Multi-trait Improvement by Predicting Genetic Correlations in Breeding Crosses.

Area of Science:

• Plant breeding
• Quantitative genetics
• Genomics

Background:

• Quantitative traits in plants are often genetically correlated, hindering selection progress.
• Identifying parent combinations with favorable genetic correlations is crucial for breeding multiple traits.
• Genomic prediction of genetic correlations (r) in crosses is a potential but untested approach.

Purpose of the Study:

• To assess the accuracy of genomic predictions for genetic correlations (r) and their impact on selection response using simulations.
• To empirically evaluate the prediction accuracy of genetic correlations in a barley breeding program.

Main Methods:

• Simulations were used to model genomewide marker segregation and predict genetic correlations (r).
• The accuracy of predicting r and long-term selection response was evaluated.
• Empirical data from 26 barley (Hordeum vulgare L.) breeding populations were analyzed.

Main Results:

• Genomic prediction accuracy for r was moderate in simulations, influenced by heritability, population size, and genetic architecture.
• Empirical prediction accuracy for r in barley ranged from low (-0.012) to moderate (0.42).
• Selecting crosses based on predicted r increased multi-trait genetic gain by 11-27% compared to selection on predicted cross means.

Conclusions:

• Prioritizing crosses based on predicted genetic correlations is a feasible strategy.
• This approach can effectively improve unfavorably correlated traits in breeding programs.
• Genomic prediction of genetic correlations offers a valuable tool for enhancing breeding efficiency.