Genomic prediction in CIMMYT maize and wheat breeding programs

Area of Science:

• Plant breeding
• Quantitative genetics
• Agricultural science

Background:

• Genomic selection (GS) is a powerful tool for accelerating genetic gains in plants and animals.
• Prediction accuracy in plant genomic selection varies based on heritability, marker density, sample size, and genotype × environment interactions (GE).

Purpose of the Study:

• To evaluate genomic prediction results within the International Maize and Wheat Improvement Center's (CIMMYT's) maize and wheat breeding programs.
• To assess the predictive ability of different models using pedigree and marker data for practical implementation of GS.

Main Methods:

• Assessed predictive ability of models using pedigree and marker information.
• Investigated the impact of population structure and relatedness on prediction accuracy.
• Incorporated genotype × environment (GE) interactions into genomic prediction models.

Main Results:

• Pedigree information significantly contributes to prediction accuracy in global populations.
• Prediction accuracy diminishes when training and prediction populations are unrelated.
• Modeling GE interactions enhances prediction accuracy by leveraging information from correlated environments.

Conclusions:

• Genomic selection shows promise for accelerating genetic gains in maize and wheat breeding.
• Further research is needed to optimize GS implementation, particularly for prediction within and between bi-parental crosses.
• Quantifying breeding value components is essential for advancing GS in plant breeding populations.