Implementing within-cross genomic prediction to reduce oat breeding costs.

Area of Science:

• Plant breeding
• Genomics
• Quantitative genetics

Background:

• Genomic prediction adoption in small breeding programs is hindered by high genotyping costs.
• Marker costs often exceed field trial expenses, posing a financial barrier.

Purpose of the Study:

• To demonstrate the utility of genomic prediction in small, narrow-base biparental oat populations across generations.
• To show how early genotyping data can optimize phenotyping by selecting siblings.
• To explore the integration of mixed marker data for enhanced prediction accuracy.

Main Methods:

• Stratifying a narrow-base biparental oat population genotyped with a modest marker set.
• Utilizing early generation genotyping data for sibling selection and reduced phenotyping.
• Integrating cheap dominant marker data (including legacy data) with higher density codominant marker data.

Main Results:

• Genomic prediction is feasible in early and later generations of stratified populations.
• Early generation genotyping and sibling selection reduce the number of lines needed for later phenotyping.
• Mixed marker data integration effectively combines different marker types for robust predictions.

Conclusions:

• Small breeding programs can implement genomic prediction using stratified populations and legacy data.
• This strategy enables early adaptation to multiple target environments.
• The approach can be scaled by incorporating higher density markers and broadening the population base.