Association-based trait mapping is an innovative methodology based on linkage disequilibrium. Studies in plants, especially in cereals, are rare. A genome-wide association study of wheat is reported, in which a large number of diversity array technology markers was used to genotype a winter wheat core collection of 96 accessions. The germplasm was structured into two sub-populations. Twenty agronomic traits were measured in field trials conducted over up to eight growing seasons. Association analysis was performed with two different approaches, the general linear model incorporating the Q-matrix only and the mixed linear model including also the kinship-matrix. In total, 385 marker-trait associations significant in both models were detected. The intrachromosomal location of many of these coincided with those of known major genes or quantitative trait loci, but others were detected in regions where no known genes have been located to date. These latter presumptive loci provide opportunities for further wheat improvement, based on a marker approach.

Genetic analyses and association mapping were performed on a winter wheat core collection of 96 accessions sampled from a variety of geographic origins. Twenty-four agronomic traits were evaluated over 3 years under fully irrigated, rainfed and drought treatments. Grain yield was the most sensitive trait to water deficit and was highly correlated with above-ground biomass per plant and number of kernels per m2. The germplasm was structured into four subpopulations. The association of 46 SSR loci distributed throughout the wheat genome with yield and agronomic traits was analyzed using a general linear model, where subpopulation information was used to control false-positive or spurious marker-trait associations (MTAs). A total of 26, 21 and 29 significant (P < 0.001) MTAs were identified in irrigated, rainfed and drought treatments, respectively. The marker effects ranged from 14.0 to 50.8%. Combined across all treatments, 34 significant (P < 0.001) MTAs were identified with nine markers, and R2 ranged from 14.5 to 50.2%. Marker psp3200 (6DS) and particularly gwm484 (2DS) were associated with many significant MTAs in each treatment and explained the greatest proportion of phenotypic variation. Although we were not able to recognize any marker related to grain yield under drought stress, a number of MTAs associated with developmental and agronomic traits highly correlated with grain yield under drought were identified.

No abstract

Genebanks are entrusted with the storage, preservation and distribution of crop germplasm. Seed longevity is an important character in this context, but little is known regarding its genetic basis, largely because it is so strongly influenced by nongenetic factors. Here we present the outcome of a genetic dissection of seed longevity in bread wheat. We applied both a standard quantitative trait locus analysis based on segregation from a biparental cross, and an association analysis using a germplasm panel to detect marker trait associations. The latter revealed more loci than the former. Some of the genomic regions identified are known to contain genes determining spike architecture or aspects of biotic and abiotic stress responses. The results open perspectives for identification of favourable longevity alleles and the more accurate prediction of seed longevity in cereal germplasm collections.

Roots are vital plant organs that determine adaptation to various soil conditions. The present study evaluated a core winter wheat collection for rooting depth under PEG induced early stage water stress and non-stress growing conditions. Analysis of phenotypic data indicated highly significant (p < 0.01) variation among genotypes. Broad sense heritability of 59 and 73% with corresponding genetic gains of 7.6 and 9.7 (5% selection intensity) were found under non-stress and stress conditions, respectively. The test genotypes were grouped in to three distinct clusters using unweighted pair group method with arithmetic mean (UPGMA) clustering based on maximum Euclidian distance. The first three principal components gave optimum mixed linear model for genome wide association study (GWAS). Linkage disequilibrium (LD) analysis showed significant LD (p < 0.05) amongst 15% of total marker pairs (25,125). Nearly 16% of the significant LDs were among inter chromosomal marker pairs. GWAS revealed five significant root length QTLs spread across four chromosomes. None of the identified QTLs were common between the two growing conditions. Stress specific QTLs, combined explaining 31% of phenotypic variation were located on chromosomes 2B (wPt6278) and 3B (wPt1159). Similarly, two of the three QTLs (wPt0021 and wPt8890) identified under the non-stress condition were found on chromosomes 3B and 5B, respectively. The B genome showed significant importance in controlling root growth both under stress and non-stress conditions. The identified markers can potentially be validated and used for marker assisted selection.