The potential of genomic‐assisted breeding to improve Fusarium head blight resistance in winter durum wheat

Abstract

AbstractDurum wheat is the most important tetraploid wheat mainly used for semolina and pasta production, but is notorious for its high susceptibility to Fusarium head blight (FHB). Our objectives were to identify and characterize quantitative trait loci (QTL) in winter durum and to evaluate the potential of genomic approaches for the improvement ofFHBresistance. Here, we employed an international panel of 170 winter and 14 spring durum lines, phenotyped forFusarium culmorumresistance at five environments. Heading date, plant height and meanFHBseverity showed significant genotypic variation with high heritabilities andFHBresistance was negatively correlated with both heading date and plant height. The dwarfing geneRht‐B1significantly affectedFHBresistance and the genome‐wide association scan identified eight additionalQTLaffectingFHBresistance, explaining between 1% and 14% of the genotypic variation. A genome‐wide prediction approach yielded only a slightly improved predictive ability compared to marker‐assisted selection based on the four strongestQTL. In conclusion,FHBresistance in durum wheat is a highly quantitative trait and in breeding programmes may best be tackled by classical high‐throughput recurrent phenotypic selection that can be assisted by genomic prediction if marker profiles are available.