Abstract
BACKGROUND: Closed populations are always at risk of losing genetic diversity needed for long-term genetic progress. When selecting, breeders are sometimes afraid of “missing out” on rare and/or beneficial genetic variation. Such genetic variation missing in a high-performing nucleus population may be found in low performing animals. The gap in performance level may hamper utilization of external genetic variation in the nucleus. In this study, we test different indicator variables to select animals as donors of diversity to be added to a commercial nucleus population via simulations. We also test a strategy of integration of external diversity found in those donors into an elite population in an ongoing genomic selection breeding program that employs optimum contribution selection. The core idea of restructuring breeding programs is to create layers through which diversity is gradually “upcycled” until it is ready for integration into the elite population. This strategy is assessed by means of simulation of a purebred pig breeding program over a time horizon of 20 generations. We screened boars as diversity donor candidates that have been selected in previous generations. RESULTS: The introduced variation did not help to outperform the genetic progress achieved by the control scenario (no diversity introduction) in any of the tested scenarios, though some scenarios did not perform much worse. Generally, the more resources are spent on the upcycling-component, the more the diversity in the elite can be increased, but this compromises genetic gain because less resources are available to advance the elite. Although the effects on genetic gain and diversity were similar among the diversity-selection-criteria, selecting boars with the lowest kinship to the elite population was slightly more effective and is the easiest strategy to implement, relative to selecting boars with unique haplotypes or haplotypes with a higher breeding value than in the elite. The efficiency of turning genetic diversity into genetic gain was always enhanced in breeding programs investing in diversity introduction. CONCLUSIONS: Though the tested layered breeding programs did not show an advantage in terms of gain over the control scenario, the strategies may be relevant for real breeding programs (1) if a population is lacking diversity for a trait or (2) if new traits become important. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12711-026-01046-9.