Genomic selection on breeding time in a wild bird population

Area of Science:

• Evolutionary Biology
• Quantitative Genetics
• Conservation Genetics

Background:

• Genomic selection (GS) offers advantages over traditional phenotype-based selection by utilizing marker genotypes.
• GS is limited in wild populations due to high effective population sizes and marker density requirements.
• Advances in high-throughput genotyping make GS potentially applicable to wild populations.

Purpose of the Study:

• To assess the feasibility and accuracy of genomic selection in a wild bird population.
• To apply the GBLUP (Genomic Best Linear Unbiased Prediction) approach using a marker-based relatedness matrix.
• To evaluate the impact of marker number and relatedness estimators on prediction accuracy.

Main Methods:

• Utilized a dataset of approximately 2000 wild great tits (Parus major) genotyped on a 650k SNP chip.
• Employed the GBLUP approach, replacing pedigree-based relatedness with a marker-based relatedness matrix.
• Calculated genomic breeding values for egg-laying date and assessed accuracy by correlating with independent phenotypes.

Main Results:

• Achieved a prediction accuracy of approximately 0.20 for egg-laying date in wild great tits.
• The number of single nucleotide polymorphisms (SNPs) strongly influenced accuracy, while the relatedness estimator had minimal effect.
• The model effectively distinguished genetic from permanent environmental effects using the marker-based relatedness matrix.

Conclusions:

• Genomic selection is a viable tool for wild populations, even those with low heritability traits like avian breeding time.
• This approach has broad implications for evolutionary studies and conservation efforts in wild species.
• Cost-effective genotyping technologies are key to enabling GS in non-domesticated populations.