Phylogenomics resolve the systematics and biogeography of the ant tribe Myrmicini and tribal relationships within the hyperdiverse ant subfamily Myrmicinae

Abstract

Ants are a globally distributed and highly diverse group of eusocial animals, playing key ecological roles in most of the world’s terrestrial ecosystems. Our understanding of the processes involved in the evolution this diverse family is contingent upon our knowledge of the phylogeny of the ants. While relationships among most subfamilies have come into resolution recently, several of the tribal relationships within the hyperdiverse subfamily Myrmicinae persistently conflict between or within studies, mirroring the controversial relationships of the Leptanillinae and Martialinae to the remaining ant subfamilies. Another persistent issue of debate in ant phylogenetics is the timing of major evolutionary events as inferred via divergence dating. Here, we test the topology of the myrmicine tribes using genome scale data, inspect gene tree-species tree concordance, and use posterior predictive checks and tests of compositional heterogeneity to infer sequence characteristics which potentially introduce systematic bias in myrmicine tribal topology. Furthermore, we test the placement of the fossil †Manica andrannaeby integrating phylogenomic and morphological data from nearly all species within the genusManica,and a broad sampling of its sister genusMyrmica.Subsequently, we demonstrate the effect of fossil placement on overall divergence times in the Myrmicinae. We then re-evaluate the historical biogeography of the Myrmicini and Pogonomyrmecini considering newly generated genetic data and insights from our phylogenomic results. We find that our current understanding of tribal topology in the Myrmicinae is strongly supported, but this topology is highly sensitive to compositional heterogeneity and gene-tree species-tree conflict. Our fossil placement analyses strongly suggest that †Manica andrannaeis a stemManicaspecies, and that placement of this fossil in the crown group affects not only divergence dates within the tribe Myrmicini, but also has broad implications for divergence times throughout the formicoid clade. The results of our biogeographic reconstructions indicate a South American origin for the Pogonomyrmecini + Myrmicini, with the MRCA ofMyrmicainhabiting the western Nearctic in the early Miocene prior to repeated dispersal across Beringia throughout the Miocene and Pliocene. The MRCA ofManica, on the other hand, was inferred to have a Holarctic range prior to vicariance during the Pliocene. Unexpectedly, we found strong support in the Pogonomyrmecini for three coordinated dispersal events from South to Central America during the early Miocene, which has been previously proposed as an early biotic interchange event prior to the more commonly accepted 3.5 Ma closure of the Isthmus of Panama.