P-smoother

P-smoother applies a positional Burrows-Wheeler transform (PBWT)-based smoothing algorithm to correct recent mutations and genotyping errors in phased VCF haplotype panels, improving haplotype matching and identical-by-descent (IBD) detection in biobank-scale datasets.

Key Features:

• Error Correction: Employs a PBWT-based smoothing algorithm on phased genotype data (VCF) with bidirectional panel scanning that flips mismatching alleles using an IBD prior.
• Improved Haplotype Matching: Reduces mismatch-induced breakpoints so PBWT exact-match haplotype matching becomes more error-tolerant and accurate.
• Enhanced IBD Segment Identification: Running PBWT over the smoothed panel identifies more pairwise IBD segments than unsmoothed panels and enables PS-cluster for multiway IBD detection.
• Efficiency and Scalability: In simulations with 4,000 haplotypes and a 0.2% error rate, corrects approximately 85% of errors and is demonstrated to be computationally efficient for biobank-scale datasets such as the UK Biobank.

Scientific Applications:

• Genetic Linkage Studies: Improves the reliability of linkage analyses by increasing the accuracy of haplotype matching.
• Population Genetics: Supports detection of shared ancestry and evolutionary patterns via improved IBD segment detection in large haplotype panels.
• Personalized Medicine: Provides more accurate haplotype data for genomic studies of disease predisposition and variant interpretation.

Methodology:

Operates on phased genotype data using bidirectional PBWT-based panel scanning that flips mismatching alleles based on an IBD prior to produce a smoothed haplotype panel.