polyHap

polyHap infers haplotypes in polyploid organisms and genomic regions with copy number variations (CNVs), enabling accurate phasing and supporting haplotype-based analyses.

Key Features:

• HMM with sampling algorithm: Uses a hidden Markov model (HMM) combined with a sampling algorithm to model phase and provide probabilistic inference.
• Joint haplotype inference: Performs joint inference of haplotypes across multiple individuals.
• Uncertainty quantification: Provides a quantifiable measure of uncertainty for phased haplotypes.
• Variable ploidy and CNV support: Handles varying ploidy levels across individuals while assuming constant ploidy within analyzed genomic regions and phases regions with copy number variations (CNVs).
• Supported ploidy levels: Manages diploid, triploid, and tetraploid genotypes.
• Reduced phasing and imputation errors: Demonstrates reduction of switch errors and imputation errors for missing genotypes in evaluations.
• Simulation-based validation: Validated using simulations that generate artificial polyploid genotypes from real haplotype data.
• Comparative performance: Shown to outperform fastPhase for diploids and SATlotyper for tetraploids in densely genotyped regions.

Scientific Applications:

• Haplotype-based association studies: Enables haplotype-based association analyses in polyploid organisms and CNV regions.
• Selection scans: Facilitates identification of genomic regions under selection using phased haplotypes.
• Ancestral inference: Supports ancestral inference through reconstructed haplotypes.
• Dense SNP dataset analysis in polyploids: Allows analysis of large SNP datasets from polyploid organisms and CNV regions.

Methodology:

Uses a hidden Markov model (HMM) combined with a sampling algorithm to jointly infer haplotypes across multiple individuals and provide measures of uncertainty.