Exploiting trait correlations for next-generation grain yield and end-use quality improvement of U.S. hard winter wheat

Authors: HALEY, SCOTT - Colorado State University; MOORE, JESSICA - Cargill, Incorporated; LATSHAW, SUSAN - Bayer Cropscience; Morris, Craig; POLAND, JESSE - Kansas State University

Submitted to: Wheat Genetics International Symposium Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 1/31/2017
Publication Date: 5/24/2017
Citation: Haley, S., Moore, J., Latshaw, S., Morris, C.F., Poland, J. 2017. Exploiting trait correlations for next-generation grain yield and end-use quality improvement of U.S. hard winter wheat. Wheat Genetics International Symposium Proceedings. BOKU - University of Natural Resources and Life Sciences, Vienna, Austria, p. 440. URL: http://iwgs2017.boku.ac.at/

Interpretive Summary:

Technical Abstract: Since the early 1980s, the land area planted to U.S. hard winter wheat and the share of U.S. wheat in global export markets have both declined dramatically. Improved profitability of other crops relative to wheat, declining or static domestic wheat flour consumption, and an increasingly competitive global wheat market are important factors underlying these trends. Although hard winter wheat breeding programs focus significant attention to end-use quality, the lack of direct market incentives for superior quality and the absence of a formal cultivar registration requirement often result in wheat cultivars with inferior functional quality being commercialized and, in some cases, gaining significant market share. Additionally, identity-preserved programs developed to segregate and maintain improved functional value have proven both difficult to establish and costly to maintain in a dynamic economic setting. In order to reverse the trends in declining plantings and market share, modern approaches that foster increased rates of genetic gain for production-related traits simultaneously with improved functional value from an end-use quality perspective must be implemented. Single-trait and multi-trait genomic selection (GS) approaches offer tremendous potential to accomplish these objectives. Phenotypic data collected within the context of a comprehensive public cultivar development program were utilized to develop GS-based prediction models for both production-related and quality-related traits. These data included spatially adjusted data from multi-year and multi-location yield trials and comprehensive experimental milling and bread-baking quality data. Genome-wide single nucleotide polymorphism (SNP) markers obtained using genotyping-by-sequencing (GBS) were used in ridge regression best-linear unbiased prediction (rrBLUP) approaches to develop GS-based predictions. Positive or negative trait associations between production-related traits (i.e., grain yield, plant height, grain protein deviation), between production- and quality-related traits (i.e., grain yield, grain protein concentration, water absorption), and between quality-related traits (i.e., water absorption, grain protein concentration, flour yield) were exploited to improve prediction accuracy of key traits through multi-trait genomic selection. Validation of predictions included both traditional cross-validation and forward prediction of unphenotyped individuals. Through focused integration of these approaches in a long-term cultivar development program we hope to be able to provide improved wheat cultivars that enhance both production- and quality-related traits, optimize economic returns for both producers and product manufacturers, and reverse the decline in plantings and market share of U.S. hard winter wheat.