A population genetic hidden Markov model for detecting genomic regions under selection

Area of Science:

• Genomics
• Population Genetics
• Computational Biology

Background:

• Hidden Markov models (HMMs) are increasingly used in genomics.
• Identifying genomic regions under recent selection is crucial for understanding evolution.
• Population genomics studies generate large datasets of genetic variation.

Purpose of the Study:

• To introduce an explicit population genetics hidden Markov model (popGenHMM).
• To utilize single nucleotide polymorphism (SNP) frequency data for detecting recent selection.
• To apply the model to real-world population genomics data.

Main Methods:

• Developed a novel hidden Markov model (popGenHMM) incorporating population genetics principles.
• Assumed independent SNP frequency emissions with partial correlation by selective state.
• Trained and applied the popGenHMM to Drosophila melanogaster resequencing data.

Main Results:

• The popGenHMM successfully identified genomic regions potentially under recent selection.
• The model's predictions were based on the site frequency spectrum (SFS).
• Demonstrated the utility of popGenHMM with early Drosophila Population Genomics Project data.

Conclusions:

• The popGenHMM is a valuable tool for detecting recent selection in population genomics.
• The model effectively leverages SNP frequency data for evolutionary inference.
• Further application of popGenHMM can enhance our understanding of adaptive evolution.