Abstract

Verticillium wilt of hop (Humulus lupulus L.), caused by the soil-borne pathogen Verticillium nonalfalfae, is a devastating disease with no effective chemical control. In European hop-growing regions, breeding resistant cultivars is the most effective strategy. The lack of response differences in earlier studies suggests constitutive resistance. We therefore conducted a genome-wide association study (GWAS) using a phased hop genome assembly to improve detection of Verticillium resistance loci. A bi-parental population of 142 genotypes, derived from a cross between resistant Wye Target and susceptible BL2/1, was phenotyped for Verticillium wilt resistance and genotyped by sequencing. Association analyses with five statistical models (MLM in TASSEL 5, MLM, MLMM, FarmCPU and BLINK in GAPIT) did not identify any significant SNPs; however, several candidate loci were identified using exploratory threshold, particularly in the phase 2 genome assembly, including a wall-associated kinase (WAK) consistently detected across both genome phases and all models. GWAS results were further assessed with a Random Forest model, which identified SNPs of high feature importance and showed adequate predictive power (accuracy ≈ 0.4, correlation ≈ 0.8) for preliminary breeding screening. These findings provide an initial set of candidate markers and exploratory prediction models for Verticillium wilt resistance in hop, representing a valuable genomic resource for future marker-assisted selection and breeding strategies.

IPC Classification

Keywords