Modeling spatial trends and selecting tropical wheat genotypes in multi-environment trials
Caique Machado e Silva, Victor Silva Signorini, Saulo Fabrício da Silva Chaves, Diana Jhulia Palheta de Souza, Gabriel Wolter Lima, Cleiton Renato Casagrande, Henrique Caletti Mezzomo, João Paulo Oliveira Ribeiro and Maicon Nardino
Abstract: In many cases, traditional analysis of breeding trials based on analysis of variance (ANOVA) do not allow a suitable genetic evaluation. Alternatively, mixed model-based approaches create the possibility of dealing with unbalanced data and modeling spatial trends. The aims of this study were to compare the goodness-of-fit of the model and the genotype ranking through different residual modeling approaches and to select the best performing tropical wheat genotypes based on the best-fitting model. A panel of tropical wheat genotypes was evaluated in three field trials conducted between 2020 and 2021 for grain yield. Linear mixed model analyses were used on the data to estimate the genetic parameters and to predict the genotypic values in analyses of single- and multi-environment trials. Accounting for spatial trends in the analyses of single-and multi-environment trials provides better outcomes than the compound symmetry model does.