Accessible, realistic genome simulation with selection using stdpopsim.

Area of Science:

• Population genetics
• Evolutionary biology
• Computational biology

Background:

• Evolutionary forces like selection shape genetic variation.
• Simulating realistic selection across complex genomes is computationally challenging.
• Existing tools like stdpopsim offer species-specific population genetic models.

Purpose of the Study:

• To extend the stdpopsim library to simulate various modes of selection.
• To enable simulations on annotated genomic subsets with specific fitness effects.
• To provide a reproducible framework for benchmarking population genetics methods.

Main Methods:

• Incorporated background selection, selective sweeps, and arbitrary distributions of fitness effects (DFE).
• Enabled simulations on species-specific genomic annotations and published DFE estimates.
• Benchmarked demographic inference, DFE estimation, and selective sweep detection methods.

Main Results:

• Demographic inference methods show robustness to selection on linked sites.
• DFE-inference methods are sensitive to underlying model assumptions.
• Genomic features like recombination rate and sequence density impact selective sweep detection power.

Conclusions:

• The extended stdpopsim framework offers a powerful resource for population genetics research.
• Facilitates reproducible exploration of selection's interplay with other evolutionary forces.
• Lowers the barrier for studying complex evolutionary scenarios.