High-resolution population-specific recombination rates and their effect on phasing and genotype imputation

Area of Science:

• Population Genomics
• Human Genetics
• Bioinformatics

Background:

• Population-specific reference panels significantly influence population genomic analyses.
• This is particularly true for population isolates, where genetic differences can be more pronounced.

Purpose of the Study:

• To develop high-resolution recombination rate maps for a Finnish population.
• To compare Finnish recombination rates with those of non-Finnish Europeans (NFE).
• To assess the downstream effects of population-specific recombination rates on haplotype phasing and genotype imputation.

Main Methods:

• Whole-genome sequencing of 55 Finnish family trios at high coverage (20-30×).
• High-resolution recombination rate mapping at 10 and 50 kb scales.
• Statistical phasing and genotype imputation using Finnish and NFE recombination maps.

Main Results:

• Finnish recombination rates are 12-14% lower than NFE rates, with moderate correlation (Spearman's ρ = 0.67-0.79).
• Finnish recombination maps showed no significant impact on haplotype phasing accuracy (switch error rates ~2%).
• Imputation concordance rates remained high across variant frequencies (common: 97-98%, low frequency: 92-96%, rare: 78-90%).

Conclusions:

• Haplotype phasing and genotype imputation primarily depend on population-specific reference panels and sample size.
• Population-specific recombination maps have minimal impact on these analyses, even in isolated populations.
• Existing HapMap recombination maps are robust for population-specific phasing and imputation pipelines.