Genetic mapping of QTL for FHB resistance and whole genome association mapping in barley

Fusarium head blight (FHB) is a difficult to control disease of barley (Hordeum vulgare L.) due to the absence of good genetic resistance and the association of resistance with undesirable agronomic traits. Ninety four random RIL developed from the cross Shenmai 3 x Rawson were evaluated in four environments in North Dakota (ND) and three environments in China. Fusarium head blight, DON accumulation (DON), heading date (HD), and plant height (HT) were measured in Fusarium nurseries and, yield (YLD) and yield related traits were measured in three environments in ND. Mean phenotypic values were used to map QTL for all traits in individual environments. Major QTL for HD were detected on chromosomes 1H, 5H, and 7H whereas chromosomes 4H and 6H contained major QTL regulating HT. One QTL that was not associated with either HD or HT was detected on 7HS for FHB resistance. In a separate study using molecular and phenotypic data generated in the USDA-CSREES Barley Coordinated Project (CAP), whole genome association mapping was validated using HD. In this study, results using the Q-matrix created by STRUCTURE and principal component analysis were compared. Overrepresentation of specific breeding lines in association mapping populations can increase false positive marker-traits associations because of reduced genetic diversity. The simple 'diversity maximization' procedure was included in this study to overcome the effect of overrepresented breeding lines. Several regions containing QTL for HD were identified using a subset population created using the 'diversity maximization' procedure. The GLM and MLM options of TASSEL detected 25 genomic regions that contained significant marker-trait association for HD. Twelve of these regions contain known function genes for HD. Among all the marker-traits associations detected, one on chromosome 2H and one on chromosome 5H each explained 25% of the variation in HD. Each of these regions is known to have two known function genes controlling HD. The magnitude of phenotypic variation found for these two marker-trait associations is the largest detected by association mapping ever in barley. This indicates that the approach used here can be used for other traits to exploit the suggested potential of whole genome association mapping.