DArT whole genome profiling provides insights on the evolution and taxonomy of edible Banana (Musa spp.)
Review statusPeer Review
MetadataShow full item record
Sardos, J.; Perrier, X.; Dolezel, J.; Hribova, E.; Christelova, P.; Van den houwe, I.; Kilian, A.; Roux, N. (2016) DArT whole genome profiling provides insights on the evolution and taxonomy of edible Banana (Musa spp.) Annals of Botany 118(7) p. 1269-1278 ISSN: 0305-7364
Permanent link to cite or share this item: http://hdl.handle.net/10568/77779
Background and Aims: Dessert and cooking bananas are vegetatively propagated crops of great importance for both the subsistence and the livelihood of people in developing countries. A wide diversity of diploid and triploid cultivars including AA, AB, AS, AT, AAA, AAB, ABB, AAS and AAT genomic constitutions exists. Within each of this genome groups, cultivars are classified into subgroups that are reported to correspond to varieties clonally derived from each other after a single sexual event. The number of those founding events at the basis of the diversity of bananas is a matter of debate. Methods: We analysed a large panel of 575 accessions, 94 wild relatives and 481 cultivated accessions belonging to the section Musa with a set of 498 DArT markers previously developed. Key Results: DArT appeared successful and accurate to describe Musa diversity and help in the resolution of cultivated banana genome constitution and taxonomy, and highlighted discrepancies in the acknowledged classification of some accessions. This study also argues for at least two centres of domestication corresponding to South-East Asia and New Guinea, respectively. Banana domestication in New Guinea probably followed different schemes that those previously reported where hybridization underpins the emergence of edible banana. In addition, our results suggest that not all wild ancestors of bananas are known, especially in M. acuminata subspecies. We also estimate the extent of the two consecutive bottlenecks in edible bananas by evaluating the number of sexual founding events underlying our sets of edible diploids and triploids, respectively. Conclusions: The attribution of clone identity to each sample of the sets allowed the detection of subgroups represented by several sets of clones. Although morphological characterization of some of the accessions is needed to correct potentially erroneous classifications, some of the subgroups seem polyclonal.