Solving the tree containment problem in linear time for nearly stable phylogenetic networks

Abstract

A phylogenetic network is a rooted acyclic digraph whose leaves are uniquely labeled with a set of taxa. The tree containment problem asks whether or not a phylogenetic network displays a phylogenetic tree over the same set of labeled leaves. It is a fundamental problem arising from validation of phylogenetic network models. The tree containment problem is NP-complete in general. To identify network classes on which the problem is polynomial time solvable, we introduce two classes of networks by generalizations of tree-child networks through vertex stability, namely nearly stable networks and genetically stable networks. Here, we study the combinatorial properties of these two classes of phylogenetic networks. We also develop a linear-time algorithm for solving the tree containment problem on binary nearly stable networks.