Graphtyper

Graphtyper realigns short-read sequence data to a pangenome graph—representing possible haplotypes as paths—to detect and genotype sequence variants for population-scale and complex-region genetic studies.

Key Features:

• Pangenome graph realignment: Realigns short-read sequence data to a graph-based pangenome that captures the diversity of human genome sequences.
• Variation-aware graph encoding: Represents possible haplotypes as paths and encodes sequence variations observed across populations in a variation-aware graph.
• Accurate and sensitive genotyping: Produces genotype calls with enhanced sensitivity for detecting sequence variants through the graph-based approach.
• Population-scale scalability: Demonstrated high-throughput genotyping by calling 89.4 million sequence variants from whole genomes of 28,075 individuals from Iceland.
• Computational efficiency: Achieved large-scale genotyping with reported computational usage under 100,000 CPU days.
• HLA-region genotyping: Provides detailed genotyping capabilities for complex regions including human leukocyte antigen (HLA) genes.

Scientific Applications:

• Population-scale variant discovery and genotyping: Enables detection and genotyping of sequence variants across large cohorts for population genetics and association studies.
• HLA and immunogenetics: Supports detailed genotyping of human leukocyte antigen (HLA) genes for immunology and disease association studies.
• Characterization of genetic diversity: Facilitates precise characterization of sequence variation to study genetic diversity and its implications in health and disease.
• Individual and cohort sequencing projects: Applicable to both small-scale and large population-based sequencing projects for comprehensive genetic analyses.

Methodology:

Realigns short-read sequence data to a pangenome graph that represents haplotypes as paths, leverages a variation-aware graph to encode and utilize population-observed sequence variations, and generates genotype calls.