Linking Direct to Consumer Genomics and Electronic Health Records to Accelerate Science

摘要 在复杂性状和医学疾病的遗传研究中需要非常大的样本量，并收集 这样的样品既昂贵又耗时。为了加速复杂的基因发现和翻译，它是 对于快速高效地利用现有数据至关重要。我们提案的目标是实施一项新颖的 可扩展的在线/移动招聘模式，可利用通过以下方式生成的大量基因组数据 直接面向消费者的测试，并将这些数据链接到包含电子健康的研究数据仓库 记录BJC Healthcare系统中的(EHR)信息。我们将在三个方面实现目标：1) 最终确定用于招募消费者基因组学客户的在线/移动工具，这些客户也是北京赛马会的患者 医疗保健系统。我们设计了一种在线/移动方法，使参与者能够为我们提供 知情同意，他们个人水平的23andMe全基因组基因型，以及需要的关键信息 将他们的基因组数据与他们基于研究的电子病历数据联系起来。2)招募23andMe客户队列，这些客户 此外，BJC医疗系统的患者还可以建立一个数据库，将他们的基因型别和EHR联系起来。我们的目标是 到第二年招募10,000名参与者，根据23andMe抽样基地的规模，北京赛马会的规模 医疗保健人群，以及我们已经确定的合格参与者。3)在我们之前的基础上 工作中，我们将评估多基因风险评分(PR)在预测物质使用表型方面的表现 健康后果，如心脏病、慢性阻塞性肺病、肺癌和相关特征。这些分析将是一个证据 原则上来检验这个数据库的价值。我们的电子招聘策略可以很容易地适应 招聘各种研究，我们将分享我们的招聘工具和详细资料，我们的资料库 基础设施与研究机构通过网络开发，作为NIH临床和翻译的一部分 科学奖。 至关重要的是，今后我们要更有效地利用研究资源，重点是建立 可在未来重复使用的数据资源，而不是招募参与者回答 单一的研究问题。通过整合在线招聘、现有的基因组数据和基于研究的 电子健康档案以一种新颖的方式，我们的招聘模式可以迅速地达到大的队列 以具有成本效益的方式进行复杂疾病遗传学研究所需的。这种方法将使 研究人员有效地快速有效地瞄准新兴的健康趋势和研究需求，而这 这种方法是对我们所有人等正在进行的大规模努力的补充。

Abstract Very large sample sizes are needed in genetic studies of complex traits and medical illnesses, and collecting such samples is expensive and time intensive. To accelerate complex genetic discovery and translation, it is critical to harness existing data quickly and efficiently. The goal of our proposal is to implement a novel and scalable online/mobile recruitment paradigm to harness the vast amount of genomic data generated through direct-to-consumer testing and to link these data to a research data warehouse containing electronic health record (EHR) information from the BJC Healthcare System. We will accomplish our goal in three aims: 1) Finalize online/mobile tools for recruiting consumer genomics customers who also are patients from the BJC Healthcare system. We have designed an online/mobile approach that enables participants to provide us with informed consent, their individual-level 23andMe genome-wide genotypes, and the key information required to link their genomic data to their research based EHR data. 2) Recruit a cohort of 23andMe customers who are also patients from the BJC Healthcare system to build a database linking their genotypes and EHR. We aim to recruit 10,000 participants by year 2, based on the size of the 23andMe sampling base, the size of the BJC Healthcare population, and the eligible participants we have already identified. 3) Building on our previous work, we will evaluate the performance of polygenic risk scores (PRS) for substance use phenotypes to predict health outcomes such as heart disease, COPD, lung cancer, and related traits. These analyses will be a proof of principle to test of the value of this database. Our electronic recruitment strategy can be easily adapted to recruitment for a variety of studies, and we will share our recruitment tools and details of our repository infrastructure with research institutions through the network developed as part of NIH Clinical and Translational Sciences Awards. It is critical that moving forward we make more efficient use of research resources, focusing on establishing data resources that can be used repeatedly in the future as opposed to recruiting participants to answer a single research question. By integrating online recruitment, existing genomic data, and research based electronic health records in a novel approach, our recruitment paradigm can rapidly attain the large cohorts needed for complex disease genetics research in a cost-effective manner. This approach will enable researchers to effectively target emerging health trends and research needs quickly and efficiently, and this approach complements the large-scale efforts underway such as All of Us.