Selection of snap bean F2 progenies for production using the REML/BLUP methodology

This study aimed to select snap bean F-2 progenies for traits related to production of pods and grains using the REML/BLUP procedure. The experiment was carried out at the experimental area of Instituto Federal Fluminense, located in Bom Jesus do Itabapoana, Rio de Janeiro State, Brazil. The 55 genotypes were sown in lines, 42 F2 progenies derived from diallelic crosses, and 13 parental genotypes used as controls. Each line consisted of 24 plants, without replications and at a spacing of 1.0x0.5 m. The plants were harvested and evaluated individually for production of green pods and grains per plant. The selection among and within progenies via BLUP using Selegen-REML/BLUP program was carried out. Based on the Restricted Maximum Likelihood (*REML), we estimated genetic and residual coefficients of variation of 19.43% and 33.53%, respectively, for green pod production and grain production. We used 7.9% of selection intensity, totaling 100 plants selected based on individual genotypic value. The estimated accuracy for the selection of progeny was 0.5014 for pod production and 0.5130 for grain production, indicating difficulties in the selection of these traits due to the high environmental influence on the production traits. Nevertheless, the prediction of genetic values via BLUP, showed the progenies derived from crosses Feltrin x UENF 15-23-4, UENF 7-5-1 x UENF 9-24-2 and Feltrin x UENF 14-3-3 as the most promising progenies with gains of 65.66 g, 61.49 g and 57.63 g, respectively, for pod production and 52.45 g, 46.96 g and 49.29 g, respectively, for grain production. The predicted additive genetic gain with selection was of 36.05% for grain production and 33.5% for green pod production. We concluded that selection via BLUP for production of pods and grains enabled to predict and obtain significant genetic gains for genetic improvement of snap beans for the next generations.