|VAN VLECK L DALE|
Submitted to: Journal of Animal Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/6/1995
Publication Date: N/A
Interpretive Summary: Variants of the animal model are commonly used for National Cattle Evaluations within breeds. Results from analyses of records of progeny of sires of 22 breeds mated to Angus and Hereford dams at MARC suggest that if a version of the animal model were implemented to carry out joint genetic evaluations of beef cattle of several breeds, then that model should take into account differences in phenotypic variances for breeds and for sexes for 200- and 365-d weights. Differences in variances due to maternal effects for breeds of dams also should be considered. In the present study of 200- and 365-d weights, correlations between random sire and dam effects for records of calves from different sex-dam breed sub- classes were not different from unity which indicates differences in variation are due to scale rather than to different partitioning of total variation. Thus, simplifications for a multivariate animal model such as those proposed by Quaas et al. (1989) could be applied for these traits. Failure to account for differences in variances can lead to incorrect selection decisions and reduction in rate of genetic improvement.
Technical Abstract: Heterogeneity of variance for 200- and 365-d weights for sex, sire and dam breed subclasses was studied. Data comprised weaning (WW; n = 7829) and yearling (YW; n = 7367) weights of progeny from 673 sires from 22 breeds evaluated in the Germ Plasm Evaluation Program at the U.S. Meat Animal Research Center. Sires were mated to Hereford and Angus cows. Three sire and dam models with records of animals of the four combinations of sex and dam breed considered different traits and differing in variance-covariance structure were fitted for each trait, and each of the factors, sex and parental breed, causing heterogeneity of variance. For each factor, (co) variances across subclasses were assumed to be: 1) the same; 2) the same as fractions of phenotypic variance that is different for each subclass; and 3) different, except that (co)variances for dam effects were estimated in common for all sire breeds. Variance components were estimated by derivative-free REML. Models were compared with likelihood ratio tests. For both traits, variances differed (P<.02) in scale, but not as fractions of phenotypic variances (P>.10) for sire breed and sex subclasses. Vari- ances were not different (P>.10) by dam breed, either in scale or as fract- ions. Phenotypic variances by sire breed for WW (kg) ranged from 233 to 749 and for YW (kg) from 686 to 1388. Variances for males were 1.2 and 1.5 times the variances for females for WW and YW. Correlations among genetic effects on weights of calves from different sex-dam breed subclasses were at least .85. Pooled estimates of sire and dam variances as fractions were, respectively, .06 and .39 for WW and .11 and .24 for YW. The concl- usion is that differences in phenotypic variances among sire breeds and between sexes should be considered in genetic evaluations.