Phylogenomic assessment of the role of hybridization and introgression in trait evolution

Trait evolution among a set of species—a central theme in evolutionary biology—has long been understood and analyzed with respect to a species tree. However, the field of phylogenomics, which has been propelled by advances in sequencing technologies, has ushered in the era of species/gene tree incongruence and, consequently, a more nuanced understanding of trait evolution. For a trait whose states are incongruent with the branching patterns in the species tree, the same state could have arisen independently in different species (homoplasy) or followed the branching patterns of gene trees, incongruent with the species tree (hemiplasy). Another evolutionary process whose extent and significance are better revealed by phylogenomic studies is gene flow between different species. In this work, we present a phylogenomic method for assessing the role of hybridization and introgression in the evolution of polymorphic or monomorphic binary traits. We apply the method to simulated evolutionary scenarios to demonstrate the interplay between the parameters of the evolutionary history and the role of introgression in a binary trait’s evolution (which we call xenoplasy). Very importantly, we demonstrate, including on a biological data set, that inferring a species tree and using it for trait evolution analysis in the presence of gene flow could lead to misleading hypotheses about trait evolution.
Author summary Traits include an organism’s appearance, form, structure, development, physiology, biochemistry and behaviour. They are subject to the same evolutionary processes as their associated genes, including convergence and incomplete lineage sorting. In the former case traits are gained or lost independently in different species, in the latter variation within ancestral species enables a present-day pattern of traits seemingly at odds with the tree of life. Advances in sequencing and new methods to reconstruct evolutionary history have made us increasingly aware of how between-species hybridization results in a network, not a tree, of life. To understand the impact of hybridization on trait evolution we introduce the concept of xenoplasy where present-day traits are shared with ancestral organisms through hybridization instead of strictly tree-like speciation. We have developed a measure called the global xenoplasy risk factor (G-XRF) to quantify the risk that xenoplasy has contributed to a present-day trait pattern, and demonstrate its effectiveness on real and simulated data.