Genomic selection -: basics and perspectives for dairy cattle breeding programs

Practical improvements of dairy cattle breeding programs through marker assisted selection were behind theoretical expectations. Genetic gain is still based on traditional recording systems and progeny testing programs, but the theoretical potential is limited due to long generation intervals. Genomic selection enables the opportunity for substantial modifications of genetic evaluations and breeding programs as well. Genomic selection is possible due to a multitude of so called SNPs, distributed more or less evenly over the whole genome, and costs for SNP-genotyping through microarray technology are relatively cheap compared to microsatellite markers. The genomic breeding value is the sum of all single SNP effects. Reliable genomic breeding values for both sexes (accuracy > 0.75) can be calculated in an early stage of an animal's life, e.g. even for embryos. This implies a shift from BLUP animal models including pedigree information towards SNP based BLUP, and putting less weight on information provided by relatives. Economic efficiency and an increase of selection response in dairy cattle breeding programs are due to the entire abdication of progeny testing. Such a strategy implies a substantial reduction in generation intervals. Model calculations for the Canadian Holstein population showed a doubling of selection response per year and a reduction of breeding costs by 90 percent when comparing genomic selection to the traditional selection scheme. However, unresolved problems related to the methodology for the estimation of SNP-effects have to be clarified in advance.