The Eberhart and Russel’s bayesian method used as an instrument to select maize hybrids

• Autor(a) principal: Nascimento, Moysés
• Data de Publicação: 2018
• Outros Autores: Oliveira, Tâmara Rebecca Albuquerque de, Carvalho, Hélio Wilson Lemos de, Costa, Emiliano Fernandes Nassau, Amaral Junior, Antonio Teixeira do, Gravina, Geraldo de Amaral, Carvalho Filho, José Luiz Sandes de
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: https://doi.org/10.1007/s10681-018-2146-y; http://www.locus.ufv.br/handle/123456789/22063
• topic: Bayes factor; Genotype × environment interaction; Informative priori; Zea mays L.

Adaptability and stability analysis methods that use a priori information allow identifying and selecting potentially productive genotypes with greater accuracy. The aim of the current study is to use the Eberhart and Russel’ Bayesian method as an instrument to analyze the adaptability and stability of hybrid maize cultivars and to assess the efficiency of using the distribution of informative and non-informative priors to select cultivars. Twenty-five (25) hybrid maize cultivars were assessed in 11 environments located in the Brazilian Northeastern region, during 2012 and 2013, according to a complete randomized block design, with two repetitions. The Eberhart and Russel’s methodology was performed in the GENES software, whereas the Bayesian procedure was implemented in the free software R, by using the MCMCregress function of the MCMCpack package. The adaptability and stability parameters values and the credibility intervals have shown that the Eberhart and Russel’s method via Bayesian technique has shown greater stability-estimation accuracy and greater efficiency in recommending cultivars adapted to favorable and unfavorable environments. The Bayesian methods using priories informative (M1) and few informative (M2) distributions have presented the same genotype classifications in the comparison between a priori distributions; however, according to the Bayes Factor, the M1 was the most adequate distribution to help finding more reliable estimates.