US-UK BBSRC-NIFA Collab-Reassembly of cattle immune gene clusters for quantitative analysis

自从大约10,000年前家畜首次被驯化以来，它们一直被有选择地饲养，以获得理想的特征。使用可测量的性状和动物系谱的传统遗传改良非常成功，特别是在增加重要牲畜物种的产量方面。今天的结果是出现了大量不同的品种，特别适合不同的环境或生产类型，例如乳牛和肉牛。然而，在大多数牲畜种群中，有相当数量的变异在选择性育种过程中从未被利用过。随着全球粮食需求的快速增长，畜牧业改良的需求也在不断升级。由于这种需求和最近的技术进步，现在可以根据动物的基因组序列来制定育种策略。在牛身上，这是通过在牛基因组中发现的近80万个单核苷酸多态(SNPs)的特征来实现的。快速识别与亲子关系、多产特征或品种组成相关的SNPs，而不是对每种动物的整个基因组进行测序，可以在动物生命的早期做出育种决定。没有表型数据的未知动物可以仅根据它们的SNP基因型进行评估，并计算它们的繁殖值。这种基因组选择的方法现在被牛养殖公司广泛使用。然而，就像任何年轻的技术一样，问题仍然存在。如果基因组的区域在个体之间非常不同和/或非常重复，就很难识别可以通过基因分型技术筛选的SNPs。哺乳动物基因组中有几个高度可变和重复的免疫基因复合体，它们在抗病和疫苗应答中起着重要作用。此外，这些区域已经进化出这种复杂性，至少在一定程度上是为了对抗快速演变的病原体。在牛身上，我们已经确定，目前的SNPs没有覆盖三个重要的大型免疫基因复合体，并且在很大程度上这些复合体没有在当前的基因组构建中组装。SNP用于基因分型的验证依赖于SNP所在区域的准确基因组组装；因此，这进一步加剧了问题。归根结底，目前的技术还不能识别与重要的免疫基因相关的遗传标记，这些基因可能会影响健康和抗病能力。猫科动物拥有一个天然的遗传多样性池，这些基因多样性是为了对抗快速进化的病原体而进化的，而这些病原体还不能用基因组学来选择。我们建议开发利用这种多样性的工具来改善牛的健康和抗病特性。在我们最初组装这些基因复合体的基础上，我们将在许多个体中组装这些基因组区域，以表征大的结构变异的程度。然后，来自30个个体的现有短全基因组序列读数将与这些较大的区域以及其他公开可用的序列数据集进行比对。通过以这些区域为目标，将有可能识别和验证合适的SNP，即使是那些频率较低的SNP，然后这些SNP将被纳入基因分型平台。然后，将通过对一群对牛结核病表现出不同抗病能力的牛进行基因分型来测试该工具的实用性。牛结核病是一种复杂的疾病，已知涉及基因成分，并受到我们在这项研究中所针对的基因复合体的影响。最终，我们设想这些标记可以被纳入当前和未来的基因分型技术，通过选择性育种来提高牛的抗病能力。

Since livestock were first domesticated approximately 10,000 years ago they have been selectively bred for desirable traits. Traditional genetic improvement using measurable traits and animal pedigrees has been very successful, particularly to increase production in important livestock species. The result today is a plethora of different breeds that are particularly suited for different environments or types of production, e.g. dairy and beef cattle. However within most livestock populations there is considerable amount of variation that has never been exploited during selective breeding. As the global demand for food increases rapidly, the demand for livestock improvement is escalating. As a consequence of this demand and recent advances in technology, it is now possible to inform breeding strategies based on the animal's genome sequence. In cattle this has been made possible through the characterisation of nearly 800,000 single nucleotide polymorphisms (SNPs) identified in cattle genomes. Rather than having to sequence the whole genome of each animal, rapid identification of SNPs that are associated with parentage, productive traits or breed composition allows for breeding decisions to be made earlier in an animal's life. Unknown animals with no phenotypic data can then be assessed solely on their SNP genotype and their breeding values calculated. This method of genomic selection is now widely used by cattle breeding companies.However, as with any young technology, problems remain. If regions of the genome are very variable between individuals and/or very repetitive it is difficult to identify SNPs that can be screened by the genotyping technology. There are several highly variable and repetitive immune gene complexes in mammalian genomes which have a fundamental role in disease resistance and responses to vaccines. Moreover these regions have evolved this complexity, at least in part, to combat rapidly evolving pathogens. In cattle, we have identified that the current SNPs do not cover three large and vital immune gene complexes, and to a large extent these complexes have not been assembled in the current genome builds. The validation of SNPs for use in genotyping relies upon an accurate genome assembly over the region the SNP is located; therefore this further compounds the problem. Ultimately the current technology is not yet able to type for genetic markers associated with important immune genes that are likely to influence health and disease resistance traits.Cattle possess a pool of natural genetic diversity that has evolved to counter rapidly evolving pathogens that cannot yet be selected for using genomics. We propose to develop the tools to utilise this diversity to improve health and disease resistance traits in cattle. Building on our initial assemblies of these gene complexes, we will assemble these genomic regions in many individuals to characterise the extent a large structural variation. Existing short whole genome sequence reads from > 30 individuals will then be aligned to these larger regions, alongside other publically available sequence datasets. By targeting these regions, it will be possible to identify and validate suitable SNPs, even those at low frequency, which will then be incorporated into a genotyping platform. The utility of this tool will then be tested by genotyping a herd of cattle that display differential disease resistance to bovine tuberculosis, a complex disease that is known to involve a genetic component and is influenced by the gene complexes we are targeting in this study. Ultimately we envisage that these markers can then be incorporated into current and future genotyping technologies to improve disease resistance in cattle through selective breeding.