Genetic Parameters for Growth Traits in Hampshire Sheep in Mexico
Univariate and bivariate linear models via Restricted Maximum Likelihood (ASReml) were used to estimate heritability, phenotypic and genetic correlation for growth traits measured at birth (BW), weaning (WW) around 60 days, 90 days (W90), 120 days (W120) and 150 days (W150) in Hampshire sheep raised in Mexico. From 2005 to 2009 a total of 1,133 individual records of lambs born on 10 farms from 612 ewes and 63 sires were analysed. Direct heritability estimates for BW, WW, W90, W120 and W150 were 0.38±0.11, 0.15±0.08, 0.17±0.09, 0.18±0.07 and 0.14±0.06, respectively. All direct and maternal permanent environmental effect correlations were positive for BM, WW, W90, W120 and W150. The phenotypic correlations between all traits were positive and ranged from 0.29 to 0.96. The genetic correlations among growth traits were positive ranging from 0.35 to 0.94. The genetic parameter estimates presented here can be used to estimate breeding values to support genetic improvement programs for the Hampshire breed in Mexico.
The impact of genotype by environment interaction on breeding values for 150-day weight in Katahdin sheep in Mexico
With the objective of evaluating the impact of genotype by environment interaction (G by E) on breeding values for 150-day weight in Mexican Katahdin sheep, data from a total of 41,323 lambs, the progeny of 1,862 sires, were used to estimate genetic correlations between seven regional environments representing the majority of Mexico. Estimates of heritability within environments ranged from 0.24 ± 0.04 for Pacific Central to 0.42 ± 0.11 for North East. Genetic correlations across environments averaged 0.51 across all pairs, ranging from 0.07 ± 0.61 to 0.86 ± 0.27, indicating the presence of G by E interaction. A validation study predicted progeny performance within each environment with and without sire by flock effects from sire breeding values (EBVs) calculated from single trait BLUP analyses of data in the remaining environments. The regression of offspring performance on sire EBV were predictable across en vironments, but at lower levels than the expected value, in the absence of G by E, of 0.5. Fitting sire by flock improved the predictability with the regression coefficient increasing from 0.31 to 0.36.