不同动物模型对高山美利奴羊早期生长性状遗传参数估计的比较

Comparison and Analysis of Genetic Parameters Estimation of Early Growth Traits of Alpine Merino Sheep by Different Animal Models

【目的】研究不同动物模型对高山美利奴羊早期生长性状遗传参数估计的影响,筛选适合高山美利奴羊早期生长性状遗传参数估计的最佳动物模型并估计早期生长性状的遗传参数,为高山美利奴羊早期阶段的选育提高提供理论依据。【方法】运用ASReml软件对固定效应进行F检验,判断固定效应中的因子是否显著,筛选出对高山美利奴羊早期生长性状具有显著影响作用的固定效应;其次运用ASReml软件中的AIREML算法通过不同动物模型估计高山美利奴羊早期生长性状的遗传参数,根据各模型随机效应的不同构建了4个动物模型,各模型中均包含固定效应、随机效应和残差效应,其中模型1的随机效应包括个体加性遗传效应,模型2的随机效应包括个体加性遗传效应、母体遗传效应,模型3的随机效应包括个体加性遗传效应、个体永久环境效应,模型4的随机效应包括个体加性遗传效应、母体遗传效应和个体永久环境效应。最后,通过赤池信息量准则(AIC)指数和似然比检验(LRT)对不同模型中的随机效应进行比较分析筛选出最佳动物模型,利用最佳动物模型估计出高山美利奴羊早期生长性状的遗传参数。【结果】(1)血统类型、出生年份、配种月份、初生月份、群别、性别以及出生类型对高山美利奴羊早期生长性状初生重、断奶重、断奶前平均日增重和断奶毛长均有极显著的影响(P<0.001),除性别和出生类型对妊娠期的影响不显著外,其余各固定效应对妊娠期的影响极显著(P<0.001)。(2)各模型估计的初生重遗传力为(0.0924±0.0160)—(0.2073±0.0226),母体效应的遗传力为0.1623±0.0113;断奶重遗传力为(0.0651±0.0126)—(0.1027±0.0159),母体效应的遗传力为(0.1097±0.0407)—(0.1098±0.0112);断奶前平均日增重的遗传力为(0.0681±0.0130)—(0.1001±0.1061),母体效应的遗传力为0.0898±0.0112;断奶毛长的遗传力为(0.0865±0.0148)—(0.0937±0.0149),母体效应的遗传力为0.0173±0.0107;妊娠期的遗传力为(0.0902±0.0174)—(0.1119±0.0189),母体效应的遗传力为0.0477±0.0146。(3)通过赤池信息量准则(AIC)指数和似然比检验(LRT)对不同模型中的随机效应进行比较分析发现,早期生长性状受个体加性遗传效应和母体遗传效应影响极显著,而受个体永久环境效应的影响可以忽略不计。因此,模型2是高山美利奴羊早期生长性状遗传参数估计的最佳动物模型。【结论】高山美利奴羊早期生长性状受母体遗传效应的影响比其他随机效应更显著;模型2为高山美利奴羊早期生长性状遗传参数估计的最佳动物模型。基于最佳动物模型估计的初生重、断奶重、断奶前平均日增重、断奶毛长、妊娠期的遗传力分别为0.0924±0.0160、0.0651±0.0126、0.0681±0.0130、0.0865±0.0148、0.0902±0.0174,母体遗传效应分别为0.1623±0.0113、0.1098±0.0112、0.0898±0.0112、0.0173±0.0107、0.0477±0.0146。

【Objective】This study aims to investigate the effect of different animal models on the estimation of genetic parameters of early growth traits of Alpine Merino Sheep, and to select the best animal model. And the best model was used to estimate the genetic parameters of early growth traits, which can provide the theoretical basis for the breeding of Alpine Merino Sheep in the early stage. 【Method】 Firstly, F-test was performed using ASReml software to determine whether the factors in the fixed effects were significant. And the fixed effects which have significant influence on the early growth traits of Alpine Merino sheep were selected for further analysis. Secondly, four animal models were constructed considering different random effects. And genetic parameters in each model were estimated by the Average Information Restricted Maximum Likelihood（AIREML） in ASReml software. Individual additive genetic effect was considered as random effect in model 1. Besides individual additive genetic effect, maternal genetic effect, individual permanent environmental effect was considered as random effects in model 2 and model3, respectively. Finally, the optimal animal model was selected by comparing the random effects in different models with the Akzo Information Criterion（AIC） Index and Likelihood Ratio Test（LRT）. 【Result】（1） The results showed that bloodline, gender, birth type, birth year, birth month, mating month and flocks highly affected the early growth traits of Alpine Merino sheep（P0.001）. Except gender and birth type, other fixed effects were significant on gestation length（P0.001）.（2） The direct heritabilities were（0.0924±0.0160）-（0.2073±0.0226）,（0.0651±0.0126）-（0.1027±0.0159）,（0.0681±0.0130）-（0.1001±0.1061）,（0.0865±0.0148）-（0.0937±0.0149） and（0.0902±0.0174）-（0.1119±0.0189） for BWT, WWT, ADG, WSL, GL, respectively. The maternal heritabilities were 0.1623±0.0113,（0.1097±0.0407）-（0.1098±0.0112）, 0.0898±0.0112, 0.0173±0.0107 and 0.0477±0.0146 for birth weight（BWT）, weaning weight（WWT）, pre-weaning average daily gain（ADG）, weaning staple length（WSL）, gestation length（GL）, respectively.（3） The comparative analysis of the random effects in different models by the Akaike information criterion（AIC） and the likelihood ratio test（LRT） showed that the early growth traits were significantly affected by the individual additive genetic effects and maternal genetic effects, but the individual permanent environmental effects were negligible. Therefore, the model 2 was the best model for the early growth traits of Alpine Merino Sheep. 【Conclusion】 The early growth traits of Alpine Merino Sheep were affected by maternal genetic effects more than other random effects. Model 2 was the best animal model for estimating the genetic parameters of early growth traits of Alpine Merino Sheep. Based on the best animal model, the direct heritabilities were 0.0924 ± 0.0160, 0.0651 ± 0.0126, 0.0681 ± 0.0130, 0.0865 ± 0.0148, 0.0902 ± 0.0174 for BWT, WWT, ADG, WSL, GL, respectively. The maternal heritabilities of these traits were 0.1623 ± 0.0113, 0.1098 ± 0.0112, 0.0898 ± 0.0112, 0.0173 ± 0.0107 and 0.0477 ± 0.0146, respectively.