ANIMAL-MODEL ESTIMATION OF ADDITIVE AND DOMINANCE VARIANCES IN EGG-PRODUCTION TRAITS OF POULTRY

An animal model analysis was used to estimate simultaneously additive (sigma(a)2) and dominance (sigma(d)2) variances for egg production traits within three White Leghorn lines. The data consisted of information for three generations on egg number (EN) produced at 18 to 25 (EN1), 26 to 65 (EN2), and 18 to 65 wk of age (EN3); egg weight (EW) measured at 30 to 35 (EW1) and 40 to 45 wk (EW2); and egg specific gravity (ESG) measured at 30 to 35 (ESG1) and 40 to 45 wk (ESG2). A transformation was used for EN2 and EN3 because of a skewed distribution. In total, 813 sires, 2,575 dams, and 28,649 daughters were involved in the analyses. Three genetic models (sire-dam, additive, and dominance) were compared in estimating heritability (h2). The sire-dam model underestimated h2 because it ignored animal relationships. The h2 estimates from the additive model were approximately 9 to 52% higher for EN and 2 to 18% higher for EW and ESG than those from the dominance model. The differences between the h2 estimates from the additive and dominance models were increased for larger dominance variance sigma(d)2. Ratios of sigma(d)2 to total variance were high for EN (10 to 20%) and low for EW and ESG (1 to 13%). Ratios of sigma(d)2 to total genetic variance for EN1, EN2, EN3, EW1, EW2, ESG1, and ESG2 were 18 to 36, 29 to 43, 29 to 56, 1 to 26, 3 to 8, 20 to 27, and 2 to 14%, respectively. The results on dominance were in good agreement with heterosis and inbreeding depression for these egg production traits described previously.