Selection leads to false inferences of introgression using popular methods

Area of Science:

• Evolutionary Biology
• Population Genetics
• Genomics

Background:

• Detecting introgression and migration rates is crucial in evolutionary biology.
• Current methods often assume neutral evolution, which is increasingly recognized as unrealistic due to pervasive selection.
• Ignoring selection can negatively affect demographic inferences.

Purpose of the Study:

• To investigate the impact of biologically realistic selection on inferences of migration.
• To evaluate how different selection models affect commonly used population genetic methods.
• To assess the reliability of existing methods under selection.

Main Methods:

• Simulated population genetic data under various selection models: background selection, selective sweeps, balancing selection, and adaptive introgression.
• Analyzed data using methods based on the site frequency spectrum and direct sequence analysis (BPP).

Main Results:

• Ignoring selection led to false migration inferences in site frequency spectrum-based methods.
• Balancing selection and some background selection models incorrectly suggested migration (isolation-with-migration models).
• The BPP method produced false positives for migration across tested conditions, particularly at recent divergence times.

Conclusions:

• Existing methods for inferring migration may be unreliable when selection is present.
• New methods robust to the effects of selection are needed for accurate demographic inference.
• Selection's influence on migration detection highlights the need for more realistic evolutionary models.