The apple REFPOP-a reference population for genomics-assisted breeding in apple

Area of Science:

• Plant genetics and breeding
• Genomics
• Horticulture

Background:

• Apple breeding is time-consuming and expensive, requiring extensive screening of candidates.
• Genomics-assisted breeding (GAB) uses genomic and phenotypic data to improve selection efficiency.
• Phenotypic data across diverse environments is crucial for GAB's broad applicability.

Purpose of the Study:

• To introduce the apple REFPOP, a large, multinational reference population designed to accelerate apple breeding.
• To evaluate the utility of REFPOP for genomics-assisted breeding programs.
• To assess the impact of SNP density on marker-trait association detection and genomic prediction.

Main Methods:

• Established the apple REFPOP with 534 genotypes across six European countries.
• Generated a high-density genome-wide dataset of 303,239 SNPs with high imputation accuracy (0.95).
• Measured two key phenological traits (floral emergence, harvest date) and analyzed marker-trait associations and genomic prediction abilities.

Main Results:

• Low linkage disequilibrium and weak population structure were observed in the REFPOP.
• Significant marker-trait associations were identified for floral emergence and harvest date.
• Maximum genomic prediction ability was achieved with approximately 10,000 SNPs, regardless of trait architecture.

Conclusions:

• The apple REFPOP is a valuable tool for accelerating genomics-assisted apple breeding, particularly for programs using related germplasm.
• A coordinated, multinational approach is advantageous for developing tailored breeding methods in the genomics era.
• The study confirms the suitability of the REFPOP design and provides insights into optimal SNP density for genomic prediction in apples.