The Fundamental Role of Character Coding in Bayesian Morphological Phylogenetics.

Area of Science:

• Evolutionary biology
• Phylogenetics
• Computational biology

Background:

• Phylogenetic trees provide historical context for organismal evolution.
• Molecular data models are based on known character state transitions (e.g., nucleotides).
• Morphological data presents challenges due to incomplete sampling and unknown character state spaces.

Purpose of the Study:

• To investigate the impact of model misspecification on phylogenetic tree estimation.
• To explore how observer bias in defining character states affects evolutionary models.
• To quantify the effect of incorrect state space assumptions on phylogenetic accuracy.

Main Methods:

• Simulated morphological character data with varying numbers of character states.
• Estimated phylogenetic trees using evolutionary models with correct and incorrect state space assumptions.
• Analyzed the resulting phylogenetic error, focusing on branch lengths.

Main Results:

• Model misspecification due to observer bias can lead to significant phylogenetic error.
• Assuming a larger state space than actual underestimates branch lengths.
• Assuming a smaller state space than actual overestimates branch lengths.

Conclusions:

• Observer bias in morphological phylogenetics can distort estimates of evolutionary history.
• Accurate character state space definition is crucial for reliable phylogenetic inference.
• Future phylogenetic studies must address potential biases in morphological character data.