Genomic selection is based on breeding values predicted from a large number of genetic marker effects across the whole genome. The marker effects are estimated using a reference dataset, that contains known marker genotypes and known phenotypic performance from typically >2000 animals. This is one of the main challenges in genomic selection: to estimate tens or hundreds of thousands of marker effects from a limited number of phenotypic records. However, the major dairy countries have methodologies developed, tested and implemented. Total breeding values for juvenile selection candidates that have no phenotypic records but known marker genotypes, are predicted as the sum of the estimates corresponding to their marker genotypes. This prediction already has a high accuracy early in life, and for instance relaxes the requirement to progeny test dairy bulls before they are used widely in the population. Costs of a breeding program are considerably reduced, the generation interval is shortened drastically, and faster genetic gains are achieved. This has a major impact on the outline of breeding programs that apply genomic selection. Current developments include extensions to multi-trait models, increasing the reference populations, and increasing the numbers of SNPs used.
|Title of host publication||Workshop Abstracts 14th Annual Conference of the European Society for Domestic Animal Reproduction (ESDAR), 15 - 18 September, 2010, Heger, Hungary|
|Place of Publication||Berlin|
|Publication status||Published - 2010|
|Event||14th Annual Conference of The European Society for Domestic Animal Reproduction (ESDAR), Eger, Hungary - |
Duration: 14 Sep 2010 → 18 Sep 2010
|Conference||14th Annual Conference of The European Society for Domestic Animal Reproduction (ESDAR), Eger, Hungary|
|Period||14/09/10 → 18/09/10|