Rooting Gene Trees via Phylogenetic Networks

Abstract

AbstractGene trees inferred from alignments of molecular sequences are usually unrooted. Since the root of a gene tree is often the desired property, one of the most classical problems in computational biology is gene tree rooting, where the goal is to infer the most credible rooting edge in an unrooted gene tree. One way to solve it is to apply unrooted reconciliation, where the rooting edge is postulated based on a given split of a rooted species tree. Here, we address a novel variant of the rooting problem, where the gene tree root is inferred using a given phylogenetic network of the species present in the gene tree. One can apply unrooted reconciliation to obtain the best rooting, where the unrooted gene tree is jointly reconciled with a set of splits inferred from the given network. Natural candidates are splits induced by display trees of the network. However, such an approach is computationally prohibiting due to the exponential size of the set. Therefore, we propose a broader and easier-to-control set of splits based on the structural properties of the network. Next, we derive exact mathematical formulas for the rooting problem with the algorithm that runs in square time and space. We verify the algorithm’s quality based on simulated gene trees and networks.