Species tree estimation using Neighbor Joining

TL;DR

This paper compares the accuracy of Neighbor Joining applied to concatenated DNA sequences with other species tree reconstruction methods, finding it to be among the most effective and computationally efficient options.

Contribution

It demonstrates that Neighbor Joining on concatenated sequences is a fast, statistically consistent method for species tree estimation, competitive with more complex algorithms.

Findings

Neighbor Joining with concatenation is highly accurate.

Neighbor Joining is computationally faster than other methods.

It is among the most effective species tree reconstruction methods.

Abstract

Recent theoretical work has demonstrated that Neighbor Joining applied to concatenated DNA sequences is a statistically consistent method of species tree reconstruction. This brief note compares the accuracy of this approach to other popular statistically consistent species tree reconstruction algorithms including ASTRAL-II Neighbor Joining using average gene-tree internode distances (NJst) and SVD-Quartets+PAUP*, as well as concatenation using maximum likelihood (RaxML). We find that the faster Neighbor Joining, applied to concatenated sequences, is among the most effective of these methods for accurate species tree reconstruction.