Comparing and Aggregating Partially Resolved Trees

TL;DR

This paper introduces efficient algorithms for measuring similarities between phylogenetic trees, effectively handling unresolved nodes by using triplet and quartet-based distances with parametric and Hausdorff approaches.

Contribution

It presents novel distance measures for rooted and unrooted trees that accommodate unresolved nodes, improving robustness and granularity over traditional methods.

Findings

Effective handling of unresolved nodes in phylogenetic trees.

Introduction of parametric and Hausdorff distance measures.

Algorithms demonstrating computational efficiency.

Abstract

We define, analyze, and give efficient algorithms for two kinds of distance measures for rooted and unrooted phylogenies. For rooted trees, our measures are based on the topologies the input trees induce on triplets; that is, on three-element subsets of the set of species. For unrooted trees, the measures are based on quartets (four-element subsets). Triplet and quartet-based distances provide a robust and fine-grained measure of the similarities between trees. The distinguishing feature of our distance measures relative to traditional quartet and triplet distances is their ability to deal cleanly with the presence of unresolved nodes, also called polytomies. For rooted trees, these are nodes with more than two children; for unrooted trees, they are nodes of degree greater than three. Our first class of measures are parametric distances, where there is a parameter that weighs the difference between an unresolved triplet/quartet topology and a resolved one. Our second class of measures are based on Hausdorff distance. Each tree is viewed as a set of all possible ways in which the tree could be refined to eliminate unresolved nodes. The distance between the original (unresolved) trees is then taken to be the Hausdorff distance between the associated sets of fully resolved trees, where the distance between trees in the sets is the triplet or quartet distance, as appropriate.