Unifying Gene Duplication, Loss, and Coalescence on Phylogenetic Networks

Abstract Statistical methods were recently introduced for inferring phylogenetic networks under the multispecies network coalescent, thus accounting for both reticulation and incomplete lineage sorting. Two evolutionary processes that are ubiquitous across all three domains of life, but are not accounted for by those methods, are gene duplication and loss (GDL). In this work, we devise a three-piece model—phylogenetic network, locus network, and gene tree—that unifies all the aforementioned processes into a single model of how genes evolve in the presence of ILS, GDL, and introgression within the branches of a phylogenetic network. To illustrate the power of this model, we develop an algorithm for estimating the parameters of a phylogenetic network topology under this unified model. The algorithm consists of a set of moves that allow for stochastic search through the parameter space. The challenges with developing such moves stem from the intricate dependencies among the three pieces of the model. We demonstrate the application of the model and the accuracy of the algorithm on simulated as well as biological data. Our work adds to the biologist’s toolbox of methods for phylogenomic inference by accounting for more complex evolutionary processes.