RESUMO
The common bean (Phaseolus vulgaris L.) is of great importance to the food and nutritional security of many populations, and exploitation of the crop's genetic diversity is essential for the success of breeding programs. Thus, the aim of the present study was to evaluate the genetic diversity of 215 common bean accessions, which included cultivars, obsolete cultivars, improved lines, and landraces using morpho-agronomic and biochemical traits, and amplified fragment length polymorphism markers (AFLP). Genetic parameters, box plots, Pearson's correlation analysis, and Ward's hierarchical clustering were used to analyze the data. The Jaccard similarity coefficient and neighbour-joining clustering method were used for molecular analysis. A wide variability among the accessions was observed for morpho-agronomic and biochemical traits. Selective accuracy (Ac) and broad-sense heritability (h2) values were high to intermediate for all traits, except seed yield. Ward's hierarchical clustering analysis generated six groups. AFLP analysis also revealed significant differences among the accessions. There was no correlation between the differences based on genetic markers and those based on morpho-agronomic and biochemical data, which indicates that both datasets are important for elucidating the differences among accessions. The results of the present study indicate great genetic diversity among the evaluated accessions.