Genotypic selection of maize varieties via mixed models methodology

Abstract

One of the techniques used that has advanced in breeding programs is the analysis using mixed models. The selection of cultivars using this methodology allows for more accurate and realistic inferences about the genotypic evaluation of the superior genotypes for selection. The objective of this work was to select corn genotypes from four harvests of the Embrapa Acre breeding program, using the mixed model methodology - REML / BLUP. The experiments were conducted in the Experimental field of Embrapa Acre. The design used was in a 5x6 rectangular lattice, with two replications. Grain yield data from 78 varieties tested in the years 2013/2014, 2014/2015, 2015/2016, 2017/2018 were used. The MHVG model was used to estimate the genotypic stability. The Selegem REML / BLUP software was used for statistical-genetic analysis. In the four years it was found that the varieties did not exceed the average production of the hybrids. The ordering of the genotypes coincided in the four years of evaluation, indicating broad behavior of the cultivars. The mixed model methodology was efficient in evaluating the cultivation and use value tests, assisting in the selection and disposal of cultivars each agricultural year. The genotypes BRS 1055 and HTCMS-SP1 are selected.