Optimal cross selection for long-term genetic gain in two-part programs with rapid recurrent genomic selection

Area of Science:

• Plant breeding
• Genetics
• Agricultural science

Background:

• Two-part breeding programs integrate population improvement with recurrent genomic selection and product development.
• Rapid recurrent genomic selection offers potential but faces challenges like genotyping costs and genetic drift.

Purpose of the Study:

• To evaluate optimal cross selection for balancing genetic gain and diversity maintenance in rapid recurrent genomic selection programs.
• To compare optimal cross selection against truncation selection in a simulated wheat breeding program.

Main Methods:

• Simulated a 20-year wheat breeding program comparing optimal cross selection with truncation selection.
• Implemented optimal cross selection using AlphaMate to jointly optimize selection, diversity, and cross allocation.
• Varied the number of cycles per year (1-6) for recurrent genomic selection.

Main Results:

• Optimal cross selection significantly increased long-term genetic gain, especially with more cycles per year.
• With four cycles/year, optimal cross selection yielded 78% (15%) more gain than truncation selection with small (large) parent numbers.
• Optimal cross selection quadrupled (doubled) the efficiency of converting genetic diversity into gain compared to truncation selection.

Conclusions:

• Optimal cross selection is superior for maximizing genetic gain in two-part programs with rapid recurrent genomic selection.
• This approach effectively manages and leverages germplasm for improved breeding outcomes.
• It enhances the efficiency of genetic diversity utilization and maintains prediction accuracy.