hap-IBD

hap-IBD detects identity-by-descent (IBD) and homozygosity-by-descent (HBD) segments in phased genotype datasets for large-scale genetic analyses.

Key Features:

• Input data: Operates on phased genotype datasets.
• Compressed haplotype representation: Uses a compressed representation of haplotype data to enhance speed and efficiency of analysis.
• Positional Burrows-Wheeler Transform (PBWT): Employs PBWT to accelerate detection of shared haplotype segments by optimizing data structures for efficient querying.
• Multi-threaded execution: Supports multi-threading to parallelize computation and reduce runtime.
• Short-segment detection: Detects short IBD segments down to 2–4 centiMorgans (cM) with high accuracy.
• IBD and HBD detection: Identifies and reports both identity-by-descent (IBD) and homozygosity-by-descent (HBD) segments.
• Benchmark performance: In comparisons using UK Biobank chromosome 20 data and simulated sequence data, hap-IBD outperformed GERMLINE, iLASH, RaPID, and TRUFFLE and was the only method to rapidly and accurately detect short IBD segments in the full dataset; using 12 threads it processed 485,346 UK Biobank samples and detected 231.5 billion autosomal IBD segments ≥2 cM in 24.4 hours.

Scientific Applications:

• Population genetics: Identification of shared ancestry and population structure through detected IBD segments.
• Linkage analysis: Mapping of trait-associated regions by leveraging detected IBD segments among relatives.
• Heritability estimation: Quantifying genetic sharing to inform estimates of trait heritability.
• Genome-wide association studies (GWAS): Supporting GWAS by providing fine-scale IBD information, including short 2–4 cM segments, for association and relatedness analyses.

Methodology:

hap-IBD combines a compressed haplotype representation with the positional Burrows-Wheeler transform (PBWT) and multi-threaded processing to detect IBD and HBD segments.