Structurally complex genome loci in human populations and human phenotypes

摘要/摘要 结构复杂基因座（SCL）是基因组动态的热点，其与人类表型的关系 变化未知。 SCL 具有多个重复 DNA 序列片段，其中可以包含或侧接 基因、外显子或调控元件；这些重复的序列相互重组以产生新的 等位基因通过非等位同源重组和基因转换，创造出许多功能不同的 具有不同基因剂量和/或蛋白质结构的等位基因。 人类遗传学尚不知道大多数 SCL 中存在的等位基因，也不知道它们与人类的关系 表型变异。 SCL 的遗传变异往往由许多等位基因引起，难以组装，并且 与附近的 SNP 和 SNP 单倍型有复杂的关系。然而，SCL 提供了一个真正的联系机会 对基因剂量或蛋白质结构域具有可解释影响的功能等位基因等位基因系列的表型 结构。 在这项工作中，我们将开发方法来确定整个基因组基因座上的 SCL 如何由等位基因组成 系列并与多种人类表型相关。为此，我们将结合多种形式的数据 基因组分析 – 明确的长读数据（n ~102 且不断增长）、全基因组和全外显子组序列 数据 (104-105) 和 SNP 阵列数据 (105-107) 以及伴随表型数据。 在目标 1 中，我们将开发方法来揭示 SCL 的全谱变异。我们将 (a) 识别变量 DNA 特征以及这些特征在数千人中变化和共同分布的方式 不同的祖先，以及（b）找到解释这种人口规模的潜在等位基因和等位基因频率 变化。 在目标 2 中，我们将利用大量现有的 SNP 数据集实现强大的基因型-表型分析；我们 将通过创建 SCL 等位基因和周围 SNP 的参考单倍型大型面板来实现这一点，并且 改进将 SCL 等位基因归入 SNP 数据的方法。 在目标 3 中，我们将推进 SCL 的遗传关联分析和精细作图方法，并探索 SCL 对数量性状和疾病风险的功能影响。 我们渴望做出更多关于结构复杂基因座上的等位基因系列如何形成的发现 人类表型。

SUMMARY/ABSTRACT Structurally complex loci (SCLs) are hotspots of genome dynamism whose relationship to human phenotypic variation is unknown. SCLs have multiple segments of duplicated DNA sequence which can contain or flank genes, exons, or regulatory elements; these repeated sequences recombine with one another to generate new alleles by non-allelic homologous recombination and gene conversion, creating many functionally distinct alleles with different gene dosages and/or protein structures. Human genetics does not yet know the alleles that are present at most SCLs, nor their relationship to human phenotypic variation. Genetic variation at SCLs tends to arise from many alleles, to be hard to assemble, and to have complex relationships to nearby SNPs and SNP haplotypes. Yet SCLs provide a real opportunity to relate phenotypes to allelic series of functional alleles with interpretable effects on gene dosage or protein domain structure. In this work, we will develop ways to ascertain how SCLs at loci across the genome are comprised of allelic series and relate to a diverse set of human phenotypes. To do this, we will combine data from many forms of genome analysis – definitive long-read data (n ~102 and growing), whole-genome and whole-exome sequence data (104-105) and SNP array data (105-107) with companion phenotype data. In Aim 1, we will develop methods to reveal the full spectrum of variation at SCLs. We will (a) identify variable DNA features and the ways in which these features vary and co-distribute across thousands of people of diverse ancestries, and (b) find the underlying alleles and allele frequencies that explain this population-scale variation. In Aim 2, we will enable powerful genotype-phenotype analyses that leverage vast existing SNP data sets; we will do this by creating large panels of reference haplotypes of SCL alleles and surrounding SNPs, and advancing methods for imputing SCL alleles into SNP data. In Aim 3, we will advance approaches for genetic association analysis and fine-mapping at SCLs, and explore the functional consequences of SCLs on quantitative traits and disease risk. We aspire to make and enable many more discoveries about how allelic series at structurally complex loci shape human phenotypes.