Wolc A, Arango J, Jankowski T, Dunn I, Settar P, Fulton J E, O'Sullivan N P, Preisinger R, Fernando R L, Garrick D J, Dekkers J C M
Department of Animal Science, Iowa State University, Ames, IA, USA; Hy-Line International, Dallas Center, IA, USA.
J Anim Breed Genet. 2014 Jun;131(3):173-82. doi: 10.1111/jbg.12086. Epub 2014 Mar 17.
Discovery of genes with large effects on economically important traits has for many years been of interest to breeders. The development of SNP panels which cover the whole genome with high density and, more importantly, that can be genotyped on large numbers of individuals at relatively low cost, has opened new opportunities for genome-wide association studies (GWAS). The objective of this study was to find genomic regions associated with egg production and quality traits in layers using analysis methods developed for the purpose of whole genome prediction. Genotypes on over 4500 birds and phenotypes on over 13,000 hens from eight generations of a brown egg layer line were used. Birds were genotyped with a custom 42K Illumina SNP chip. Recorded traits included two egg production and 11 egg quality traits (puncture score, albumen height, yolk weight and shell colour) at early and late stages of production, as well as body weight and age at first egg. Egg weight was previously analysed by Wolc et al. (2012). The Bayesian whole genome prediction model--BayesB (Meuwissen et al. 2001) was used to locate 1 Mb regions that were most strongly associated with each trait. The posterior probability of a 1 Mb window contributing to genetic variation was used as the criterion for suggesting the presence of a quantitative trait locus (QTL) in that window. Depending upon the trait, from 1 to 7 significant (posterior probability >0.9) 1 Mb regions were found. The largest QTL, a region explaining 32% of genetic variance, was found on chr4 at 78 Mb for body weight but had pleiotropic effects on other traits. For the other traits, the largest effects were much smaller, explaining <7% of genetic variance, with regions on chromosomes 2, 12 and 17 explaining above 5% of genetic variance for albumen height, shell colour and egg production, respectively. In total, 45 of 1043 1 Mb windows were estimated to have a non-zero effect with posterior probability > 0.9 for one or more traits.
多年来，育种者一直对发现对经济重要性状有重大影响的基因很感兴趣。高密度覆盖全基因组且更重要的是能以相对低成本对大量个体进行基因分型的SNP芯片的发展，为全基因组关联研究（GWAS）带来了新机遇。本研究的目的是使用为全基因组预测而开发的分析方法，找到与蛋鸡产蛋量和蛋品质性状相关的基因组区域。使用了来自一个褐壳蛋鸡品系八代的4500多只鸡的基因型和13000多只母鸡的表型数据。用定制的42K Illumina SNP芯片对鸡进行基因分型。记录的性状包括产蛋早期和晚期的两个产蛋量性状以及11个蛋品质性状（穿刺评分、蛋白高度、蛋黄重量和蛋壳颜色），还有体重和开产日龄。蛋重此前已由Wolc等人（2012年）进行过分析。采用贝叶斯全基因组预测模型——BayesB（Meuwissen等人，2001年）来定位与每个性状关联最紧密的1兆碱基区域。将1兆碱基窗口对遗传变异有贡献的后验概率用作该窗口中存在数量性状基因座（QTL）的判断标准。根据性状不同，发现了1至7个显著（后验概率>0.9）的1兆碱基区域。最大的QTL位于4号染色体78兆碱基处，解释了体重32%的遗传变异，但对其他性状有多效性影响。对于其他性状，最大效应要小得多，解释的遗传变异<7%，2号、12号和17号染色体上的区域分别解释了蛋白高度、蛋壳颜色和产蛋量超过5%的遗传变异。总共，1043个1兆碱基窗口中有45个被估计对一个或多个性状有非零效应，后验概率>0.9。
