Genome-wide association and genomic selection in Brassica napus L.
| dc.contributor.author | Sun, Jia | |
| dc.contributor.examiningcommittee | Brûlé-Babel, Anita (Plant Science) | en_US |
| dc.contributor.examiningcommittee | Zvomuya, Francis (Soil Science) | en_US |
| dc.contributor.examiningcommittee | Pauls, K. Peter (University of Guelph) | en_US |
| dc.contributor.supervisor | Duncan, Robert (Plant Science) | en_US |
| dc.date.accessioned | 2022-02-03T18:12:18Z | |
| dc.date.available | 2022-02-03T18:12:18Z | |
| dc.date.copyright | 2022-02-03 | |
| dc.date.issued | 2021 | en_US |
| dc.date.submitted | 2022-02-03T18:06:12Z | en_US |
| dc.degree.discipline | Plant Science | en_US |
| dc.degree.level | Doctor of Philosophy (Ph.D.) | en_US |
| dc.description.abstract | Linkage and association mapping are two of the most common ways to identify genes or quantitative trait loci (QTL) associated with quantitative traits. The identified genes or QTL can then be used in marker-assisted selection (MAS) with various breeding methodologies. Although MAS has gained great success in improving traits controlled by fewer genes or QTL with large effects, its application is limited in improving traits controlled by many loci with smaller effects. Genomic selection (GS), as a variant of MAS, was proposed to address this issue by utilizing markers along the whole genome instead of only focusing on the major-effect markers. Currently, the application of GS in Brassica napus breeding is at a preliminary stage. Therefore, this research was proposed to explore the potential of applying GS in B. napus breeding. Chapter 3 demonstrated the application of genome-wide association mapping (GWAS) in B. napus on important agronomic and seed quality traits. In total, 141 significant MTAs were detected. Thirty candidate genes had been previously identified in B. napus associated with abiotic stress responses and pathogen infection. Chapter 4 investigated the factors that could affect GS prediction accuracies in hybrid B. napus including training population (TP) size and composition, marker density and the choice of GS model. The prediction accuracy significantly improved by combining 91 parents and 345 hybrids in the TP, indicating the composition and size of the TP are crucial to GS performance. Higher marker density did not necessarily increase the prediction accuracy, which was possibly due to the high relatedness among the individuals in the target population. Chapter 5 explored the application of GWAS-guided GS and the prediction accuracies were compared across different traits, marker sets and GS models. Compared to conventional GS, GWAS-guided GS showed improvements in prediction accuracy, yet the improvements were not consistent across traits, models or marker sets. In addition, Bayesian models required significantly longer computational time than penalized approaches (rrBLUP and GBLUP). Taken together, the work presented here demonstrated the potential and impact of GS in assisting and optimizing hybrid B. napus breeding programs. | en_US |
| dc.description.note | February 2022 | en_US |
| dc.identifier.uri | http://hdl.handle.net/1993/36265 | |
| dc.rights | open access | en_US |
| dc.subject | Genome-wide association study | en_US |
| dc.subject | Genomic selection | en_US |
| dc.subject | Brassica napus breeding | en_US |
| dc.subject | Agronomic traits | en_US |
| dc.subject | Seed quality traits | en_US |
| dc.title | Genome-wide association and genomic selection in Brassica napus L. | en_US |
| dc.type | doctoral thesis | en_US |