Divergência genética em acessos de feijoeiro comum coletados no estado de Goiás

• Autor(a) principal: Santos, Flávio Pereira dos
• Data de Publicação: 2017
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UFG
• dARK ID: ark:/38995/001300000bgft
• Texto Completo: http://repositorio.bc.ufg.br/tede/handle/tede/7164
• topic: Phaseolus vulgaris L.; Divergencia genética; Modelo de Harrisson; Phaseolus vulgaris L.; Genetic divergence; Harrison model; GENETICA::GENETICA VEGETAL

Common bean (Phaseolus vulgaris L.) is widely cultivated in Brazil, which is the first biggest producer and consumer in the world. The knowledge about the genetic divergence across landraces is very useful for breeders, once it allows them to organize the genetic resources and explore the genetic variability available. The genetic analysis can be done predicting similarities or dissimilarities coefficients that are estimated by morphological differences between accessions. The objective of this study was to identify the diversity across the accessions of common bean that were collected in Goiás State (Brazil), available in Embrapa Rice and Beans’ germplasm bank, through morphological descriptors and agronomic information. The experimental material was composed by 156 common bean accessions. Two experiments were performed, one of them in a greenhouse (morphologic characterization) and the other one in the field (yield evaluation). The experiment in the greenhouse was conducted using two vases with three seeds each one per accessio), without experimental design. The experiment conducted in the field was done under Federer’s Augmented Blocks design, with four blocks of 43 plots (39 accessions + 4 checks). The accessions were characterized by 39 morpho-agronomic qualitative descriptors and ten quantitative descriptors. The quantitative variables were converted in multicategorical variables. The 39 morpho-agronomic qualitative and the ten quantitative descriptors were transformed in binary variables by creating 236 fictitious variables. Though, the similarity matrix was built, using the model proposed by Harrison, which was converted in a dissimilarity matrix. Then the cluster analysis was performed by UPGMA method. The accessions 101 (Rosinha) x 145 (Pintado)) and 120 (Dobra morro) x 152 (Doidão ou bonitão), were the most divergent, because they showed the lowest similarities value, 0,11. The biggest divergences were observed in accession 152, with the similarity coefficients between 0,11 and 0,52. No redundant accessions were found. The pair 86 (Paraná) x 103 (Amarelinho). showed the biggest similarity (0,84). The accessions were clustered in 17 groups, with cophenetic correlation coefficient equal to 0,75, that was significat by Mantel’s test (P < 0,001).The number of accession per group varied from 43 to one. Four groups with only one accessions were formed, which showed the lowest similarities coefficients. No significance was observed for grain yield a mong the accessions, nor between the accession and the commercial checks.