OPTIMIZING STRATEGIES FOR MARKER-ASSISTED SELECTION

225 Kildee Hall, Department of Animal Science, Iowa State University, Ames, IA 50011-3150 USA

Genetic improvement of livestock primarily focuses on selection for quantitative traits in outbred populations. To date, most genetic improvement has been achieved through selection on breeding values estimated from phenotype of the individual and/or its relatives. Molecular genetics is now providing tools to enhance rates of genetic improvement by being able to select on quantitative trait loci (QTL) or on linked markers. Sophisticated statistical methods have been developed to estimate the effects of QTL in complex pedigrees. The use of this QTL information in strategies for marker-assisted selection (MAS) has, however, received less attention. This is best illustrated by recent simulation studies (e.g. Gibson, 1994, Proc. Wld. Congr. Genet. Appl. Livest. 21:201), which showed that, although current strategies for MAS on a known QTL increase response to selection in the short term, they can lead to less response in the longer term than selection based on phenotype. Reduced longer term response is caused by the impact of increased emphasis on the known QTL on response in other genes that affect the trait. We (Dekkers and van Arendonk, 1998, Genetical Research 71:257) recently developed methods to optimize selection on a known QTL. Results show that QTL information can lead to greater responses to selection in both the short and longer term, in particular for QTL that exhibit dominance, provided selection on the QTL is optimally balanced with selection on phenotypic information. Implications for strategies for MAS and the benefits that can be expected from MAS in livestock breeding programs will be discussed.