Genomic Predicted cross performance: a tool for optimizing parental combinations in breeding programs

Area of Science:

• Plant breeding
• Quantitative genetics
• Genomics

Background:

• Genomic prediction accelerates genetic gains by estimating breeding values.
• Traditional genomic prediction focuses on additive effects.
• Predicting cross-performance offers greater value for optimizing parental combinations.

Purpose of the Study:

• Introduce the genomic predicted cross-performance (GPCP) tool.
• Evaluate GPCP's effectiveness against traditional genomic estimated breeding values (GEBVs).
• Highlight scenarios where cross-performance prediction is more advantageous.

Main Methods:

• Developed the GPCP tool using a mixed linear model incorporating additive and directional dominance.
• Integrated GPCP into the BreedBase environment and as an R package.
• Assessed GPCP using simulated traits with varying dominance and four yam traits.

Main Results:

• The GPCP tool outperformed traditional GEBVs for traits with significant dominance effects.
• GPCP effectively identified optimal parental combinations for enhanced crossing strategies.
• Demonstrated superiority in predicting cross-performance over individual GEBVs.

Conclusions:

• The GPCP tool offers a robust solution for predicting cross-performance, benefiting breeding programs targeting dominance-influenced traits.
• GPCP is particularly valuable for clonally propagated crops facing inbreeding depression and heterosis.
• The tool enhances crossing strategies, leading to improved genetic outcomes.