CHAP-GWAS: Leveraging Chromosomal Haplotypes to Improve Genome-Wide Association Studies

<title>Abstract</title> <p> Conventional genome-wide association studies (GWAS), which analyze individual single nucleotide polymorphisms (SNPs), have been instrumental in dissecting the genetic basis of quantitative traits. While short-range haplotype variants have recently demonstrated superior statistical power by capturing local linkage disequilibrium, they are often constrained by predefined, fixed-length blocks. Advances in long-read sequencing and computational phasing are making chromosome-scale haplotypes increasingly accessible, opening new opportunities for association studies. However, frameworks that effectively exploit this long-range genetic information remain limited. Here, we introduce CHAP-GWAS, a novel, dynamically adaptive algorithm designed to identify trait-associated haplotype variants from chromosome-scale phased data. Unlike existing methods, CHAP-GWAS utilizes a phenotype-driven approach that dynamically defines and extends haplotype blocks from "seed" loci, allowing for the precise capture of complex genetic effects. Using comprehensive simulations and real-world datasets from three diverse plant species ( <italic>Arabidopsis</italic> , rice, and maize), we demonstrate that CHAP-GWAS successfully identifies quantitative trait loci overlooked by conventional single-SNP and static haplotype-based approaches. By leveraging chromosome-length haplotypes, our framework significantly enhances statistical power and addresses a critical component of missing heritability. CHAP-GWAS offers a robust, scalable, and synergistic approach to genetic mapping, with the potential to advance precision agriculture and medicine through more comprehensive and accurate genetic discovery. </p>