Utilizing genomics and historical data to optimize gene pools for new breeding programs: A case study in winter wheat

Area of Science:

• Plant breeding
• Genomics
• Quantitative genetics

Background:

• Genomic and phenotypic data are rapidly accumulating for crop genotypes.
• Emerging breeding programs can leverage these resources for genetic foundation and prediction platforms.
• Integrating genomics-enabled breeding saves costs and reallocates resources to advanced field evaluations.

Purpose of the Study:

• To understand best practices for leveraging genomic and phenotypic resources in a winter wheat breeding program.
• To determine optimal genetics for a specific target population of environments.
• To establish robust genomic prediction platforms for selecting future breeding lines.

Main Methods:

• Compiled historical phenotype data (1,285 lines) and performed GGE biplots/PCA for yield.
• Clustered locations into 22 subsets and calculated Estimated Marginal Means (EMMs) and Best Linear Unbiased Predictions (BLUPs).
• Determined representative training populations (TPs) using genetic relatedness, then generated Genomic Estimated Breeding Values (GEBVs) with mixed models (rrBLUP). Examined QTL-by-environment interactions.

Main Results:

• Achieved an average cross-validation accuracy of r = 0.42 for yield across TPs.
• Validation with elite lines showed higher accuracy (r = 0.62) when TPs included historical data.
• Identified QTL-by-environment interactions, where major QTL expression varied in benefit depending on disease pressure.

Conclusions:

• Genomics-enabled breeding accelerates cultivar development.
• Multi-institutional partnerships are crucial for leveraging diverse data and resources.
• Tailoring genomic prediction to specific target environments optimizes breeding outcomes.