Genetic study and progenies selection of leaf kale half siblings

?rea de concentra??o: Produ??o Vegetal.Objetivou-se com este trabalho selecionar, estudar a associa??o entre caracteres, avaliar a diversidade gen?tica, al?m de realizar estudo de repetibilidade e estabiliza??o fenot?pica em prog?nies de meios-irm?os de couve de folhas. O experimento foi conduzido nos anos de 2015/2016, sendo avaliada uma popula??o formada por 33 prog?nies de meios-irm?os e tr?s cultivares comerciais, utilizadas como testemunhas. O delineamento utilizado foi o de blocos aos acaso com quatro repeti??es e seis plantas por parcela. Al?m da produtividade de folhas, avaliou-se em cada planta caracteres morfoagron?micos de ordem quantitativa e qualitativa. A sele??o das prog?nies foi realizada via modelos mistos (REML/BLUP), al?m do estudo de correla??es gen?tica entre caracteres. O estudo de diverg?ncia gen?tica foi realizado utilizando-se vari?veis can?nicas e agrupamentos pelo m?todo UPGMA. J? o estudo de repetibilidade e estabiliza??o fenot?pica foi realizado estimando-se os par?metros gen?ticos ??a, ??g, ??e, coeficiente de varia??o ambiental e repetibilidade, pelas metodologias frequentista e bayesiana. Para avaliar a confiabilidade destas estimativas, obtiveram-se intervalos por m?todos frequentista, bayesiano e bootstrap (percentil, BCPB e BCa). Verificou-se que maiores ganhos gen?ticos s?o obtidos a partir da sele??o baseada no n?mero de folhas e que o maior grau de associa??o entre caracter?sticas foi estabelecido entre o n?mero de folhas com a produtividade de folhas e o di?metro do caule. Verificou-se a que h? diverg?ncia gen?tica entre os gen?tipos em estudo, sendo as prog?nies P1, P2, P6, P8, P11, P12, P13, P15, P17, P19, P20, P22, P21, P23, P24, P25, P26, P29 e P30 as mais indicadas para posteriores recombina??es. Observou-se que a sele??o confi?vel de prog?nies de meios-irm?os de couve de folhas pode ser realizada com quatro colheitas e que as metodologias frequentista e bayesianas s?o mais indicadas para obten??o da confiabilidade das estimativas dos par?metros gen?ticos avaliados.Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico (CNPq)Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES)Funda??o de Amparo ? Pesquisa do estado de Minas Gerais (FAPEMIG)Tese (Doutorado) ? Programa de P?s-Gradua??o em Produ??o Vegetal, Universidade Federal dos Vales do Jequitinhonha e Mucuri, 2018.The objective of this work was to select and study the association between characters, to evaluate the genetic diversity, besides performing a study of repeatability and phenotypic stabilization in progenies of leaf kale half-siblings. The experiment was conducted in the years 2015/2016, with a population of 33 progenies of half siblings and three commercial cultivars used as controls. The design used was randomized blocks with four replications and six plants per plot. In addition to leaf productivity, morphagronic characters of quantitative and qualitative order were evaluated in each plant. Selection of progenies was performed using mixed models (REML / BLUP), as well as genetic correlations between characters. The study of genetic divergence was performed using canonical variables and groupings by the UPGMA method. The phenotypic repeatability and stabilization study was performed by estimating the genetic parameters ?2a, ??g, ??e, coefficient of environmental variation and repeatability, by frequentist and Bayesian methodologies. To evaluate the reliability of these estimates, intervals were obtained by frequentist, bayesian and bootstrap methods (percentile, BCPB and BCa). It was verified that greater genetic gains are obtained from the selection based on the number of leaves and that the highest degree of association between characteristics was established between the number of leaves with the leaf yield and the diameter of the stem. It was verified that there is genetic divergence between the genotypes under study, being the progenies P1, P2, P6, P8, P11, P12, P13, P15, P17, P19, P20, P22, P21, P23, P24, P25, P26, P29 e P30 the most indicated for later recombination. It was observed that the reliable selection of progenies of leaf kale half siblings can be performed with four harvests and that the frequentist and bayesian methodologies are more indicated to obtain the estimates reliability of the genetic parameters evaluated