Evolution of Quantitative Traits under a Migration-Selection Balance: When Does Skew Matter?

Area of Science:

• Evolutionary genetics
• Population genetics
• Quantitative genetics

Background:

• Quantitative-genetic models often assume normal distributions for genotypic and phenotypic values.
• Subdivided populations with differing selection pressures can experience skewed local distributions due to migration.
• Existing models may lack accuracy when distributions deviate from Gaussianity.

Purpose of the Study:

• To develop more accurate quantitative-genetic models for population differentiation under migration-selection balance.
• To investigate the impact of non-Gaussian distributions, specifically skew, on evolutionary predictions.
• To assess the importance of higher moments in modeling selection in heterogeneous environments.

Main Methods:

• Developed a simplified two-habitat model.
• Derived formulas for migration-selection balance without assuming Gaussian distributions.
• Incorporated higher moments, such as skew, into the models.
• Utilized simulations to validate theoretical predictions.

Main Results:

• Formulas incorporating higher moments provide more accurate predictions of divergence than Gaussian models.
• Skew in local distributions is a natural outcome of migration-selection balance in heterogeneous environments.
• Simulations confirmed that traits with large-effect loci exhibit the most pronounced skew.

Conclusions:

• Higher moments, particularly skew, are crucial for accurately modeling population differentiation under migration-selection balance.
• The assumption of Gaussian distributions can lead to inaccurate predictions in heterogeneous environments.
• Understanding distribution shapes is vital for predicting evolutionary responses to selection.