Statistical inference of selection and divergence from a time-dependent Poisson random field model

Area of Science:

• Evolutionary biology
• Population genetics
• Bioinformatics

Background:

• Estimating species divergence time and selection pressures is crucial for understanding evolutionary processes.
• Existing models often assume population equilibrium, which may not reflect real evolutionary scenarios.
• A time-dependent Poisson random field model offers a new approach to analyze DNA sequence data.

Purpose of the Study:

• To apply a novel time-dependent Poisson random field model to estimate divergence time and selection coefficients between two related species.
• To assess the accuracy of the model using numerical simulations.
• To investigate evolutionary dynamics without assuming population equilibrium.

Main Methods:

• Utilized Markov chain Monte Carlo (MCMC) methods for parameter estimation.
• Applied a time-dependent Poisson random field model to aligned DNA sequences.
• Analyzed a dataset of 91 genes from Drosophila melanogaster and Drosophila simulans.

Main Results:

• Estimated species divergence time (t(div)) as 2.16 N(e) (1.68 million years).
• Calculated a mean selection coefficient per generation (μ(γ)) of 1.98/N(e).
• Observed a small, positive average selection coefficient, indicating weak selective pressure.

Conclusions:

• The time-dependent model provides a robust framework for estimating evolutionary parameters.
• The findings suggest limited selective effects on the analyzed genes between Drosophila species.
• The model's ability to handle non-equilibrium populations enhances its applicability in evolutionary studies.