Multispecies coalescent analysis of the early diversification of neotropical primates: phylogenetic inference under strong gene trees/species tree conflict

Abstract

Neotropical primates (NP) are presently distributed in the New World from Mexico to northern Argentina, comprising three large families, Cebidae, Atelidae, and Pitheciidae, consequently to their diversification following their separation from Old World anthropoids near the Eocene/Oligocene boundary, some 40 Ma. The evolution of NP has been intensively investigated in the last decade by studies focusing on their phylogeny and timescale. However, despite major efforts, the phylogenetic relationship between these three major clades and the age of their last common ancestor are still controversial because these inferences were based on limited numbers of loci and dating analyses that did not consider the evolutionary variation associated with the distribution of gene trees within the proposed phylogenies. We show, by multispecies coalescent analyses of selected genome segments, spanning along 92,496,904 bp that the early diversification of extant NP was marked by a 2-fold increase of their effective population size and that Atelids and Cebids are more closely related respective to Pitheciids. The molecular phylogeny of NP has been difficult to solve because of population-level phenomena at the early evolution of the lineage. The association of evolutionary variation with the distribution of gene trees within proposed phylogenies is crucial for distinguishing the mean genetic divergence between species (the mean coalescent time between loci) from speciation time. This approach, based on extensive genomic data provided by new generation DNA sequencing, provides more accurate reconstructions of phylogenies and timescales for all organisms.

Keywords: Miocene; South America; coalescence; primate evolution; species tree.

Fig. 1.—

(a) Size distribution of orthologous alignments; (b) distribution of orthologous alignments with human chromosomes; (c) distribution of the frequency of variable sites per alignment used; (d) average frequency of nucleotides on each species analyzed.

Fig. 2.—

Frequency of gene tree topologies within the inferred species tree.

Fig. 3.—

Distribution of statistical supports of topological NP arrangements. (a) Distribution of aLRT supports along gene trees; (b) distribution of posterior clade probabilities along consensus Bayesian gene trees; (c) violin plot showing distribution of the Bayesian posterior probabilities for the full topologies. In (d), each point of the triangle plot depicts the posterior probability of full topologies for each gene tree. Average values are highlighted.

Fig. 4.—

Timescale of the early NP diversification. ${\Delta }_t$ refers to the time interval between the LCA of living NP and the diversification time between atelids and cebids. D indicates the difference between genetic divergence time and speciation time. The width of internal branches is proportional to the effective size of ancestral populations, which are shown above branches.