Estimation of genetic parameters for average daily gain using models with competition effects

Area of Science:

• Animal Breeding and Genetics
• Quantitative Genetics
• Swine Production

Background:

• Accurate estimation of genetic parameters is crucial for effective genetic improvement programs in swine.
• Competition among pigs within a pen can influence individual performance and potentially bias genetic variance estimates.
• Previous models often did not fully account for complex competition effects (genetic and environmental).

Purpose of the Study:

• To estimate variance components for average daily gain (ADG) in swine, specifically examining the impact of competition effects.
• To compare different modeling approaches for partitioning genetic and environmental variances, including competition and pen effects.
• To determine the most practical and accurate method for accounting for competition in genetic evaluations.

Main Methods:

• Utilized data from 11,235 pigs across 4 selected lines.
• Employed mixed models including fixed effects (line, sex, contemporary group, initial age) and random effects (direct genetic, competition, pen, litter, residual).
• Compared models with and without specific competition (genetic/environmental) and pen effects to assess their influence on variance component estimates.

Main Results:

• Genetic competition variance was small but significant. Environmental competition effects appeared to be the source of pen variance.
• Ignoring competition or pen effects led to inflated estimates of direct-competition and genetic competition variances.
• Excluding both pen and competition effects inflated direct genetic variance estimates; including pen effects or environmental competition effects avoided bias in genetic variances.

Conclusions:

• Including either pen effects or environmental competition effects as random effects in the model is necessary to avoid bias in genetic variance estimates for ADG.
• While both methods correct for bias, incorporating pen effects is practically simpler for swine genetic evaluations.
• Competition effects, though small, have a measurable impact on variance partitioning and model interpretation.