Genomic prediction in the wild: A case study in Soay sheep

Abstract

Genomic prediction, the technique whereby an individual’s genetic component of their phenotype is estimated from its genome, has revolutionised animal and plant breeding and medical genetics. However, despite being first introduced nearly two decades ago, it has hardly been adopted by the evolutionary genetics community studying wild organisms. Here, genomic prediction is performed on eight traits in a wild population of Soay sheep. The population has been the focus of a >30 year evolutionary ecology study and there is already considerable understanding of the genetic architecture of the focal Mendelian and quantitative traits. We show that the accuracy of genomic prediction is high for all traits, but especially those with loci of large effect segregating. Five different methods are compared, and the two methods that can accommodate zero-effect and large-effect loci in the same model tend to perform best. If the accuracy of genomic prediction is similar in other wild populations, then there is a real opportunity for pedigree-free molecular quantitative genetics research to be enabled in many more wild populations; currently the literature is dominated by studies that have required decades of field data collection to generate sufficiently deep pedigrees. Finally, some of the potential applications of genomic prediction in wild populations are discussed.

The utility of genomic prediction models in evolutionary genetics

“…[16]European ash ( Fraxinus excelsior )ash dieback damage150 f >10 k g DNS0.35RRBLUPAshraf et al . [17] h Soay sheep ( Ovis aries )weight1168∼36 k0.320.51BayesRjaw length897∼36 k0.590.38BayesRforeleg length1126∼36 k0.530.62BayesRhindleg length1139∼36 k0.500.59BayesRmetacarpal length890∼36 k0.620.65BayesRmale horn length472∼36 k0.420.67BayesRcoat colour4737∼36 kDNS1.00BayesRcoat pattern4737∼36 kDNS0.98BayesRHunter et al . [15] i Soay sheep ( Ovis aries )adult body weight1168∼36 k0.34–0.49 [58]DNSBayesR a Heritability from random additive line effect g in ASReml; DNS, ‘data not shown’. b Broad-sense heritability. c Depending on model, SNPs used and sex.…”

“…Thus, the predicted breeding value is typically defined only for a target set of environmental conditions. One interesting example of how to deal with this complicating factor comes from [15,17] who opted to fit mixed models that incorporated fixed effects, non-genetic random effects and repeated measures. The random effect of individual identity was extracted and incorporated as the phenotype in genomic prediction.…”

Using genomic prediction to detect microevolutionary change of a quantitative trait

“…Genomic prediction of GEBVs was performed using the BayesR method [30] implemented in the BayesR v0.75 software package [31]. We have previously shown that BayesR-derived GEBVs of Soay sheep morphological traits, including adult weight, have a high accuracy (~0.64) [29]. BayesR models SNP effects as a number of distributions of different effect sizes, including one of zero effect.…”

“…In the BayesR models all phenotyped and genotyped animals (n=1168) were treated as a training population and all animals for whom we had genotypes but no phenotypes (n=5627) were the test population. The phenotypes used in the BayesR analysis were obtained by first fitting a linear mixed model (see [29]) that included individual identity (to account for repeated measures), birth year and capture year as random effects and sex and age as fixed effects. The random effect of individual identity was used as the phenotype.…”

“…Soay sheep adult weight is a polygenic trait with a moderate heritability and no individual loci were significant in previous GWAS [20,25,29]. Genomic prediction of GEBVs was performed using the BayesR method [30] implemented in the BayesR v0.75 software package [31].…”