When can we reconstruct the ancestral state? Beyond Brownian motion.

Area of Science:

• Evolutionary Biology
• Phylogenetics
• Computational Biology

Background:

• Reconstructing ancestral states is fundamental for understanding evolutionary history.
• Accurate ancestral state estimation is crucial for answering key evolutionary questions.
• Existing methods focus on discrete traits and Brownian motion models.

Purpose of the Study:

• To extend the 'big bang condition' for accurate ancestral state reconstruction.
• To establish conditions for reliable reconstruction across diverse continuous trait evolution models.
• To broaden the applicability of accurate phylogenetic inference methods.

Main Methods:

• Generalizing the 'big bang condition' to continuous trait evolution.
• Analyzing stochastic processes with regularity conditions along phylogenetic trees.
• Verifying conditions for Ornstein-Uhlenbeck, reflected Brownian Motion, bounded Brownian Motion, and Cox-Ingersoll-Ross models.

Main Results:

• The 'big bang condition' is extended to a broad class of continuous trait evolution models.
• The applicability of accurate reconstruction is demonstrated for several popular models.
• A unified framework for assessing reconstruction accuracy is provided.

Conclusions:

• The study provides a general condition for accurate ancestral state reconstruction in continuous trait evolution.
• This work enhances the reliability of phylogenetic analyses involving continuous traits.
• Findings are applicable to a wide range of evolutionary models and biological data.