Most recent common ancestor probability distributions in gene genealogies under selection

Area of Science:

• Population Genetics
• Computational Biology
• Evolutionary Dynamics

Background:

• Understanding the genetic makeup of populations over time is crucial for evolutionary studies.
• Existing models of genetic ancestry, like Wright's formula, may not fully capture dynamics under selection.
• Computational methods are increasingly important for analyzing complex population genetic data.

Purpose of the Study:

• To computationally investigate the conditional probability distribution of the most recent common ancestor's allelic type in selected populations.
• To compare these conditional distributions with unconditional cases and quantify their differences.
• To enhance the ancestral selection graph model for improved computational efficiency in nonneutral evolutionary studies.

Main Methods:

• Utilized computational methods to analyze conditional probability distributions of allelic types in gene genealogies.
• Developed and enhanced the ancestral selection graph model, simplifying its structure and reducing branching rates.
• Applied the model to biallelic haploid and diploid population models.

Main Results:

• Quantified differences between conditional and unconditional distributions of the most recent common ancestor's allelic type.
• Demonstrated that unconditional distributions deviate from the stationary distribution described by Wright's formula.
• Showcased an improved ancestral selection graph structure for nonneutral coalescent simulations.

Conclusions:

• The study provides a refined computational framework for analyzing genetic ancestry in selected populations.
• The enhanced ancestral selection graph offers greater efficiency for likelihood-inference techniques in population genetics.
• Findings contribute to a deeper understanding of evolutionary processes under selection and offer practical tools for research.