Genomic Selection in Pigs: Are We There Yet ?

Yes!  High density (HD) SNP panels have revolutionized the field of animal breeding. The technology has moved beyond a “proof of concept” stage to where genomic selection can be implemented in breeding programs to improve the accuracy of the estimated breeding values (pigs), reduce generation interval (cattle) and advance genetic progress. The “first generation” of genomic selection using trait-specific marker panels was implemented by PIC in 2010 with 39-58% increases in accuracy for lowly heritable and hard to measure traits. This approach requires a large “training” data set with high quality phenotypes to estimate marker effects. Small panels continue to be an attractive option for phenotypes not available in a commercial setting (e.g., disease). The “second generation” of genomic selection no longer requires estimating individual marker effects. Using SNP genotypes we can predict with greater accuracy the fraction of the genome that is identical by decent - increasing the accuracy of relationships between individuals. Genus/PIC has HD genotyped >15k genetic nucleus animals and will HD genotype an additional 10k individuals by the end of the year for implementation of multivariate single step genomic evaluation. Imputation further allows a significant reduction in overall program cost of genotyping. This new technology provides an affordable alternative to HD genotyping of selection candidates required to differentiate between litter-mates. Genome sequencing shows promise to be the “third generation” of genomic selection, where breeding values could be estimated from actual or imputed individual sequence data.