PHASEBOOK

PHASEBOOK reconstructs haplotypes from genotype data by combining familial Mendelian segregation and population linkage information to improve phasing and genotype imputation.

Key Features:

• Robustness to Errors: Phasebook is robust to genotyping errors and map inaccuracies and performs well on large half-sib families common in plant and animal genetics.
• Dual Information Utilization: Combines population information (linkage disequilibrium) and familial data (Mendelian segregation and linkage) for phasing.
• Hidden Markov Model Integration: Employs Hidden Markov Models derived from fastPHASE or Beagle to assign reconstructed haplotypes to hidden states corresponding to clusters of genealogically related chromosomes and to impute missing genotypes.
• Application in Genetic Analysis: Supports mapping trait loci, predicting genomic breeding values, identifying signatures of selection, and fine-mapping quantitative trait loci (QTL).
• Computational Efficiency: Handles large datasets from high-density SNP panels suitable for extensive genetic studies.

Scientific Applications:

• Mapping trait loci: Fine-mapping and mapping of trait loci using reconstructed haplotypes and cluster states.
• Predicting genomic breeding values: Estimation of genomic breeding values in animal and plant genetics using phased genotypes.
• Identifying selection signatures: Detection of signatures of selection via haplotype structure.
• Fine-mapping QTLs: Fine-mapping quantitative trait loci (QTL) using haplotype-derived cluster states.

Methodology:

Integrates LinkPHASE, HiddenPHASE, DualPHASE, and DAGPHASE in a two-step approach: first reconstructs haplotypes using familial information based on Mendelian segregation and linkage; then employs Hidden Markov Models (HMMs) from fastPHASE or Beagle to fill gaps and assign haplotypes to hidden cluster states.