A resource for the genetic analysis of complex traits

 描述(申请人提供)：每年数百万美国人被诊断出患有糖尿病、心脏病、阿尔茨海默氏症和各种形式的癌症。与环境变化相一致，这些疾病的风险由大量的、不同种类的遗传因素控制。通过表征那些增加风险的基因，生物医学界可以描述与人类健康和疾病有关的分子途径，最终使合理设计新的治疗方法成为可能。不幸的是，在确定遗传风险等位基因目录方面进展缓慢。尽管我们有能力在人类中进行大规模的基于人群的病例对照遗传关联研究，但任何给定的复杂疾病都只有一小部分致病部位是已知的。在过去的几年里，包括我们自己在内的几个小组一直在探索高级世代、多亲本作图面板的用途，用于常规、强大和高分辨率地剖析模式遗传系统(例如，小鼠合作杂交、大鼠NIH异源种群、果蝇合成种群资源或DSPR)中的复杂性状变异。模型系统展示了与人类相似的遗传、细胞、生理和行为过程，并为深入了解导致复杂性状变异的因素提供了一种补充途径。我们成功地开发了 DSPR在第一期赠款资助期间，并将其作为免费可用的社区资源，用于对苍蝇性状变异的遗传解剖。我们的项目提供了一个框架，可以将致病变异映射到较小的基因组间隔，并有能力绘制罕见的变异和含有多个致病等位基因的基因，这两种基因都是人类关联研究难以探究的。在这里，我们将继续扩展我们的工作，开发DSPR作为一套强大的使能资源，为果蝇生物医学社区。首先，我们将使用长时间读取的高通量测序来为创建DSPR的所有品系生成基因组组装。这些新颖的组装将使我们能够识别品系中的结构和拷贝数变异，补充我们现有的数据，以提供DSPR分离变异的完整目录。其次，我们将对大量品系的多个组织进行全基因组基因表达谱分析，使我们能够识别导致表达变异的基因。此外，我们将确定基因组中可能招募调控蛋白的区域。整合这些数据集将有助于检测致病因素， 被认为对性状变异有显著影响的调节变种。最后，在我们成功地绘制了DSPR中与生物医学相关的性状变异的潜在基因座图的基础上，我们将对测交设计中的两个性状进行超大规模的表型筛选(将每个DSPR系与一组不相关的菌株杂交)。这项工作将使我们能够详细说明所绘制的基因座在多大程度上具有依赖于绘制人群的遗传背景的影响。这些结果将对模型系统中映射实验的设计具有重要的指导意义。

 DESCRIPTION (provided by applicant): Each year millions of Americans are diagnosed with diseases such as diabetes, heart disease, Alzheimer's, and various forms of cancer. In concert with environmental variation, risk for these diseases is controlled by large, heterogeneous sets of genetic factors. By characterizing those genes that increase risk, the biomedical community can describe the molecular pathways involved in human health and disease, ultimately enabling the rational design of novel treatments. Unfortunately, progress towards identifying the catalog of genetic risk alleles has been slow. Despite our ability to carry out massive population-based case-control genetic association studies in humans, only a tiny fraction of causative sites are known for any given complex disease. Over the last few years several groups, including ourselves, have been exploring the utility of advanced generation, multiparental mapping panels for routine, powerful, and high-resolution dissection of complex trait variation in model genetic systems (e.g., the mouse Collaborative Cross, the rat NIH heterogeneous stock, the Drosophila Synthetic Population Resource, or DSPR). Model systems exhibit similar genetic, cellular, physiological, and behavioral processes to humans, and offer a complementary avenue for obtaining insight into the factors that underlie complex trait variation. We successfully developed the DSPR during the first period of grant funding, and provide this as a freely-available community resource for the genetic dissection of trait variation in flies. Our project delivered a framework that can map causative variants to small genomic intervals, and has the power to map rare variants, and genes that harbor multiple causative alleles, both of which human association studies struggle to interrogate. Here we will continue to extend our work developing the DSPR as a powerful set of enabling resources for the Drosophila biomedical community. First, we will employ long-read, high-throughput sequencing to generate genome assemblies for all lines founding the DSPR. These novel assemblies will allow us to identify structural, and copy number variants in the lines, complementing our existing data to provide the complete catalog of segregating variation in the DSPR. Second, we will carry out genomewide gene expression profiling on multiple tissues for a large set of lines, allowing us to identify genes contributing to expression variation. Additionally, we will identify regions of the genome likely t recruit regulatory proteins. Integrating these datasets will facilitate the detection of causative, regulatory variants, which are thought to contribute significantly to trait variation. Finally, buiding on our considerable success mapping loci underlying biomedically-relevant trait variation in the DSPR, we will carry out extremely large-scale phenotyping screens for two traits in a testcross design (crossing each DSPR line to a set of unrelated strains). This work will allow us to detail the degree to which mapped loci have effects that are dependent on the genetic background of the mapping population. These results will have important implications for the design of mapping experiments in model systems.