Comparison of evaluation-selection systems for maximizing genetic response at the same level of inbreeding.

Use of improved methods of genetic evaluation and intense selection lead to higher rates of genetic change but also lead to higher rates of inbreeding. A number of methods have been proposed to attain high rates of genetic response with moderate or low inbreeding (i.e., restrict the number of close relatives selected, use false high heritability in the genetic evaluation, use assortative mating, adjust estimated breeding values for the relationship with those already selected, avoid matings of related individuals, and increase the number of parents selected). These various methods were compared here using stochastic simulation of an infinitesimal genetic model. The results were studied graphically plotting response against inbreeding, so allowing comparison of the methods at each of a range of inbreeding levels. None of the methods was best over all conditions. At high rates of inbreeding and with a fixed number of sires, several of the methods gave large decreases in inbreeding for small losses in response. However, these were matched by simply increasing the number of sires selected. There is an optimum response-inbreeding "front" attained by several methods, and this front seems hard to breach. Characterization of the front would be useful in designing selection programs.