Inference of directional selection and mutation parameters assuming equilibrium.

Area of Science:

• Population genetics
• Evolutionary biology
• Genomics

Background:

• Wright's classical model describes allele frequency evolution under mutation, selection, and drift.
• The equilibrium distribution is key for inferring evolutionary parameters from site frequency spectra (SFS).
• A boundary-mutation model simplifies inference by partitioning parameter estimation.

Purpose of the Study:

• To derive maximum likelihood estimators for mutation and selection parameters within an equilibrium framework.
• To apply these estimators to both simulated and empirical population genetic data.
• To refine the analysis of allele frequency spectra for evolutionary inference.

Main Methods:

• Approximation of Wright's model using a boundary-mutation model for low mutation rates.
• Derivation of maximum likelihood estimators for mutation and selection parameters.
• Application to simulated site frequency spectra (SFS) data and empirical Drosophila simulans data.

Main Results:

• The boundary-mutation model allows for partitioned inference of selection and mutation parameters.
• Selection parameters are inferred from the shape of the SFS within polymorphic regions.
• Mutation parameters are inferred from the counts of polymorphic and monomorphic alleles.

Conclusions:

• The derived maximum likelihood estimators provide a robust method for inferring evolutionary parameters.
• The boundary-mutation model offers an efficient approach for analyzing population genomic data.
• This method was successfully applied to both simulated and real-world genetic datasets.