From conservation to utilization : data-driven and genome-wide approaches to unlock barley genebank resources for disease resistance breeding

Abstract

To capitalize on the potential of plat genetic resources (PGRs), we focused on a barley core collection from the German ex situ genebank and contrasted it with an elite collection, with particular emphasis on four disease traits. Genome-wide association study (GWAS) was conducted, identified 932 marker–trait associations. Three PGR donors with high counts of novel alleles exhibited exceptional resistance, offering valuable resources for targeted pre-breeding in future varieties. To further extend the use of genebank diversity, we implemented genome-wide prediction to enable large-scale phenotypic imputation. Using the core collection, we trained genomic prediction models and imputed resistance scores for 20,458 PGRs. Empirical validation revealed strong correlations between predicted and observed resistance levels. Subsequent GWAS identified five novel marker–trait associations, demonstrating that prediction-informed sampling can effectively expand trait-relevant genetic diversity.