Files and links (1)
CC BY V4.0, Open Access
Dissertation
INTO THE UNKNOWN: PREDICTING WHEAT DISEASE RESISTANCE AND SEEDLING EMERGENCE
Washington State University
Doctor of Philosophy (PhD), Washington State University
01/2022
DOI:
https://doi.org/10.7273/000004406
Handle:
https://hdl.handle.net/2376/124891
Abstract
Winter wheat (Triticum aestivum L.) is the most widely grown cereal grain in Washington state. However, it generally takes 12 years to develop a cultivar from crossing to release, which can limit the ability for rapid improvement of traits. Genomic selection (GS) is posed to increase genetic gains by reducing the generation interval, also known as cycle time, while increasing response to selection. To fully utilize GS in breeding programs we reviewed and proposed implementing a two-part breeding strategy by differentiating population improvement and product development to increase the genetic gain of a breeding program. We then compared GS models for various scenarios for two complex traits, seedling emergence and stripe rust (Puccinia striiformis Westend. f. sp. tritici Erikss.) resistance, using two training populations composed of a diverse association mapping panel and breeding lines that can be used as a guideline for the implementation of GS into breeding programs. We compared GS models for seedling emergence and showed the consistent moderate accuracy of the parametric models indicates little advantage of using non-parametric models within individual years, but the non-parametric models show a slight increase in accuracy when combing years for complex traits. We then compared single-trait and multi-trait genome-wide association (GWAS) models with covariates, and were able to identify many small effect markers, while identifying the large effect markers on chromosome 5A. Stripe rust resistance is controlled by both major and minor resistance genes, and it is recommended to combine both major and minor genes for durable resistance. We compared the accuracy of GS models with major gene and GWAS markers as fixed effects and compared them to marker-assisted selection. The major gene and GWAS markers had only a small to zero increase in prediction accuracy over the base GS models but showed a statistical increase in accuracy using major markers when the mean accuracy decreased. Further, stripe rust resistance phenotypes are commonly skewed due to high levels of resistance. We compared classification and regression models and showed that breeders can use linear and non-parametric regression models within their own breeding lines over combined years to accurately predict skewed phenotypes.
Metrics
6 File views/ downloads
26 Record Views
Details
- Title
- INTO THE UNKNOWN: PREDICTING WHEAT DISEASE RESISTANCE AND SEEDLING EMERGENCE
- Creators
- Lance Farley Merrick
- Contributors
- Arron H. Carter (Advisor)Kimberly Garland Campbell (Committee Member)Zhiwu Zhang (Committee Member)Xianming Chen (Committee Member)
- Awarding Institution
- Washington State University
- Academic Unit
- Crop and Soil Sciences, Department of
- Theses and Dissertations
- Doctor of Philosophy (PhD), Washington State University
- Publisher
- Washington State University
- Number of pages
- 386
- Identifiers
- OCLC#: 1365112500; 99900883237901842
- Language
- English
- Resource Type
- Dissertation