利用广义线性模型预测家畜离散性状育种值

将广义线性混合模型(GLMM)引入动物离散性状的遗传分析及个体的遗传评定,初步比较了GLMM方法与一般线性方法(LM)的估计效果。模拟研究的性状为单阈值二项分类性状,选用的连接函数为对数连接μi=eη/(1+eη),方差函数为V(μi)=μ(i1-μi)/n,试验设计为全同胞-半同胞混合家系,参数估计采用Fisher迹法。结果表明:GLMM方法能较准确地估计公畜的个体育种值,在个体的遗传评定效果方面要明显优于常规的线性方法,其预测的育种值排序结果与真实育种值的排序之间存在极显著的相关性(P<0.001)。     

家畜离散性状遗传分析方法比较研究

采用广义线性方法（GLMM）、阈性状法（TM）和常规线性方法（LM）对不同参数组合下离散性状的遗传力及准确度进行了估计,模拟研究的性状为单阈值二项分类性状,试验设计为全同胞一半同胞混合家系。研究结果表明,在遗传力的估计准确度方面,GLMM方法具有较大的优势,LM方法估计遗传力的效果较差,离真值的偏差也较大;不同参数组合下,GLMM方法的遗传力估计结果普遍高于线性方法。另外,性状遗传力真值和性状表型发生率对估计遗传力及其准确度也有明显的影响。     

复合育种值在畜牧生产中的应用研究

复合育种值是根据家畜个体本身、祖先、同胞及后裔的性能资料，采用估计育种值的方法计算所得。利用宁夏地区某牛场14头荷斯坦奶牛及其各种亲属（母亲、同胞、女儿）的产奶量对其产奶量复合育种值进行了计算，并根据复合育种值的高低对所选14头奶牛进行了排序，其中9号、13号、8号及12号奶牛具有较高的复合育种值，分别为6611．78、6519．58、6511．78和6506．08kg。据调查，此种排列与实际生产中表现相一致。因此，复合育种值为奶牛选种选配提供了重要依据，在家畜育种生产中具有重要的实践意义。     

优质鸡睾丸重的遗传选择

为了提高优质鸡睾丸重,试验采用屠宰法测定天露黄鸡N409品系08世代第1批次的睾丸重,结合系谱信息,采用平均信息约束最大似然法(AI-REML)估计其遗传参数,采用动物模型最佳线性无偏预测(BLUP)法估计同胞个体睾丸重育种值,基于育种值进行选种和组建双向选择系。结果表明:未经正态转换和经过Box-Cox转换的105天睾丸重遗传力分别为0. 64和0. 66,属于高遗传力性状;表型正态与否并不影响候选同胞个体的育种值排序;经睾丸重育种值排序后筛选出小睾丸、中睾丸和大睾丸公鸡,与母鸡组建双向系,其后代公鸡睾丸重差异显著(P0. 05),而体重差异不显著(P0. 05)。说明基于上述方法提高优质鸡睾丸重可行,而且选择睾丸重不影响后代体重。     

优质型黄羽肉鸡腹脂的遗传选择

为降低优质鸡腹脂重,本实验以天露黄鸡父系作为研究材料,通过屠宰测定腹脂重,结合系谱信息,利用最佳线性无偏预测(BLUP)估计同胞腹脂率育种值后选留优秀个体,组建家系并繁殖下一世代。结果表明:天露黄鸡的腹脂重、腹脂率遗传力均为0.36;屠宰测定8世代参考群体,并以公母合并腹脂率作为已知表型估计第2批次候选群体育种值,对其进行排序选种,后代公、母鸡腹脂率分别下降了45.6%和14.1%;如果只以母鸡腹脂率作为已知表型估计候选群体育种值,则有87.7%的选留个体和公母合并腹脂率估计育种值进行选种时的选留个体相同,因此只屠宰母鸡有利于在生产上节约劳动成本,而同批次公鸡则可以作为候选个体,对其进行选择可进一步提高公鸡的留种率。     

种鹅产蛋数不同选育方法模拟研究

利用连续3个世代鹅的产蛋记录和系谱数据,对鹅产蛋性状不同的选育方法进行了模拟研究。利用系谱数据,G1和G2世代产蛋数据,估计个体育种值,并依据表型和育种值数据,对G2群体按照个体表型、家系表型、个体育种值、家系育种值进行排序,各选择30%个体,寻找其在G3世代中的后代,计算不同选择方法所产生后代的表型和遗传进展,评估不同选择方法对鹅产蛋数选育的最优方法。结果表明:个体育种值选择家系育种值选择家系选择个体表型选择效果;对公鹅的选择效果对母鹅的选择效果。     

约克夏猪繁殖性状直接的和母体的加性遗传效应分析

畜禽遗传改良的速度有赖于使生产者能准确地度量生产性能和评估遗传价值的所用手段。建立在动物模型基础上的混合模型方法已成为育种值估计的重要方法,这不仅是因为方法提供了育种值的最佳线性无偏预测值（ＢＬＵＰ）,而且更是因为方法考虑了动物个体间的亲缘关系,同时...     

奶牛体型线性性状的动物模型遗传评估

本文利用北京市7个奶牛场的奶牛体型线性评定资料,累计收集了4年共1647头母牛的15个线性性状数据,采用最佳线性无偏预测(BLUP,Best Linear Unbiased Prediction)方法,配合个体动物模型(AM,Animal Model),估计了胎次、泌乳月、鉴定日期、奶牛场以及鉴定员等固定环境效应,并计算出了每头奶牛15个线性性状的个体育种值。鉴于奶牛体型线性评定方法在国内正处于推广应用阶段,而BLUP的动物模型方法目前在国内的研究应用也少见报道。本文通过将两者结合起来,估计动物个体的种用价值,结果表明奶牛体型线性评定以第三胎、第三个泌乳月效果较好。经济效益好的场队其牛群也较整齐。不同的鉴定日期、不同的鉴定员对评定结果也略有影响。最后,可以得到剔除固定效应并经过排序用于选择的奶牛个体育种值。处理分析表明效果比较理想,为推动我国的动物育种工作,建议今后在国内大力推广奶牛体型线性评定和动物模型BLUP评估相结合的育种方法和手段。     

畜禽基因组选择方法研究进展

畜禽的选种选育在生产中至关重要,育种值估计是选种选育的核心。基因组选择（genomic selection,GS）是利用全基因组范围内的高密度标记估计个体基因组育种值的一种新型分子育种方法,目前已在牛、猪、鸡等畜禽育种中得到应用并取得了良好的效果。该方法可实现畜禽育种早期选择,降低测定费用,缩短世代间隔,提高育种值估计准确性,加快遗传进展。基因组选择主要是通过参考群体中每个个体的表型性状信息和单核苷酸多态性（single nucleotide polymorphism,SNP）基因型估计出每个SNP的效应值,然后测定候选群体中每个个体的SNP基因型,计算候选个体的基因组育种值,根据基因组育种值的高低对候选群体进行合理的选择。随着基因分型技术快速发展和检测成本不断降低,以及基因组选择方法不断优化,基因组选择已成为畜禽选种选育的重要手段。作者对一些常用的基因组选择方法进行了综述,比较了不同方法之间的差异,分析了基因组选择存在的问题与挑战,并展望了其在畜禽育种中的应用前景。     

动态性状育种值的估计(预测)方法探讨

为探讨更具一般性的动态性状育种值估计方法,在二步估计法的基础上建立了用于动态性状育种值估计(预测)的数学模型,采用该模型可实现对动态性状育种值的一步估计(预测),并保证估计结果的最优性。按线性与非线性模型两种情况分别论述了动态性状育种值的估计(预测)方法。结果表明:与随机回归模型相比,所提出的数学模型不仅具有明显的优越性,更具一般性;还证明了当每个个体动态性状的观测点一一对应且个体之间不存在亲缘关系时,一步分析法与二步分析法的估计结果相同。     

Overview of genetic evaluation in dairy cattle

It is costly and time‐consuming to carry out dairy cattle selection on a large experimental scale. For this reason, sire and cow evaluations are almost exclusively based on field data, which are highly affected by a large array of environmental factors. Therefore, it is crucial to adjust for those environmental effects in order to accurately estimate the genetic merits of sires and cows. Index selection is a simple extension of the ordinary least squares under the assumption that the fixed effects are assumed known without error. The mixed‐model equations (MME) of Henderson provide a simpler alternative to the generalized least squares procedure, which is computationally difficult to apply to large data sets. Solution to the MME yields the best linear unbiased estimator of the fixed effects and the best linear unbiased predictor (BLUP) of the random effects. In an animal breeding situation, the random effects such as sire or animal represent the animal's estimated breeding value, which provides a basis for selection decision. The BLUP procedure under sire model assumes random mating between sires and dams. The genetic evaluation procedure has progressed a long way from the dam‐daughter comparison method to animal model, from single trait to multiple trait analysis, and from lactational to test‐day model, to improve accuracy of evaluations. Multiple‐trait evaluation appears desirable because it takes into account the genetic and environmental variance‐covariance of all traits evaluated. For these reasons, multiple‐trait evaluation would reduce bias from selection and achieve a better accuracy of prediction as compared to single‐trait evaluation. The number of traits included in multiple‐trait evaluation should depend upon the breeding goal. Recent advances in molecular and reproductive technologies have created great potential for quantitative geneticists concerning genetic dissection of quantitative traits, and marker‐assisted genetic evaluation and selection.     

Accuracy of genome-wide evaluation for disease resistance in aquaculture breeding programs

Current aquaculture breeding programs aimed at improving resistance to diseases are based on challenge tests, where performance is recorded on sibs of candidates to selection, and on selection between families. Genome-wide evaluation (GWE) of breeding values offers new opportunities for using variation within families when dealing with such traits. However, up-to-date studies on GWE in aquaculture programs have only considered continuous traits. The objectives of this study were to extend GWE methodology, in particular the Bayes B method, to analyze dichotomous traits such as resistance to disease, and to quantify, through computer simulation, the accuracy of GWE for disease resistance in aquaculture sib-based programs, using the methodology developed. Two heritabilities (0.1 and 0.3) and 2 disease prevalences (0.1 and 0.5) were assumed in the simulations. We followed the threshold liability model, which assumes that there is an underlying variable (liability) with a continuous distribution and assumed a BayesB model for the liabilities. It was shown that the threshold liability model used fits very well with the BayesB model of GWE. The advantage of using the threshold model was clear when dealing with disease resistance dichotomous phenotypes, particularly under the conditions where linear models are less appropriate (low heritability and disease prevalence). In the testing set (where individuals are genotyped but not measured), the increase in accuracy for the simulated schemes when using the threshold model ranged from 4 (for heritability equal to 0.3 and prevalence equal to 0.5) to 16% (for heritability and prevalence equal to 0.1) when compared with the linear model.     

利用广义线性方法定位家畜抗性等级性状的QTL

在广义线性模型的框架内模拟研究了家畜抗性等级性状的QTL定位方法，QTL参数的估计采用最大似然方法，比较了阈模型方法与一般线性方法的QTL定位效率，并对影响等级性状QTL定位效率的主要因素（QTL效应、性状的遗传力）进行了模拟研究。试验为多个家系的女儿设计，资源群体大小为500头。结果表明：在QTL位置参数估计及检验功效方面，阈模型方法具有一定的优势，对抗性等级性状QTL定位的功效也高于线性方法。另外，性状遗传力和QTL效应的大小对QTL定位的准确度也有直接的影响，随着性状遗传力QTL效应的增大，2种方法QTL定位的效率均有不同程度的提高。     

Genotype x environment interactions and genetic parameters for fecal egg count and production traits of Merino sheep

Breeding for host resistance to parasites has become an imperative in many sheep industries. Because of the widespread use of AI in sheep breeding schemes, it is important to understand how the performance of offspring from rams varies in different flock environments, both for resistance to parasites and key production traits. This study used both variance component and reaction norm models to investigate the level of genotype x environment interaction for fecal egg count (FEC) and important Merino production traits in a range of flock environments in Australia. These flocks were linked by the use of common rams in a sire-referencing scheme. Both linear and quadratic polynomial reaction norm models were used. The heritability of these traits and the genetic correlation between them and FEC also was investigated using the reaction norm model. A contemporary group (CG) was defined by a flock, year, age class, sex, and paddock combination. Each CG environment was characterized by the mean value of any given trait for that CG. The recorded data used in the study were analyzed in a standardized form. Standardization for each trait was achieved within a CG by subtracting the CG mean from each observation and dividing by the CG SD. The genotype x environment effect accounted for <0.05 of the phenotypic variance for all traits. In most traits the heritability varied little across environments. The exceptions were FEC, BW, and both greasy and clean fleece weights, which had a higher heritability at the lower end of the environmental range. Fecal egg count also had a higher heritability in high-FEC environments. Genetic correlations between FEC and several key production traits were similar in the flock environments studied. Quadratic polynomial models and models with a variable residual fitted the data better than linear models. The genotype x environment effect for FEC and the genetic correlations between FEC and production traits were effectively zero; thus, sheep breeding programs for increased parasite resistance can be run effectively by ignoring these factors. Some account should be taken of the high heritabilities of FEC and fleece and BW in different flock environments.     

Application of Bayesian causal inference and structural equation model to animal breeding

Optimized breeding goals and management practices for the improvement of target traits requires knowledge regarding any potential functional relationships between them. Fitting a structural equation model (SEM) allows for inferences about the magnitude of causal effects between traits to be made. In recent years, an adaptation of SEM was proposed in the context of quantitative genetics and mixed models. Several studies have since applied the SEM in the context of animal breeding. However, fitting the SEM requires choosing a causal structure with prior biological or temporal knowledge. The inductive causation (IC) algorithm can be used to recover an underlying causal structure from observed associations between traits. The results of the papers, which are introduced in this review, showed that using the IC algorithm to infer a causal structure is a helpful tool for detecting a causal structure without proper prior knowledge or with uncertain relationships between traits. The reports also presented that fitting the SEM could infer the effects of interventions, which are not given by correlations. Hence, information from the SEM provides more insights into and suggestions on breeding strategy than that from a multiple‐trait model, which is the conventional model used for multitrait analysis.     

比较机器学习等算法对肉鸡产蛋性状育种值估计的准确性

我国白羽肉鸡育种中,通过遗传途径提高产蛋数和控制合适的蛋重是培育优良品系的一个重要方面。为探索适合我国白羽肉鸡育种中的基因组选择模型,本研究以2 474只白羽肉鸡品系的产蛋性状为研究对象,主要分析了机器学习算法KAML、BLUP(包括：PBLUP、GBLUP、SSGBLUP)和Bayes(包括：Bayes A、Bayes B和Bayes Cπ)方法对产蛋数和蛋重性状的预测准确性,准确性以5倍交叉验证进行评估。利用系谱以及基因组信息估计了产蛋数和蛋重性状的遗传力和遗传相关。结果表明,产蛋数性状遗传力为0.061~0.16,属于低遗传力性状;蛋重遗传力为0.28~0.39,属于中等遗传力性状;产蛋数与蛋重是中等遗传负相关(-0.518~-0.184),不同阶段产蛋数之间是强的遗传正相关(0.736~0.998)。不同模型预测43周产蛋数和52周蛋重的育种值估计准确性结果表明,KAML方法对两者的预测准确性分别为0.115和0.266,与GBLUP方法(准确性分别为0.118和0.283)和SSGBLUP方法(准确性分别为0.136和0.259)的准确性差异显著,同时显著低于Bayes方法(准确性分别为0.230~0.239、0.336~0.340)的预测准确性, PBLUP方法预测准确性最低(准确性分别为0.095和0.246)。因此,在白羽肉鸡产蛋数和蛋重性状中应用Bayes方法将获得最高的育种值估计准确性。     

An alternative derivation method of mixed model equations from best linear unbiased prediction (BLUP) and restricted BLUP of breeding values not using maximum likelihood

Two mixed model equations (MME) for best linear unbiased prediction (BLUP) of breeding values and for restricted BLUP of breeding values were derived by maximum likelihood from the joint normal probability distribution of the observations and breeding values. As a result, MME is actually more general than maximum likelihood because we can prove that each set of solutions of MME are identical to BLUP and restricted BLUP of breeding values and then it does not depend on normality. In the present study, the author shows deriving directly each MME from BLUP and restricted BLUP equations for breeding values without assuming the joint normal distribution of the data and random effects. However, if we cannot assume the multivariate normal density distribution of the estimated aggregate breeding value and each breeding value for selected traits, the response to selection by restricted BLUP may deviate from the expected values.