Novel Statistical Methods for Gene-Environment Interactions in Complex Diseases

描述(由申请人提供)： 与单基因疾病不同，遗传和环境因素在控制冠心病、哮喘、癌症和精神疾病等复杂疾病的风险方面发挥着基本的和相互作用的作用。全基因组关联(GWA)扫描和更有针对性的研究面临的一个主要挑战是，难以检测嵌入相当大的统计和技术变异中的真实遗传信号，以及与检查大量标记相关的多样性。在这些问题和机遇的推动下，我们将开发一套互补的方法来解决基因环境(GxE)在多个水平上的相互作用，包括GWA研究、通过连锁的基因定位和候选基因方法。对于每个级别，我们将开发新的技术，以填补现有方法的空白，或开发和评估新的战略方法。这里提出的研究特别侧重于增强方法来识别GxE相互作用，以寻找控制风险的生物重要基因。我们建议解决以下具体目标：(1)开发和评估新的统计方法，以通过解决GxE交互作用的全基因组关联研究中的适当排序来确定基因的优先顺序；(2)开发和评估新的统计方法，以定位因果基因，作为连锁和精细定位研究的一部分，同时考虑GxE交互作用；(3)开发和评估新的统计方法，以在候选基因研究中识别环境变量与SNPs之间的高阶交互作用；(4)调整现有的统计方法，并开发新的统计方法，以解决环境和遗传测量的不精确和缺失；以及(5)开发和传播用于GxE分析的高效算法，并将这些方法应用于几个正在进行的复杂疾病的遗传学研究。我们的新统计方法将适应多样性，识别和纳入测量不确定性，利用基因组结构和来自相关连锁、精细定位和候选基因研究的丰富信息，并通过生物相关的统计模型进行结构推断。他们将有效利用现有信息，并在科学相关的框架内报告调查结果。我们拟议的研究将带来科学上的好处，其中许多涉及公共卫生问题，即改善控制复杂疾病风险的基因和环境因素之间相互作用的特征。特别是，量化GxE相互作用可以帮助识别高危人群进行有重点的干预和预防，并提高我们对特定疾病的病因机制的理解。(摘要结束)

DESCRIPTION (provided by applicant): Unlike monogenic diseases, both genetic and environmental factors play essential and interactive roles in controlling risk to complex diseases such as coronary heart disease, asthma, cancer and psychiatric disorders. A principal challenge facing genome-wide association (GWA) scans and more focused investigations is the difficulty in detecting true genetic signals embedded in considerable statistical and technical variation, and the multiplicity associated with examining a large number of markers. Motivated by these issues and opportunities, we will develop a suite of complementary methods to address gene environment (GxE) interactions at many levels, including GWA studies, gene localization via linkage, and candidate gene approaches. For each level we will develop new techniques that fill gaps in currently available methods or develop and evaluate new strategic approaches. The research proposed here specifically focuses on enhancing methods to identify GxE interactions in the search for biologically important genes controlling risk. We propose to address the following specific aims: (1) Develop and evaluate new statistical methods to prioritize genes through proper ranking in genome-wide association studies that address GxE interactions; (2) Develop and evaluate new statistical methods to localize causal genes as part of linkage and fine mapping studies while considering GxE interactions; (3) Develop and evaluate new statistical methods to identify higher order interactions between environmental variables and SNPs in candidate gene studies; (4) Adapt existing and develop new statistical methods to address imprecise and missing environmental and genetic measurements; and (5) Develop and disseminate efficient algorithms for GxE analyses, and apply these methods in several ongoing genetic studies of complex diseases. Our novel statistical approaches will accommodate multiplicity, identify and incorporate measurement uncertainty, take advantage of genomic structure and the wealth of information from related linkage, fine mapping, and candidate gene studies, and structure inferences through biologically relevant statistical models. They will make efficient use of available information and report findings in a scientifically relevant framework. Our proposed research will confer scientific benefits, many with public health implications, of improved characterization of the interaction between genes and environmental factors controlling risk to complex diseases. In particular, quantifying GxE interaction can help identify high risk groups for focused intervention and prevention, and improve our understanding of etiologic mechanisms for specific diseases. (End of Abstract)