Median Tree Problems Research Articles

Consensus of All Solutions for Intractable Phylogenetic Tree Inference.

Solving median tree problems is a classic approach for inferring species trees from a collection of discordant gene trees. Median tree problems are typically NP-hard and dealt with by local search heuristics. Unfortunately, such heuristics generally lack provable correctness and precision. Algorithmic advances addressing this uncertainty have led to exact dynamic programming formulations suitable to solve a well-studied group of median tree problems for smaller phylogenetic analyses. However, these formulations allow computing only very few optimal species trees out of possibly many such trees, and phylogenetic studies often require the analysis of all optimal solutions through their consensus tree. Here, we describe a significant algorithmic modification of the dynamic programming formulations that compute the cluster counts of all optimal species trees from which various types of consensus trees can be efficiently computed. Through experimental studies, we demonstrate that our parallel implementation of the modified dynamic programming formulation is more efficient than a previous implementation of the original formulation. Finally, we show that the parallel implementation can rapidly identify novel reassorted influenza A viruses potentially facilitating pandemic preparedness efforts.

Exact median-tree inference for unrooted reconciliation costs

BackgroundSolving median tree problems under tree reconciliation costs is a classic and well-studied approach for inferring species trees from collections of discordant gene trees. These problems are NP-hard, and therefore are, in practice, typically addressed by local search heuristics. So far, however, such heuristics lack any provable correctness or precision. Further, even for small phylogenetic studies, it has been demonstrated that local search heuristics may only provide sub-optimal solutions. Obviating such heuristic uncertainties are exact dynamic programming solutions that allow solving tree reconciliation problems for smaller phylogenetic studies. Despite these promises, such exact solutions are only suitable for credibly rooted input gene trees, which constitute only a tiny fraction of the readily available gene trees. Standard gene tree inference approaches provide only unrooted gene trees and accurately rooting such trees is often difficult, if not impossible.ResultsHere, we describe complex dynamic programming solutions that represent the first nonnaïve exact solutions for solving the tree reconciliation problems for unrooted input gene trees. Further, we show that the asymptotic runtime of the proposed solutions does not increase when compared to the most time-efficient dynamic programming solutions for rooted input trees.ConclusionsIn an experimental evaluation, we demonstrate that the described solutions for unrooted gene trees are, like the solutions for rooted input gene trees, suitable for smaller phylogenetic studies. Finally, for the first time, we study the accuracy of classic local search heuristics for unrooted tree reconciliation problems.

Cophenetic Median Trees.

Median tree inference under path-difference metrics has shown great promise for large-scale phylogeny estimation. Similar to these metrics is the family of cophenetic metrics that originates from a classic dendrogram comparison method introduced more than 50 years ago. Despite the appeal of this family of metrics, the problem of computing median trees under cophenetic metrics has not been analyzed. Like other standard median tree problems relevant in practice, as we show here, this problem is also NP-hard. NP-hard median tree problems have been successfully addressed by local search heuristics that are solving thousands of instances of a corresponding (local neighborhood) search problem. For the local neighborhood search problem under a cophenetic metric, the best known (naïve) algorithm has a time complexity that is typically prohibitive for effective heuristic searches. Building on the pioneering work on path-difference median trees, we develop efficient algorithms for Manhattan and Euclidean cophenetic search problems that improve on the naïve solution by a linear and a quadratic factor, respectively. We demonstrate the performance and effectiveness of the resulting heuristic methods in a comparative study using benchmark empirical datasets.

Computing Manhattan Path-Difference Median Trees: A Practical Local Search Approach.

Median tree problems are powerful tools for inferring large-scale phylogenetic trees that hold enormous promise for society at large. Such problems seek a median tree for a given collection of input trees under some problem-specific distance. Here, we introduce a median tree problem under the classic Manhattan path-difference distance. We show that this problem is NP-hard, devise an ILP formulation, and provide an effective local search heuristic that is based on solving a local search problem exactly. Our algorithm for the local search problem improves asymptotically by a factor of $n$n on the best-known (naïve) solution, where $n$n is the overall number of taxa in the input trees. Finally, comparative phylogenetic studies using considerably large empirical data and an accuracy analysis for smaller phylogenetic trees reveal the ability of our novel heuristic.

The asymmetric median tree — A new model for building consensus trees

Inferring the consensus of a set of different evolutionary trees for a given species set is a well-studied problem, for which several different models have been proposed. In this paper, we propose a new optimization problem for consensus tree construction, which we call the asymmetric median tree, (AMT). Our main theoretical result is the equivalence between the asymmetric median tree problem on k trees and the maximum independent set (MIS) problem on k-colored graphs. Although the problem is NP-hard for three or more trees, we have polynomial-time algorithms to construct the AMT for two trees and an approximation algorithm for three or more trees. We define a measure of phylogenetic resolution and show that our algorithms (both exact and approximate) produce consensus trees that on every input are at least as resolved as the standard models in use (strict consensus, majority tree, Nelson tree). Finally, we show that the AMT combines desirable features of many of the standard consensus tree models in use.