Genomic selection for the high-hanging fruit in livestock breeding programs

D.P. Berry, Y. de Haas, R.F. Veerkamp, M. Coffey, M.P.L. Calus

Research output: Contribution to conferenceAbstractAcademic

Abstract

Genomic selection has by now been adopted for most of the “low-hanging” traditionally measured traits in developed breeding programs. The benefit of incorporating genomic information into genetic evaluations, however, is greatest for economically important traits not routinely recorded and thus where only low accuracy of selection is being achieved using traditional approaches. Irrespective of species, difficult to measure traits (i.e., the high-hanging fruit) include feed intake and efficiency, environmental footprint, product quality, and animal health and disease. Options to develop genomic predictions for these high hanging fruits include (1) collation of (inter)national databases which in themselves are too small to achieve high accuracy of predictions, (2) exploitation of low-cost, easy to measure predictors, (3) development of an optimal reference population based on phenotypic and genomic diversity (and possible financial incentives or investment in collection of same), and (4) detection and exploitation of the causal mutations. A successful international initiative (gDMI) based on the collation of feed intake data from 9 countries in dairy cows concluded that exchange of phenotypic and genomic information can augment the accuracy of genetic/genomic evaluations. Such an approach should also be embarked on for other breeds and species. Steps should, however, be taken now in anticipation of such an initiative including the exchange of germplasm between research centers or the use of a pan-global list of sires. Maximizing phenotypic and genomic diversity (within the constraints of relatedness to the candidate population) could improve the accuracy of genomic predictions; whole populations could be screened using low-cost predictor traits (e.g., sensors, mid-infrared spectroscopy) and divergent animals collated to a centralized unit for deep phenotyping. Combining transcriptomic and genomic data could aid in the detection of causal mutations. Successful genomic selection will also require a re-evaluation of current phenotyping strategies and may include measurements on the parents themselves (e.g., feed intake on bulls) as currently undertaken in other species or half-/full-sibs (e.g., deliberate infection with pathogens).
Original languageEnglish
Publication statusPublished - 2015

Fingerprint Dive into the research topics of 'Genomic selection for the high-hanging fruit in livestock breeding programs'. Together they form a unique fingerprint.

Cite this