Abstract
<title>Abstract</title> <p>Fusarium head blight (FHB) remains the most economically devastating disease of wheat in the Northern Great Plains. Despite three decades of intensive breeding effort, the genomic architecture underlying resistance improvement has not been characterized at the haplotype level. We analyzed 33 years of multi-environment phenotypic data from two USDA-ARS nursery systems, the Uniform Regional Nursery (URN) and the Uniform Regional Scab Nursery (URSN), integrating genome-wide haplotype information to track how resistance alleles moved through breeding populations over time. The genetic architecture of FHB resistance proved highly polygenic, with most variation residing in small-effect loci that escape detection by conventional single-marker analysis. The haplotype-based framework identified multiple validated genomic regions that together explain substantial phenotypic variation yet individually contribute modestly to resistance. This study focuses on the genomic architecture and haplotype frequency trajectories underlying three decades of resistance improvement. At the genomic level, we observed directional selection at multiple resistance loci, including blocks overlapping known major resistance genes and novel regions across multiple chromosomes. Resistance haplotype frequencies increased significantly over the study period, while expected heterozygosity at these loci remained stable, demonstrating that breeders enriched resistance alleles without local diversity loss. A drift-control analysis confirmed that allele frequency shifts at validated resistance loci exceeded neutral expectations, providing statistical evidence for selection rather than drift. This study demonstrates that haplotype-based frameworks can reveal the polygenic architecture of quantitative disease resistance and track evolutionary responses to selection in real populations.</p>