Genome ancestry patterns revealed unknown contributor(s) to cultivated banana

Hybridizations between closely related species are involved in the domestication process of many crops. Banana cultivars are derived from such hybridizations between species and subspecies of the genus Musa that diverged in various regions and archipelagos of tropical South-East Asia. Among the generated diploid and triploid hybrids, those with seedless parthenocarpic fruits were selected by humans and dispersed through vegetative propagation. Musa acuminata subspecies are involved in the vast majority of these cultivars. We analyzed sequence data from 14 M. acuminata wild accessions, 10 M. acuminata-based cultivars including diploids and one triploid to characterize the ancestral contributions along their chromosomes. Using multivariate analysis and SNP clustering, we identified five ancestral groups contributing to these cultivars. Four of them corresponded to known M. acuminata subspecies. A fifth group present only in cultivars was defined based on cultivar '??Pisang Madu'?? and represented two uncharacterized genetic pools. The genome mosaics of cultivars were diverse and more complex than expected with at least three and up to five ancestries, implying multiple hybridization steps. Contributions from at least three known M. acuminata subspecies and from the fifth ancestral group identified here were revealed in the triploid genome of '??Cavendish'??, the most economically important banana cultivar. Our results highlighted the complexity of cultivated banana origins and revealed unknown ancestry contributions suggesting that important banana wild ancestors are still to be identified.