Heritability of complex traits via IBD and IBS in related and unrelated individua

描述(由申请人提供)： 全基因组关联研究已经确定了数百个与常见疾病和其他特征相关的风险变异，但在大多数情况下，只解释了一小部分遗传遗传性。这种“缺失的遗传性”的来源是人类遗传学中的一个关键问题。解决这一问题的努力在很大程度上集中在寻找特定的疾病风险变异上，留下了大部分遗传性仍然无法解释。我们将通过研究通过全基因组范围或跨基因组分割的遗传片段遗传相同(IBD)或状态相同(IBS)共享的遗传内容如何与广泛的人类特征的表型相似性相对应，来系统地解构遗传遗传性。我们将使用IBD来表征所有风险变异体的集合，并使用IBS来表征由Gwas芯片进行基因分型或标记的风险变异体的子集。我们将探索这样一种观点，即在量化和理解遗传遗传力方面，“无关”个体之间的遗传相似性可以与近亲之间的相似性一样有用，或者更有用。根据基因组位置划分遗传力将使我们能够得出关于各种人类特征的遗传结构的结论，同时考虑到所有风险变量的集合和Gwas芯片捕获的风险变量的子集。我们将通过模拟来发展这些想法，并将它们应用于包含70,000个样本的真实Gwas数据。我们将继续发布实用的、公开可用的软件包，实现我们开发的方法。

DESCRIPTION (provided by applicant): Genome-wide association studies (GWAS) have identified hundreds of risk variants associated to common diseases and other traits, yet in most cases have explained only a small fraction of genetic heritability. The source of this "missing heritability" is a critical question in human genetics. Efforts to address this question have largely focused on the search for specific disease risk variants, leaving the bulk of heritability still unexplained. We will systematically deconstruct genetic heritability by investigating how genetic content shared either via segments inherited identical-by-descent (IBD) or genotyped alleles identical-by-state (IBS), either genome-wide or partitioned across the genome, corresponds to phenotypic similarity across a broad range of human traits. We will use IBD to characterize the set of all risk variants, and IBS to characterize the subset of risk variants genotyped or tagged by GWAS chips. We will explore the idea that genetic similarity between "unrelated" individuals can be as useful, or more useful, than similarity between close relatives in quantifying and understanding genetic heritability. Partitioning heritability by genomic location will enable us to draw conclusions about the genetic architecture of various human traits, considering both the set of all risk variants and the subset of risk variants captured by GWAS chips. We will develop these ideas via simulation and apply them to real GWAS data comprising 70,000 samples. We will continue to release practical, publicly available software packages implementing the methods that we develop.