Macneil, Michael  
Lamberson, W  
Golden, B 
Submitted to: Journal of Animal Science Supplement
Publication Type: Abstract Only Publication Acceptance Date: 6/1/2001 Publication Date: 6/1/2001 Citation: MACNEIL, M.D., LAMBERSON, W.R., GOLDEN, B.L. IDENTIFICATION OF GERMPLASM FOR PRESERVATION FROM PEDIGREED POPULATIONS. JOURNAL OF ANIMAL SCIENCE SUPPLEMENT. 2001. v. 79(Suppl. 1). p. 11. Interpretive Summary: Cryogenic conservation programs seek to maximize genetic diversity in the conserved sample of germplasm. Breed associations record and maintain extensive pedigree databases for a wide variety of livestock populations. The objective of this research was to develop methods for identifying genetically diverse samples of animals from pedigree databases. Given a list of candidates and their pedigrees, coefficients of relationship (R) among them can be calculated. For large numbers of candidates, one suitable approach is to generate "pseudo progeny" from all possible pairs of candidates and compute the R for each pair of candidates from inbreeding coefficients of the pseudo progeny. The R can be used directly as in a procedure proposed for use in swine. That algorithm is initiated with an arbitrary animal or a set of preselected animals (perhaps ones that already have pools of semen available) and sequentially selects the animal with the lowest cumulative relationship to the previous set until a desired complement is attained. This procedure was tested on a simulated set of relationships among 100 animals. Repeated sets of 10 animals were chosen from the population by three methods: 1) random sampling; 2) use of the algorithm initiated with a random seed animal; and 3) use of the algorithm initiated with a set of five random seed animals. The mean of the relationships for the three procedures were: 0.102, 0.067, and 0.076, respectively, and SD were 0.014, 0.005, and 0.006, respectively. Alternatively, the R may be transformed (reciprocal, 1R, etc.) to distance measures. A cluster analysis procedure can then be used to identify a set of animals for cryopreservation. Technical Abstract: Cryogenic conservation programs seek to maximize genetic diversity in the conserved sample of germplasm. Breed associations record and maintain extensive pedigree databases for a wide variety of livestock populations. The objective of this research was to develop methods for identifying genetically diverse samples of animals from pedigree databases. Given a list of candidates and their pedigrees, coefficients of relationship (R) among them can be calculated. For large numbers of candidates, one suitable approach is to generate "pseudo progeny" from all possible pairs of candidates and compute the R for each pair of candidates from inbreeding coefficients of the pseudo progeny. The R can be used directly as in a procedure proposed for use in swine. That algorithm is initiated with an arbitrary animal or a set of preselected animals (perhaps ones that already have pools of semen available) and sequentially selects the animal with the lowest cumulative relationship to the previous set until a desired complement is attained. This procedure was tested on a simulated set of relationships among 100 animals. Repeated sets of 10 animals were chosen from the population by three methods: 1) random sampling; 2) use of the algorithm initiated with a random seed animal; and 3) use of the algorithm initiated with a set of five random seed animals. The mean of the relationships for the three procedures were: 0.102, 0.067, and 0.076, respectively, and SD were 0.014, 0.005, and 0.006, respectively. Alternatively, the R may be transformed (reciprocal, 1R, etc.) to distance measures. A cluster analysis procedure can then be used to identify a set of animals for cryopreservation.
