The full potential of genomic selection is not yet achieved, since that would require DNA testing of all the selection candidates.
Today, boars selected for boar testing are chosen based on the parents breeding value, and thereafter genotyped before the final selection at the end of the boar test. In addition, 50 % of the sows that are producing boars to test are currently genotyped. However, several ongoing activities are aimed at bringing us closer to the full potential, including a new project delivering genotypes (DNA test) to a lower cost. The plan is to genotype all selected sows in the nucleus herds already in 2014.
Potentials are also seen in genotyping AI selection candidates and/or reference populations. This would increase the accuracy of the ≪satellite≫ breeding values for traits recorded in Norway only (e.g. CT recordings). Furthermore, genotyping of a reference population with recordings not present in Norway (e.g. ≪heat stress≫) would contribute to increased accuracy of selection.
Incredible achievements have been made with genomic selection in Norsvin, and in a few years the results will be reflected in the Norsvin herds. Genomic selection is, together with phenotyping and data recording, seen as the most important instruments over the next few years for further improving the accuracy of the breeding value estimations. However, we are still working on optimizing the models and design to produce an even better genetic program.