Back

FROM FOREST TO SAVANNA AND BACK TO FOREST: EVOLUTIONARY HISTORY OF THE GENUS Dimorphandra (LEGUMINOSAE)

Rocha, V. D. d.; Dal'Sasso, T. C. S.; Williams, C. C. V.; Simon, M. F.; Bueno, M. L.; Oliveira, L. O. d.

2023-01-19 evolutionary biology
10.1101/2023.01.16.524261 bioRxiv
Show abstract

The tree genus Dimorphandra comprises 26 species, which are circumscribed into three subgenera. The subgenus Dimorphandra is associated with both rainforests (Amazon and Atlantic Forest) and savanna-like vegetation (Cerrado); whereas the subgenera Pocillum and Phaneropsia are restricted to the Amazon. We obtained DNA sequence data from six gene regions of the chloroplast genome (cpDNA) and the nuclear internal transcribed spacer (ITS) from 17 species of Dimorphandra and 12 closely related species. Bayesian phylogeny and haplotype network analyses together with both ancestral area reconstructions and ecological niche modeling allowed for exploring the late evolutionary history of the genus Dimorphandra. Species within the subgenus Phaneropsia were more closely related to species of the genus Mora than to the remaining congeners in the plastid tree (but not in the ITS tree), casting doubts on the monophyly of Dimorphandra. Such incongruence may be the result of incomplete lineage sorting of ancient polymorphisms. Amazonian lineages (subgenera Pocillum and Phaneropsia) were highly polymorphic and divergent; whereas lineages from either the Cerrado or the Atlantic Forest were genetically depauperate. The Amazon seems to be the likely source of the lineage that gave rise to the extant species of Dimorphandra of the Cerrado. In turn, a lineage that occupied the Cerrado likely gave rise to the extant species that occur in the Atlantic Forest. Habitat shifts may have been a key driving force that shaped the late evolutionary history of Dimorphandra.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.