Associação entre caracteres e formação de grupos divergentes em linhagens de milho
Carregando...
Data
2018-06-29
Autores
Marconato, Marcela Bonafin [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A pesquisa teve como principais objetivos estudar a correlação entre caracteres, estimar a capacidade de combinação e discriminar grupos em linhagens de milho. Para isso, foram avaliadas três safras (2015/2016; 2016; 2016/2017), nas quais 85 linhagens de milho foram cruzadas com um testador de base ampla em topcross. O delineamento experimental utilizado foi o de blocos casualizados, com duas repetições. Foram analisados os caracteres produtividade, prolificidade, altura de planta e de espiga, posição relativa de espiga, acamamento e quebramento. O programa computacional GENES foi utilizado para a realização das análises de variância individual e conjunta, bem como análise de trilha. Com o programa computacional Statsoft foram feitas duas análises multivariadas, uma de agrupamento hierárquico de Ward e outra de agrupamento não hierárquico de K-means. A partir dos resultados da análise de trilha, observou-se que nenhuma variável seria adequada para seleção indireta dos genótipos, sendo indicada a seleção pela produtividade. A capacidade geral de combinação para produtividade foi realizada e esta permitiu destacar genótipos superiores e com bom desempenho nas três safras. As análises multivariadas possibilitaram a separação dos genótipos em grupos distintos e foram complementares para direcionar a escolha dos cruzamentos mais divergentes.
The aim of this study was to evaluate the correlation between traits, estimate the combining ability and discriminate groups in maize lineages. For this, three harvest seasons (2015/2016, 2016, 2016/2017) were evaluated, in which 85 maize lineages were crossed with a broad genetic base tester by topcross. The experimental design was a randomized block design, with two replicates. The studied traits were grain yield, prolificacy, plant height and ear height, ear placement, lodging and breaking. The GENES software was used to perform individual and joint analysis of variance, as well as pathway analysis. The Statsoft software was used to perform two multivariate analysis, Ward hierarchical cluster analysis and K-means nonhierarchical cluster analysis. The results from pathway analysis revealed that no variable would be suitable for indirect selection of the genotypes, so that selection by the grain yield is indicated. The general combining ability for grain yield was obtained, highlighted the superior genotypes with good performances in the three harvest seasons. Multivariate analysis allowed the separation of the genotypes into distinct groups and were complementary to point out the most divergent crosses.
The aim of this study was to evaluate the correlation between traits, estimate the combining ability and discriminate groups in maize lineages. For this, three harvest seasons (2015/2016, 2016, 2016/2017) were evaluated, in which 85 maize lineages were crossed with a broad genetic base tester by topcross. The experimental design was a randomized block design, with two replicates. The studied traits were grain yield, prolificacy, plant height and ear height, ear placement, lodging and breaking. The GENES software was used to perform individual and joint analysis of variance, as well as pathway analysis. The Statsoft software was used to perform two multivariate analysis, Ward hierarchical cluster analysis and K-means nonhierarchical cluster analysis. The results from pathway analysis revealed that no variable would be suitable for indirect selection of the genotypes, so that selection by the grain yield is indicated. The general combining ability for grain yield was obtained, highlighted the superior genotypes with good performances in the three harvest seasons. Multivariate analysis allowed the separation of the genotypes into distinct groups and were complementary to point out the most divergent crosses.
Descrição
Palavras-chave
Agrupamento, Análise multivariada, Correlação, Melhoramento de plantas, Zea mays L., Clustering, Multivariate analysis, Correlation, Plant breeding