|Van vleck, Lloyd|
Submitted to: Journal of Animal Science Supplement
Publication Type: Abstract only
Publication Acceptance Date: 11/13/2008
Publication Date: N/A
Citation: Interpretive Summary:
Technical Abstract: Estimates of variance components were obtained from 475 records of average (AFI) and residual feed intake (RFI). Covariates in various (8) models included average daily gain (G), age (A) and weight (W) on test, and slaughter (S) and ultrasound (U) carcass measures (fat thickness, ribeye area and marbling score or intramuscular fat). All models included contemporary group (days on feed x pen x year). For AFI, choice of model affected estimates of genetic and residual variances and heritability. Heritability was largest (0.55) for model with A, W, and G. Other models resulted in similar heritability estimates (0.31-0.37) although components of variance varied (genetic: 0.10-0.23; residual: 0.20-0.50). Models including either S or U in addition to A and W resulted in similar estimates of parameters. For RFI (AFI adjusted for ADG and metabolic body weight), models with only A and W as covariates and with no covariates resulted in near null estimates of parameters. The largest estimate of heritability was with A, W, and G as covariates (0.61). Estimates of heritability were greater with covariates for slaughter as compared with ultrasound carcass measures (0.41 and 0.42 vs. 0.34 and 0.35). With the exception of the two sets of near null estimates for RFI, estimates of phenotypic variances were similar for AFI and RFI. Although heritability estimates (and accuracy of EBV) for AFI are similar with or without adjustment to constant carcass measures, standard errors of prediction with adjustment for carcass measures would be smaller because of smaller genetic variance.