HiFiMAP: High-resolution fast identity-by-descent mapping test

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Traditional genome-wide association studies (GWAS) often overlook phased haplotype data.
• Identity-by-descent (IBD) mapping utilizes shared ancestral segments but faces scalability challenges with large cohorts and short segments.
• Existing scalable IBD methods use window-based approaches, limiting association signal resolution.

Purpose of the Study:

• Introduce High-resolution Fast IBD Mapping test (HiFiMAP) for efficient and precise IBD mapping.
• Overcome computational bottlenecks in standard IBD mapping frameworks.
• Enhance the detection of rare variants and haplotype effects using short IBD segments.

Main Methods:

• Developed HiFiMAP, a novel test for IBD mapping at single nucleotide polymorphism (SNP) resolution.
• Utilized simulation studies to validate type I error rates and statistical power.
• Benchmarked HiFiMAP against existing window-based methods using UK Biobank data (N=407,681).

Main Results:

• HiFiMAP demonstrated well-controlled type I error rates and superior power for detecting rare variants.
• Achieved a 50x speed improvement over window-based methods in a UK Biobank benchmark.
• Identified five novel genetic associations for brain imaging phenotypes, including CNS regulators like NR2F1 and NSF/WNT3.

Conclusions:

• HiFiMAP provides a scalable, high-resolution solution for IBD mapping in biobank-scale datasets.
• Enables accurate pinpointing of complex trait architectures by refining association signals to the SNP level.
• Facilitates the discovery of novel genetic associations previously missed by standard GWAS.