Crum Tamar E, Schnabel Robert D, Decker Jared E, Taylor Jeremy F
Division of Animal Sciences, University of Missouri, Columbia, MO, United States.
Informatics Institute, University of Missouri, Columbia, MO, United States.
Front Genet. 2021 Oct 18;12:758394. doi: 10.3389/fgene.2021.758394. eCollection 2021.
Development of the American Breeds of beef cattle began in the 1920s as breeders and U. S. Experiment Station researchers began to create × hybrids using Brahman as the source. By 1954, U.S. Breed Associations had been formed for Brangus (5/8 Angus × 3/8 Brahman), Beefmaster (½ Brahman × ¼ Shorthorn × ¼ Hereford), and Santa Gertrudis (5/8 Shorthorn × 3/8 Brahman). While these breeds were developed using mating designs expected to create base generation animals with the required genome contributions from progenitor breeds, each association has now registered advanced generation animals in which selection or drift may have caused the realized genome compositions to differ from initial expected proportions. The availability of high-density SNP genotypes for 9,161 Brangus, 3,762 Beefmaster, and 1,942 Santa Gertrudis animals allowed us to compare the realized genomic architectures of breed members to the base generation expectations. We used RFMix to estimate local ancestry and identify genomic regions in which the proportion of Brahman ancestry differed significantly from expectations. For all three breeds, lower than expected levels of Brahman composition were found genome-wide, particularly in early-generation animals where we demonstrate that selection on beef production traits was likely responsible for the taurine enrichment. Using a proxy for generation number, we also contrasted the genomes of early- and advanced-generation animals and found that the indicine composition of the genome has increased with generation number likely due to selection on adaptive traits. Many of the most-highly differentiated genomic regions were breed specific, suggesting that differences in breeding objectives and selection intensities exist between the breeds. Global ancestry estimation is commonly performed in admixed animals to control for stratification in association studies. However, local ancestry estimation provides the opportunity to investigate the evolution of specific chromosomal segments and estimate haplotype effects on trait variation in admixed individuals. Investigating the genomic architecture of the American Breeds not only allows the estimation of indicine and taurine genome proportions genome-wide, but also the locations within the genome where either taurine or indicine alleles confer a selective advantage.
美国肉牛品种的培育始于20世纪20年代，当时育种者和美国实验站的研究人员开始以婆罗门牛为亲本培育杂交品种。到1954年，美国已经成立了勃莱牛（5/8安格斯牛×3/8婆罗门牛）、肉牛王（1/2婆罗门牛×1/4短角牛×1/4赫里福德牛）和圣格特鲁迪斯牛（5/8短角牛×3/8婆罗门牛）的品种协会。虽然这些品种是通过预期能培育出具有亲本品种所需基因组贡献的基础代动物的交配设计培育出来的，但现在每个协会都已登记了高级代动物，在这些动物中，选择或漂变可能导致实际的基因组组成与最初预期的比例不同。9161头勃莱牛、3762头肉牛王和1942头圣格特鲁迪斯牛的高密度单核苷酸多态性（SNP）基因型数据，使我们能够将品种成员的实际基因组结构与基础代预期进行比较。我们使用RFMix来估计局部祖先，并识别婆罗门牛祖先比例与预期有显著差异的基因组区域。对于所有这三个品种，在全基因组范围内都发现婆罗门牛成分低于预期水平，特别是在早期代动物中，我们证明对牛肉生产性状的选择可能是导致牛磺酸富集的原因。使用世代数的代理变量，我们还对比了早期代和高级代动物的基因组，发现基因组的印度牛成分可能由于对适应性性状的选择而随着世代数增加。许多高度分化的基因组区域是品种特异性的，这表明不同品种之间的育种目标和选择强度存在差异。在混合血统动物中通常进行全基因组祖先估计，以控制关联研究中的分层现象。然而，局部祖先估计提供了研究特定染色体片段进化以及估计混合个体中单体型对性状变异影响的机会。研究美国肉牛品种的基因组结构不仅能够在全基因组范围内估计印度牛和普通牛的基因组比例，还能确定基因组中普通牛或印度牛等位基因具有选择优势的位置。
