Bayesian GGE model for heteroscedastic multienvironmental trials.

• Autor(a) principal: OLIVEIRA, L. A. de
• Data de Publicação: 2022
• Outros Autores: SILVA, C. P. da, SILVA, A. Q. da, MENDES, C. T. E., NUVUNGA, J. J., NUNES, J. A. R., PARRELLA, R. A. da C., BALESTE, M., BUENO FILHO, J. S. de S.
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da EMBRAPA (Repository Open Access to Scientific Information from EMBRAPA - Alice)
• Texto Completo: http://www.alice.cnptia.embrapa.br/alice/handle/doc/1142572; https://doi.org/10.1002/csc2.20696
• topic: Interação meio ambiente; Modelo misto; Ensaio de rendimento; Ensaio de cultivar; Estabilidade; Melhoramento Vegetal; Variedade; Genótipo

The dissection of genotype×environment interaction (GEI) is a crucial aspect ofthe final stages of plant breeding pipelines and recommendation of cultivars. Linear-bilinear models used to analyze this interaction, such as the additive main effectsand multiplicative interaction (AMMI) and genotype plus GEI (GGE), often assumehomogeneity of the residual variances across environments which affects the esti-mates and therefore, interpretations and conclusions. Our main objective was topropose a GGE model that considers heteroscedasticity across environments usingBayesian inference and to evaluate its implications in the interpretation of real andsimulated data. The GGE model assuming common variance was also fitted for com-parison purposes. The great flexibility of the Bayesian inference is transferred to thebiplots, allowing the construction of credible regions for genotypic and environmen-tal scores. The inference on the stability and adaptability of genotypes might changewhen heteroscedasticity is ignored. When real data are used, different patterns of cor-relations between environments also affect the representativeness and discriminationof the target environment. The modeling of heteroscedasticity allowed the clusteringof environments into subgroups, with similar effects for GEI. The proposed GGEmodel was more adequate and realistic to deal with scenarios of heterogeneous vari-ance in multienvironment trials, which can be useful for exploiting the GEI.