SelAction: software to predict selection response and rate of inbreeding in livestock breeding programs.

Software was developed to predict selection responses and rates of inbreeding from livestock breeding programs. Variance components are needed as input for the program, which subsequently enables the user to evaluate various breeding program designs. Consequently, breeding program structures with discrete (e.g., fish or poultry breeding) or overlapping generations (e.g., pig or cattle breeding) can be evaluated for selection on multiple traits. For discrete generations, selection response of multistage selection schemes can be predicted as well. Rates of inbreeding can be predicted for breeding programs with discrete generations and single-stage selection. With this software program, BLUP breeding values (animal model) can be used as information sources. It also accounts for reduction of genetic variance due to selection (‘‘Bulmer effect’’), and for reduction of selection intensity due to finite population size and correlated index values of family members. Predefined information sources can be either switched on or off, according to the user’s needs. The number of traits is limited to a maximum of 20. Quantitative genetic research has resulted in a large amount of theory to predict selection response and rates of inbreeding of livestock breeding programs. Application of this theory in practical livestock breeding programs, however, is hampered by the lack of computer software that implements the available theory. Virtually all methods to predict responses of breeding programs are based on selection index theory. After the basics of selection index theory were developed (Hazel 1943), further developments have aimed at, for example, the effect of selection on genetic (co)variances (Bulmer 1971), the effect of correlated index values on selection intensity (Meuwissen 1991), the effect of multistage selection (Tallis 1961; Young 1964), selection in overlapping generations (Meuwissen 1989), and prediction of selection response on BLUP breeding values (Villanueva et al. 1993; Wray and Hill 1989). In addition, methods to predict rates of inbreeding in selection programs have recently been developed (Bijma et al. 2001; Bijma and Woolliams 2000; Woolliams and Bijma 2000). In this article, a computer program called SelAction is presented, which implements existing theory in order to provide a tool to predict selection response and rates of inbreeding in livestock breeding programs for a wide range of population structures and selection strategies. SelAction requires little computing time, allowing it to be used as an interactive optimization tool.