| The purpose of this study are to simulate non-linear curve and estimate variance components and genetic parameters of growth traits in different months of Sanhe cattle and cross offspring.Data collected in2007-2010records at Rochelle Tara cattle Farm in Hailar Agriculture and animal husbandry Bureau. The traits in this study were body length, body height, scrotal circumference, body weight. This study consists of two parts:(1) Traits were simulated with DUD method of SAS8.0NLIN process. The best model was selected in line with the maximum of goodness of fit and minimum of Residual sum of squares, then draw growth curve and reveal grow and development rule for each trait.(2) First, non-genetic elements were analyzed by SAS8.0GLM process and highly significant factors were as the fixed effects in next study. Then, By ignoring or including maternal additive genetic effects,and considering whether direct additive genetic and maternal additive genetic effects were correlated or not, three different single-trait models were obtained. A log likelihood ratio test was conducted to choose the most appropriate model. Finally, Variance components and genetic parameters were estimated by using AIREML methodology under single-trait and multi-traits animal model.The results of non-linear simulation with growth traits shows the appropriate model of body height, body length, scrotal circumference, body weight were Logistic, Bertalanffy, Gompertz, Bertalanffy, respectively. The corresponding best models for the traits of Germen progeny were Logistic, Gompertz, Gompertz, Bertalanffy. Meanwhile the best non-linear models for France progeny were Logistic, Brody, Logistic, Brody.The results of non-genetic factors analyzed indicated year had highly significant effect on scrotal circumference and body weight at birth(P<0.01), but had only significant effect on body length and body height(P<0.05) at birth. Other three factors (season, breed and parity had no significant effect on each trait at birth (P>0.05). Body height, body length and scrotal circumference at weaning were affected highly by year and season (P<0.01), but other factors had no effect on them (P>0.05).Year had highly significant effect on body weight at weaning (P<0.05).And season had significant effect on body weight at weaning (P<0.01). Year and season had highly significant on body height, scrotal circumference and body weight at yearling (P<0.01). Year had significant effect on body length (P<0.05), and season had highly significant effect on body length at yearling (P<0.01).However breed and parity had no significant effect on each trait at different ages.The estimate of variance components of growth traits at different ages demonstrated the log likelihood ratio test had no difference among three models with each trait. In accordance with the principle of model single and minimum of residual sum of squares, individual direct addictive genetic effect had significant effect on each trait at different ages. Otherwise, maternal direct addictive genetic effect had no effect on each trait. Therefore, the appropriate model for every trait at different ages was model (1).The heritability for body weight, body height, body length and scrotal circumference at birth were0.3297、0.3289、0.4157、0.4135. The values for corresponding traits at weaning were0.2941、0.3424、0.4637、0.3921. And the heritability for above traits at yearling were0.3024、0.3397、0.4665、0.3024. The genetic correlation between body measurements traits ranging from0.3023to0.7527, the phenotypic correlation between them were0.1329-0.5719. The genetic correlation for body weight at different ages were0.4180-0.5956, the phenotypic correlation rang from0.1778to0.3758. The genetic correlation among four traits at birth, weaning and yearling in this study were moderate to high, the value vary0.7339-0.990,0.7131-0.9263,0.6226-0.9792. the phenotypic correlation were0.6305-0.8853,0.5682-0.7593,0.5081-0.6736,respectively. |
PDF Full Text Request