The aim of the work was to use the Bayesian approach, modeling the interaction of coffee genotypes with the environment, using a bisegmented regression to identify stable and adapted genotypes. A group of 43 promising genotypes of Coffea canephora was chosen. The genotypes were arranged in a randomized block design with three replications of seven plants each. The experimental plot was harvested four years in the study period, according to the maturation cycle of each genotype. The proposed Bayesian methodology was implemented in the free program R using rstanarm and coda. After fine adjustments in the approach, it was possible to make inferences about the significant GxE interaction, and to discriminate the coffee genotypes regarding production, adaptability and stability.

Markers are an important tool in plant breeding, which can improve conventional phenotypic breeding, generating more accurate information outcoming better decision making. This study aimed to apply and compare the fit of different Bayesian models BRR, BayesA, BayesB, BayesB (setting the value from very low to Π = 10-5 ) and BayesC and Bayesian Lasso (LASSO) for predictions of the genomic genetic values of productivity and quality traits of a guava population. A randomized block design with two replications was used. Seventeen full-sib families were evaluated. Fruit mass, pulp mass, soluble solids content, number of fruits, and production per plant were measured. These variables were used in the genomic prediction with SSR markers, obtained through the CTAB extraction method with 200 primers.The Bayesian ridge regression model showed the best results for all variables and was chosen to predict the individuals' genomic values according to the cross-validation data.A good stabilization of the Markov and Monte Carlo chains was observed with the mean values corresponding to the observed phenotypic means. Heritabilities showed good predictive accuracy. The model showed strong correlations between some variables, allowing indirect selection.