Genomic dissection reveals QTLs for grain biomass and correlated traits under drought stress in Ethiopian durum wheat (Triticum turgidum ssp. durum)

Drought stress seriously challenges wheat production and productivity. Grain biomass (GB) and related traits were assessed under drought stress and non-stress for 285 and 185 durum wheat genotypes, respectively, in field and climate chamber experiments to identify quantitative trait loci (QTL). Significant correlations between traits estimated in field and climate chamber trials were observed (p < .001). Genotyping with the wheat 90 K iSelect single nucleotide polymorphism (SNP) array revealed 11,919 polymorphic SNP markers distributed across the durum wheat genome. The FarmCPU (Fixed and random model Circulating Probability Unification) method was used for genome-wide association studies (GWAS). A total of 191 significant (log10p ≥ 4) marker-trait associations (MTAs) were detected at a linkage disequilibrium (LD, r2 ≥ .2) at 4.78 Mb and were clustered into 70 QTLs. A total of 69 (36%) of the MTAs passed a false discovery rate (FDR) of 5%. The numbers of QTLs detected were 21, 31, 9 and 9 under field drought stress (FDS), field non-stress (FNS), climate chamber drought stress (CCDS) and climate chamber non-stress (CCNS) conditions, respectively. About 43% and 57% of the QTLs were located on the A and B genomes, respectively. Some of the detected QTLs were in agreement with previously reported QTLs, while others are novel ones for the traits investigated. QTLs on 1A between 495694477and -501944537bp, on 3B between 416256124 and 430507900 bp, on 3B between 745357158 and 759608934 bp, on 4B between 593416763 and 605142497 bp and on 4B between 658785890 and 670511624 bp were selected for validation and may be used to increase grain yield under drought stress in marker-assisted selection (MAS) schemes.