Design of Reference Populations for Genomic Selection in Crossbreeding Programs

In crossbreeding (CB) programs, genomic selection offers the opportunity to make efficient use of information on crossbreds in the selection of purebred candidates (PB). In a deterministic simulation study, we evaluated the use of various proportions of crossbred animals in a reference population for genomic selection of purebred animals used in a crossbreeding program. When the breeding objective is to improve crossbred performance, the optimal proportion of CB in the reference population is 100%, but most of the potential gains are achieved if this proportion is only 50%. Optimal proportions and relative loss depend on the emphasis in the breeding objective on PB and the correlation between PB and CB performance. In most cases, adding CB animals to the reference population is beneficial, but the added value is lower and can become quite marginal if the proportion is higher than ~50%