evaluated traits. The mean heritability varied from 39.15 to 92.78%, and accuracy varied from 62.57 to 96.32% in the selection of populations. The narrow-sense individual heritability was low within populations for all traits. The accuracy for individual selection had a moderate value for plant height, and low values for the other traits. Individual heritability contributes to a greater gain for the traits plant height and harvest index with the use of individual BLUP, in comparison to population BLUP. The segregating populations Embrapa22/BRS207, Embrapa22/VI98053, Embrapa22/IVI01041, BRS254/BRS207, BRS254/VI98053, BRS254/UFVT1Pioneiro, and BRS264/BRS207 stand out with high additive genetic value, for two or more traits.

Triticum aestivum; deviance analysis; unbalanced data; breeding strategies; segregating populations; REML/BLUP