Genome scans for selection and introgression based on k-nearest neighbour techniques

Area of Science:

• Population Genetics
• Genomics
• Evolutionary Biology

Background:

• Genome-scan methods are crucial for identifying local selection and introgression.
• Existing methods often focus on either selection or introgression, limiting combined analyses.
• A unified approach is needed to study both phenomena concurrently.

Purpose of the Study:

• To introduce versatile genome-scan methods capable of detecting both selection and introgression.
• To provide a robust framework for analyzing combined evolutionary forces in genomic data.
• To offer a novel approach for population genomics research.

Main Methods:

• Development of nonparametric k-nearest neighbour (kNN) based genome-scan methods.
• Incorporation of pairwise Fixation Index (FST) and pairwise nucleotide differences (dxy) as key features.
• Benchmarking through extensive simulations with varied genetic and demographic parameters.

Main Results:

• kNN-based methods demonstrate high performance in detecting both selection and introgression.
• The methods show remarkable efficacy across diverse simulation scenarios, including varying recombination rates and gene flow.
• Performance is competitive with and often surpasses state-of-the-art methods.

Conclusions:

• The proposed kNN-based genome-scan methods offer a versatile and powerful tool for population genomics.
• These methods provide a significant advancement for studying the interplay of selection and introgression.
• The approach is readily applicable to real-world genomic data using the R-package popgenome.