Genomic selection using different marker types and densities.

Area of Science:

• Animal Breeding and Genetics
• Quantitative Genetics
• Genomic Prediction

Background:

• Genomic selection (GS) offers potential for accelerated genetic gain compared to traditional methods.
• High-density marker maps and cost-effective genotyping enable advanced genomic selection strategies.
• Understanding factors influencing GS accuracy is crucial for optimizing breeding programs.

Purpose of the Study:

• To investigate the impact of marker density and type (microsatellites vs. SNPs) on genomic selection accuracy.
• To compare the efficacy of using marker haplotypes versus individual marker genotypes.
• To evaluate the influence of marker density on the accuracy and bias of genomic estimated breeding values (GEBVs).

Main Methods:

• Simulated a population over 1,000 generations to achieve mutation-drift balance and obtain marker characteristics.
• Tested various marker densities for microsatellites (0.25-2N(e)/morgan) and SNPs (1-8N(e)/morgan).
• Evaluated prediction accuracy by estimating marker effects and haplotype effects (combining 2 markers).

Main Results:

• Increased marker density significantly improved selection accuracy for both microsatellites and SNPs.
• SNP markers required 2-3 times higher density than microsatellites for comparable accuracy.
• Using direct marker effects yielded higher or similar accuracies compared to using haplotypes, avoiding phase estimation errors.
• GEBVs showed small bias, suggesting potential need for rescaling in practical applications.

Conclusions:

• Higher marker density, particularly with SNPs, enhances genomic selection accuracy.
• Direct marker effects are preferable to haplotypes for practical genomic selection due to simplicity and accuracy.
• Approximately 24,000 SNPs are recommended for a 30-morgan genome with an effective population size of 100 for accurate genomic prediction.