Fine-mapping causal tissues at disease-associated loci to infer disease subtypes

项目摘要(摘要) 全基因组关联研究(GWAS)在识别遗传位点方面已经非常成功 与复杂的疾病有关，但这种联系背后的原因组织和作用机制 一般情况下，每个基因座都是未知的。最近的研究表明，跨组织的基因表达可以作为 一种中间表型来填补这些缺失的基因和疾病之间的联系。一张准确的地图 从致病部位到致病组织和介导该致病部位的基因将加深我们对疾病部位的理解 并对疾病的组织相关机制提供了新的见解。然而，准确的 致病组织和致病基因的推断可能被遗传共同调节所掩盖 (对基因表达的不同影响之间的相关性)跨组织和基因。这样做的第一个目的是 建议开发一种新的统计方法来准确推断导致疾病的组织和基因 同时显式地模拟组织和基因之间的遗传协同调节。这样做的第二个目的是 建议是使用组织特异性多基因风险评分(PR)来推断与特定组织相关的疾病亚型 通过根据病因组织类型对疾病部位进行注释而生成的，使用 第一个目标。我希望组织特异性PR能揭示个体中组织特异性遗传风险的模式 与疾病的不同亚型相对应。如果实现，这项工作将提炼我们对因果关系的认识 为推断疾病亚型提供了一种新的范例 在疾病关联研究中的每一个个体。更广泛地说，这项工作有可能提炼出 药物靶向候选疾病基因并促进针对以下疾病的患者护理和治疗的发展 疾病亚型。此外，我提出了一项深入的培训计划，利用研究社区的力量 在整个哈佛为我提供必要的培训和技能，使我的职业生涯从一个岗位- 从博士生到独立调查员。

Project Summary (Abstract) Genome-wide association studies (GWAS) have been extremely successful in identifying genetic loci associated with complex diseases, but the causal tissue and functional mechanism underlying the association at each locus is generally unknown. Recent work has shown gene expression across tissues can be used as an intermediate phenotype to fill in these missing links between genotype and disease. An accurate mapping from disease locus to the causal tissue and gene mediating that disease locus would further our understanding of disease biology and provide novel insights into tissue-linked mechanisms of disease. However, accurate inference of the causal tissue and gene driving a disease locus can be obscured by genetic co-regulation (correlation among variant effects on gene expression) across tissues as well as genes. The first aim of this proposal is to develop a new statistical method to accurately infer the causal tissue and gene driving a disease locus, while explicitly modeling genetic co-regulation between tissues and genes. The second aim of this proposal is to infer disease subtypes linked to specific tissues using tissue-specific polygenic risk scores (PRS) generated by annotating disease loci according to their causal tissue type, utilizing the method developed in the first aim. I expect tissue-specific PRS to uncover patterns of tissue-specific genetic risk across individuals that correspond to different subtypes of the disease. If achieved, this work will refine our knowledge of causal tissues and genes underlying disease loci, as well as provide a new paradigm for inferring the disease subtype of each individual in a disease association study. More generally, this work has the potential to refine the set of drug target candidate disease genes and facilitate the development of patient care and treatments specific to disease subtypes. In addition, I propose an in-depth training plan that leverages the research community throughout Harvard to provide me with the training and skills necessary to advance my career from a post- doctoral fellow to an independent investigator.