The path-label reconciliation (PLR) dissimilarity measure for gene trees.

Area of Science:

• Phylogenetics
• Computational Biology
• Evolutionary Biology

Background:

• Comparing reconciled gene trees is crucial for understanding evolutionary relationships.
• Existing metrics like Robinson-Foulds (RF) and its labeled variants (LRF, ELRF) have limitations in capturing topological and event-based differences.
• A tunable parameter allows adjustment of the balance between species map and event labeling components.

Purpose of the Study:

• Introduce and evaluate the novel Path-Label Reconciliation (PLR) dissimilarity measure for comparing reconciled gene trees.
• Assess PLR's ability to quantify differences in topology, ancestral gene-species maps, and speciation/duplication events.
• Compare PLR's performance against LRF and ELRF in terms of distance distribution, sensitivity, and computational efficiency.

Main Methods:

• Developed the Path-Label Reconciliation (PLR) dissimilarity measure, a novel semi-metric.
• Analyzed the theoretical properties of PLR, including its computation in linear time and its diameter.
• Simulated gene tree reconciliations and compared PLR with LRF and ELRF, also applying it to evaluate gene tree rootings.

Main Results:

• PLR is a computable semi-metric in linear time, offering a more evenly distributed range of distances than LRF/ELRF.
• PLR is less susceptible to overestimating differences from small topological changes and is computationally efficient.
• PLR effectively distinguishes the best gene tree among candidates and theoretical diameters are rarely reached in practice.

Conclusions:

• The PLR measure advances phylogenetic reconciliation by integrating theoretical rigor with practical utility.
• PLR provides a more nuanced comparison of reconciled gene trees, considering both topology and evolutionary events.
• The PLR measure is available in the open-source PyPI package parle for broader application in evolutionary analyses.