Methods for Integrating Functional Data into Complex Disease Genetic Analyses

 描述（由申请人提供）：人类基因组计划的最新发展和不同类型的高通量技术的突破已经改变了研究人员如何通过跨学科研究来处理复杂疾病，收集疾病各个方面的数据。该应用程序的目的是开发有效的遗传学和计算方法，以整合遗传学，基因组学和流行病学数据，以了解遗传学和环境在复杂疾病中的相互作用，长期目标是制定个性化的策略来预防和治疗这些疾病。 全基因组关联研究已经确定了数千个性状相关的遗传变异，并为这些性状的遗传结构提供了有价值的见解。然而，迄今为止发现的大多数变异赋予相对较小的风险增量，并且只能解释一小部分遗传性，导致许多人质疑如何解释剩余的“缺失”遗传性。本申请从几个方面解决了这种“缺失”的遗传力：罕见变异关联分析，基因-环境相互作用，以及遗传力估计超出加性遗传效应。 因此，我们提出以下具体目标。目的1是开发将功能信息整合到罕见变异关联分析中的方法。为了实现这一目标，目标1包括开发组织特异性功能注释数据库，并从大型合作项目（如DNA元件百科全书和基因型组织表达）产生的公共数据中构建调控表达网络（eQTL）。还将研究稀有变体分析的理论性质， 考虑到基因组特征，如连锁不平衡和稀疏信号的强大测试。目的2是开发结合功能信息的罕见变异基因-环境相互作用（GxE）的方法。高效和通用的筛选策略也将被开发用于GxE的全基因组发现。尽管这一目标集中在GxE上，但这些方法也适用于基因-基因相互作用（GxG）。目标3是开发估计遗传性的方法，其中结合GxE和GxG，以了解遗传易感性与环境之间复杂的相互作用。拟议的工作是由一个大型结直肠癌联盟发起的，该联盟拥有超过40，000名来自特征良好的研究的参与者，其中包括详细的数据。环境风险因素和GWAS和全基因组测序数据。所开发的方法将应用于该联盟，以获得对结直肠癌的新见解，并证明该方法的可行性。由于这些方法适用于其他复杂的疾病和特征，因此将开发基于R的开源软件，并提交给综合R档案网络进行广泛传播。

 DESCRIPTION (provided by applicant): Recent developments in The Human Genome Project and breakthroughs in different types of high throughput technologies have changed how researchers approach complex diseases by moving toward cross- disciplinary studies, collecting data on all facets of disease. The objective of this application is to develop efficient statistica and computational approaches to integrating genetics, genomics and epidemiologic data for understanding the interplay of genetics and environment in complex diseases, with the long-term goal of devising personalized strategies to prevent and treat these diseases. Genome-wide association studies have identified thousands of trait associated genetic variants, and provided valuable insights into the genetic architecture of these traits. However, most variants identified so far confer relatively small increments in risk, and explain only a small proportion o heritability, leading many to question how the remaining 'missing' heritability can be explained. This application addresses this 'missing' heritability from several aspects: rare variant association analysis, gene-environment interaction, and heritability estimation beyond additive genetic effects. Accordingly, we propose the following specific aims. Aim 1 is to develop methods for integrating functional information into rare variants association analysis. To achieve this goal, Aim 1 includes developing databases of tissue-specific functional annotation and constructing regulatory expression networks (eQTL) from public data generated from large collaborative projects such as the Encyclopedia of DNA Elements and the Genotype Tissue Expression. The theoretical properties of the rare variants analysis will also be studied to devise powerful tests in consideration of genomic features such as linkage disequilibrium and sparse signals. Aim 2 is to develop methods for rare variants gene-environment interaction (GxE) that incorporates functional information. Efficient and versatile screening strategies will also be developed for genome-wide discovery of GxE. Even though this aim is focused on GxE, the methods are also applicable to gene-gene interaction (GxG). Aim 3 is to develop methods for estimating heritability that incorporates GxE and GxG to understand the complex interplay between genetic susceptibility and environment The proposed work is motivated by a large consortium on colorectal cancer, which has over 40,000 participants from well-characterized studies with detailed data on both environmental risk factors and GWAS and whole genome sequencing data. The developed methods will be applied to the consortium to gain new insights in colorectal cancer and demonstrate the feasibility of the methods. Since the methods are applicable to other complex diseases and traits, R-based open source software will be developed and submitted to the Comprehensive R Archive Network for broad dissemination.