Using Avida to test the effects of natural selection on phylogenetic reconstruction methods

Area of Science:

• Evolutionary biology
• Computational biology
• Bioinformatics

Background:

• Phylogenetic trees visualize organismal ancestral relationships using molecular data.
• Various reconstruction algorithms exist, but differing results highlight the need for robust testing.
• Simulations are crucial for evaluating algorithm performance due to limited known phylogenies.

Purpose of the Study:

• To assess the impact of natural selection on the accuracy of phylogenetic tree reconstruction methods.
• To compare the performance of two common tree reconstruction algorithms under simulated evolutionary conditions.
• To identify factors influencing the success or failure of phylogenetic inference.

Main Methods:

• Utilized the Avida digital evolution platform to generate simulated phylogenetic trees.
• Introduced simulated natural selection between branching points in the evolutionary process.
• Applied two widely used phylogenetic reconstruction algorithms to the simulated data.
• Analyzed algorithm performance based on the accuracy of reconstructed trees.

Main Results:

• Phylogenetic reconstruction methods demonstrated high success rates when natural selection was present during evolution, even over extended time scales.
• Conversely, the tested algorithms frequently failed to accurately reconstruct phylogenetic trees in the absence of natural selection.
• The presence or absence of selection significantly influenced the reliability of tree inference.

Conclusions:

• Natural selection plays a critical role in the accuracy of phylogenetic tree reconstruction.
• Current tree-building algorithms are more reliable when evolutionary pressures are simulated.
• Further research is needed to improve algorithms for scenarios lacking strong selective forces.