GENOME-WIDE ASSOCIATION MAPPING AND GENOMIC PREDICTION FOR RESISTANCE TO RUSTS (PUCCINIA SPP.) IN WHEAT GERMPLASM COLLECTIONS
Muleta, Kebede Tadesse
MetadataShow full item record
This study presents the genetic characterization, genome-wide association study (GWAS) and genomic selection (GS) of resistance to stripe rust (Puccinia striiformis f. sp. tritici (Pst)) and stem rust (P. graminis f. sp. tritici (Pgt)) diseases in spring-habit hexaploid wheat germplasm collection. A total of 1,163 accessions representing major global wheat production environments were characterized for adult plant and seedling resistance to Pst in six field experiments and greenhouse tests, respectively. A genome-wide set of 5,619 informative Single Nucleotide Polymorphic (SNP) markers were used to examine population structure, linkage disequilibrium and marker-trait associations in the germplasm panel. GWAS identified 18 loci significantly associated to adult plant and seedling resistance to Pst at false discovery rate (FDR)-adjusted probability (P) < 0.10. The genetic map positions of two of the resistance loci (on chromosomes 5B and 4B) were far from previously identiﬁed Pst resistance genes/QTL, and may represent novel QTL. In addition, GWAS for resistance to Pgt and Pst was conducted using 190 Ethiopian bread wheat germplasm using a set of 24,281 genome-wide SNPs filtered from the wheat 90K iSelect genotyping assay. GWAS identified 11 and 18 genomic loci significantly (FDR P <0.1) associated to Pgt and Pst resistance, respectively. Many of the identiﬁed resistance loci were mapped close to previously identiﬁed resistance genes; however, two on chromosome 3BL and 7BL for Pgt resistance and three on chromosomes 3AL, 5AS and 7BS for Pst resistance may be new QTL. Genomic selection has the potential to enhance the utilization of germplasm collections through prediction of genomic estimated breeding values (GEBV) for as many traits as have been measured. We assessed the effect of different population genetic properties and marker density scenarios on GEBV accuracy in the context of applying GS for wheat germplasm utilization. The results of the cross-validation tests demonstrated that prediction accuracy increased with increase in population size and marker density. The results of the current GS suggests that larger germplasm collections may be more efficiently sampled based on lower-density genotyping methods, while genetic relationships between the training and validation populations remain critical when exploiting GS to select from germplasm collections.