|Van Vleck, Lloyd|
Submitted to: Journal of Animal Science Supplement
Publication Type: Abstract Only
Publication Acceptance Date: 3/1/1998
Publication Date: N/A
Technical Abstract: Weaning weights from nine sets of Angus field data were 8128 to 11675 observations were analyzed. Six animal models were compared. Model 1 included random direct, maternal genetic, permanent maternal environmental, management-sex combination. Effects in other models were regression on maternal phenotype (B) in Model 2, sire by herd-year (SH) interaction (Model 3), B and SH interaction (Model 4), grandmaternal genetic and permanent environmental effects (Model 5), and interaction and grand- maternal genetic and permanent environmental effects (Model 6). Estimates of B and (co)variances were obtained by REML using either a derivative-free or an average information method. Estimates from Model 1 ranged from .25 to .50 for direct heritability (h2), from .11 to .17 for maternal herit- ability (m2), and from -.07 to -.57 for the correlation between direct and maternal genetic effects (ram). Small estimates of B from Models 2 and 4 were obtained (-.003 to .014) and (co)variance estimates were similar to models 1 and 3. Estimates of the relative variance due to SH interaction were .02 to .10 with estimates of h2 and m2 smaller and ram higher than in models without the interaction. Grandmaternal heritability estimates ranged from .02 to .07 with estimates of m2 and ram larger compared to Model 1. Likelihoods showed SH interaction effects to be more important than B and grandmaternal effects. Likelihoods were highest for Model 6 where estimates for h2, m2, and ram ranged from .17 to .32, .11 to .24 and -.35 to .06. Results suggest that the suspected downward bias in ram may not be due to assuming the dam-offspring environmental covariance to be zero but due to not considering SH interaction and grandmaternal effects.