|Van Vleck, Lloyd|
Submitted to: Journal of Animal Science Supplement
Publication Type: Abstract Only
Publication Acceptance Date: 11/17/2006
Publication Date: 7/20/2007
Citation: Chen, C., Kachman, S.D., Johnson, R.K., Van Vleck, L.D. 2007. Estimation of genetic parameters in selected lines of swine using models with competition effects [abstract]. Journal of Animal Science. 85 (Supplement 2):61. (Abstract #42) Interpretive Summary: No interpretive summary is required.
Technical Abstract: Components of variance for average daily gain (ADG, g) were estimated from data provided by PIC on 11,235 pigs from 4 lines. Pigs with average age of 71 d were randomly assigned to pens of size 15 by line and sex and taken off test after approximately 89 days (weights ranged from 61 to 158 kg). Models included fixed effects of line, sex and contemporary group, with random direct genetic, competition (genetic and environmental) and pen effects. With the full model, estimates were 3630, 13 and 18 for direct (d), direct-competition (dc) and genetic competition (cg) (co)variance components. Pen (pn), environmental competition (ce) and residual (e) variances could not be partitioned. Estimable combinations were (pn) + 14 (ce) + (e) and (pn) + 13 (ce) corresponding to the diagonal and off diagonal elements of the pooled environmental (co)variance matrix. Model 2 (without environmental competition effects) allowed (co)variances to be partitioned (3636, 13, 18, 1746 and 6863 for d, dc, cg, pn and e) with the same likelihood as the full model. With pen effects ignored with Model 3, 13 (ce) (13 x 134 = 1742) corresponded to pen variance with Model 2 with a decreased estimate of (e = 6734) but with similar estimates of genetic variances and the same likelihood as the full model. The smaller estimate of (e) with Model 3 was associated with (ce) compared with Model 2 (6734+134=6868 vs. 6863). With Model 4 (both pen and genetic competition effects ignored), the estimate of (ce = 175) increased significantly with little change in (d = 3679) compared with Model 3. The impact of competition effects can be important depending on group size even with a relatively small estimate of (cg). Including environmental competition effects as permanent environmental effects in the model did not change estimates of genetic (co)variances. Competition environmental effects seemed to be a cause of observed pen variance. Therefore including either pen or environmental competition as random effects in the model seems desirable to avoid bias in estimates of genetic variances.