SelEstim

SelEstim detects signatures of natural selection and quantifies selection intensity from allele frequency and SNP genotype data in subdivided populations (demes).

Key Features:

• Model-Based Approach: Employs a model-based method to distinguish selected polymorphisms from neutral or nearly neutral variants.
• Diffusion Approximation Model: Uses a diffusion approximation to describe allele frequency distributions in subdivided populations exchanging migrants.
• Hierarchical Bayesian Framework: Implements a hierarchical Bayesian model to represent population- and locus-specific parameters.
• MCMC Parameter Estimation: Applies Markov chain Monte Carlo (MCMC) algorithms to sample from the joint posterior distribution of model parameters.
• Simulation Validation and Empirical Analysis: Validated by stochastic simulations and applied to SNP data such as the Stanford HGDP-CEPH panel, detecting signals including positive selection upstream of the LCT gene associated with lactase persistence.

Scientific Applications:

• Evolutionary and Population Genetics: Identifies loci under selection and quantifies selection intensity to inform studies of adaptive evolution.
• Mapping Adaptive Traits: Assesses selection at SNPs to investigate the genetic basis of phenotypic traits such as lactase persistence.
• Structured Population Analysis: Analyzes allele frequency dynamics in subdivided populations and comparisons across human populations and other species.

Methodology:

Inputs are large-scale genotype data from high-throughput sequencing or genotyping technologies; a diffusion approximation model is applied within a hierarchical Bayesian framework; MCMC is used to estimate selection and allele frequency parameters; validation is performed with stochastic simulations and analysis of real-world SNP data (e.g., HGDP-CEPH).