Support patterns from different outgroups provide a strong phylogenetic signal.

Area of Science:

• Phylogenetics
• Computational Biology
• Evolutionary Biology

Background:

• Phylogenetic reconstruction accuracy can decline with the use of increasingly distant outgroups.
• Existing methods may struggle with phylogenetic artifacts like long-branch attraction.

Purpose of the Study:

• To introduce and validate a novel scoring method, the outgroup score (pOG), for assessing phylogenetic reconstruction accuracy.
• To evaluate the performance of the pOG score in simulations and real-world phylogenetic analyses.

Main Methods:

• Development of the outgroup score (pOG) to measure the decrease in branch support with increasing outgroup distance.
• Large-scale simulations to test the pOG score's ability to capture expected patterns and identify correct topologies.
• Application of the pOG score as an optimization criterion in constructing a mammalian phylogenetic tree.

Main Results:

• The pOG score accurately reflects the expected decrease in phylogenetic support with distant outgroups.
• The pOG method successfully identified correct tree topologies where primary reconstruction methods failed, particularly under model violations.
• Simulations showed the pOG analysis outperformed standard methods in scenarios with rate variation and in quartet-based reconstruction.
• Phylogenetic analysis of 23 mammals using the pOG score supported established clades and proposed novel groupings like Atlantogenata and Artiodactyla-Chiroptera.

Conclusions:

• The pOG score is a robust tool for evaluating phylogenetic accuracy and improving tree reconstruction, especially when evolutionary models are violated.
• This method offers a model-agnostic approach to detect and mitigate phylogenetic reconstruction errors.
• The pOG analysis provides valuable insights into evolutionary relationships, as demonstrated by its application to mammalian phylogeny.