An experimental study of Quartets MaxCut and other supertree methods

Area of Science:

• Phylogenetics and Evolutionary Biology
• Computational Biology
• Bioinformatics

Background:

• Supertree methods are crucial for estimating the Tree of Life.
• Matrix representation with parsimony (MRP) is the dominant algorithmic supertree method despite recent innovations.

Purpose of the Study:

• To evaluate the performance of various supertree methods, including those based on Quartets MaxCut (QMC).
• To compare these methods against MRP and other existing supertree approaches.
• To investigate the impact of taxon sampling and optimality criteria on supertree accuracy.

Main Methods:

• Evaluation of supertree methods based on the Quartets MaxCut (QMC) algorithm.
• Comparison of QMC-based methods with MRP and five other supertree algorithms.
• Analysis of performance under various realistic model conditions and taxon sampling scenarios.

Main Results:

• Two QMC-based supertree methods generally outperformed MRP and other studied methods.
• QMC-based methods exhibit scalability issues, limiting their use on large datasets.
• Taxon sampling significantly affects supertree accuracy; sparse sampling yields poor results.
• Minimizing topological distance to source trees shows weak correlation with supertree accuracy.

Conclusions:

• Development of supertree methods superior to MRP is feasible.
• Scalable and robust implementations of accurate supertree methods are needed.
• Taxon sampling strategies and source tree selection are critical for large phylogenetic tree construction.
• Methodological challenges exist in developing and testing supertree methods due to weak correlation between topological error and distance metrics.