DILS

DILS infers demographic histories and detects linked selection patterns from multilocus FASTA population genomic data using an Approximate Bayesian Computation (ABC) framework.

Key Features:

• Hierarchical analysis framework: Performs a three-level analysis comprising demographic model identification, genomic model determination, and barrier identification.
• Demographic model identification: Determines the most suitable demographic model by evaluating gene flow and population size changes that shape polymorphism and divergence.
• Genomic model determination: Assesses whether effective population size (Ne) and migration rate (N.m) are heterogeneously distributed across the genome, indicating linked selection.
• Barrier identification: Identifies loci or genomic regions that act as barriers to gene flow.
• Input formats: Accepts single-population and two-population datasets in multilocus FASTA format.
• Statistical framework: Implements Approximate Bayesian Computation (ABC) for parameter inference and model comparison.
• Performance evaluation: Performance and limitations assessed using simulations.

Scientific Applications:

• Speciation genomics: Infers demographic histories and linked selection patterns relevant to species divergence.
• Detection of genomic barriers: Identifies genomic regions associated with reduced gene flow between populations.
• Empirical population studies: Applied to a divergence continuum comprising 28 pairs of Mytilus mussel populations/species.

Methodology:

Approximate Bayesian Computation (ABC) applied to multilocus FASTA sequence data within a hierarchical three-level framework (demographic model identification, genomic model determination of heterogeneous Ne and N.m, and barrier identification); performance assessed by simulations and applied to empirical Mytilus data.