Haploclusters

Haploclusters identifies excess haplotype sharing between case and control haplotypes to detect candidate disease susceptibility loci in genome-wide association studies.

Key Features:

• Detection of Excess Haplotype Sharing: Uses a chi-square based statistic to identify regions with excess shared ancestral haplotypes indicative of disease loci.
• Model-Free Approach: Operates without assumptions about population history or background linkage disequilibrium (LD) patterns.
• Efficient Genome-Wide Scanning: Optimized for genome-wide scans to serve as an initial-pass analysis that identifies candidate regions and inferred susceptibility haplotypes prior to likelihood-based methods.
• Statistical Rigor: Assesses significance using permutation testing, yielding low false-positive rates and greater power than single-marker tests in low to moderate LD scenarios.
• Case-Control LD Comparison: Targets regions where LD persists longer among cases than controls (background LD) to localize susceptibility signals.

Scientific Applications:

• Genome-wide association studies (GWAS): Facilitates initial identification of disease susceptibility loci for qualitative and complex traits using high-throughput genotyping technologies.
• Disease mapping: Applied to mapping loci in diseases such as cystic fibrosis and multiple sclerosis.
• Initial-pass candidate identification: Provides candidate regions and inferred susceptibility haplotypes for follow-up with complex likelihood-based fine-mapping.

Methodology:

Chi-square based statistic for quantifying excess haplotype sharing; permutation testing to assess significance; model-free, genome-wide scanning applied to simulated and real haplotype datasets.