A New Quartet Approach for Reconstructing Phylogenetic Trees: Quartet Joining Method

摘要

In this paper we introduce a new quartet-based method for phylogenetic inference. This method concentrates on reconstructing reliable phylogenetic trees while tolerating as many quartet errors as possible. This is achieved by carefully selecting two possible neighbor leaves to merge and assigning weights intelligently to the quartets that contain newly merged leaves. Theoretically we prove that this method will always reconstruct the correct tree when a completely consistent quartet set is given. Intensive computer simulations show that our approach outperforms widely used quartet-based program TREE-PUZZLE in most of cases. Under the circumstance of low quartet accuracy, our method still can outperform distance-based method such as Neighbor-joining. Experiments on the real data set also shows the potential of this method. We also propose a simple technique to improve the quality of quartet set. Using this technique we can improve the results of our method.