Cloud-Based Biomedical Data Storage and Analysis: Implications for Trustworthy Governance

项目总结 存储、共享和分析个体水平的基因组、环境和连锁的表型和/或 健康结果数据对精确医学研究提出了深刻的技术和后勤挑战。 因此，新的基于云的计算和存储平台正在开发，以支持便捷的数据 处理和基于云的分析共用。然而，对负责任的数据治理的影响 这些机制目前还没有得到检验。为促进负责任和值得信赖的数据治理(即 关于如何存储、访问和使用生物医学数据的决策，以及 向研究参与者传达)，我们将对政策进行深入的定性分析， 与三个新兴的基于云的精准医疗平台相关的实践和程序： 生物数据催化剂或BDCatalyst(NHLBI)，分析、可视化和信息学实验室-空间或砧座 (NHGRI)和研究中心(我们所有人的研究计划)。这次探索性的直接目标是 调查将检查基因组和关联临床数据的控制和管理如何存储在 这些平台不同于早期的数据共享努力。我们还将探索哪些研究利益相关者， 包括平台开发人员、调查人员(数据贡献者和数据使用者)、机构官员和 资助者认为这是与新方法相关的最相关的治理权衡。特别是， 我们将就保护参与者数据的机制征求意见，确保研究用途 与知情同意保持一致，并向感兴趣的参与者提供经过充分验证的结果。草稿 基于这些观察结果的建议将与研究界分享，并构成基础 对于后续研究，将介绍这些新平台及其支持的研究实践 不同的研究参与者提供反馈和批判性反思。为了达到这些研究目标，我们将 追求以下目标：(1)描述当前存储、共享和分析 由新兴的基于云的分析平台支持的大规模基因组和关联数据；(2)探索 利益相关者对当前和拟议的基于基因组云的分析治理方法的看法 平台；以及(3)为基因组和关联的可信治理提出和审查建议 云平台背景下的环境和表型数据。拟议的调查将 生成有关大规模存储、访问和预期用途的新颖、及时和详细的信息 基因组和关联的临床数据保存在新兴的基于云的数据存储和分析平台中。这些 数据将为确定值得信赖的数据治理的最佳实践提供可靠的基础， 为关键的精确医学研究工具提供更高的透明度。

PROJECT SUMMARY The storage, sharing, and analysis of individual-level genomic, environmental, and linked phenotypic and/or health outcome data poses profound technical and logistical challenges for precision medicine research. Accordingly, new cloud-based computing and storage platforms are being developed to support facile data processing and cloud-based analysis commons. However, the implications for responsible data governance of these mechanisms are currently unexamined. To promote responsible and trustworthy data governance (i.e. decision-making about how biomedical data are stored, accessed, and used by researchers, as well as communicated to research participants), we will conduct an in-depth qualitative analysis of the policies, practices, and procedures associated with three emerging cloud-based precision medicine platforms: the BioData Catalyst or BDCatalyst (NHLBI), the Analysis, Visualization, and Informatics Lab-space or AnVIL (NHGRI), and the Research Hub (All of Us Research Program). The immediate goal of this exploratory investigation will be to examine how the control and management of genomic and linked clinical data stored on these platforms differs from earlier data sharing efforts. We will also explore what research stakeholders, including platform developers; investigators (data contributors as well as data users); institutional officials; and funders regard as the most relevant governance tradeoffs associated with the new approaches. In particular, we will solicit views on mechanisms employed to protect participant data, ensure that research uses are aligned with informed consent, and make well-validated results available to interested participants. Draft recommendations based in these observations will be shared with the research community and form the basis for subsequent research, which will introduce these new platforms and the research practices they enable to diverse research participants for feedback and critical reflection. To achieve these research objectives, we will pursue the following Aims: (1) Characterize current approaches to the storage, sharing, and analysis of largescale genomic and linked data enabled by emerging cloud-based analysis platforms; (2) Explore stakeholder views on current and proposed approaches to the governance of genomic cloud-based analysis platforms; and (3) Propose and vet recommendations for the trustworthy governance of genomic and linked environmental and phenotypic data in the context of cloud-based platforms. The proposed investigation will generate novel, timely, and detailed information about the storage, access, and intended use of large-scale genomic and linked clinical data held in emerging cloud-based data storage and analysis platforms. These data will provide a robust basis from which to identify best practices for trustworthy data governance and provide enhanced transparency about a key precision medicine research tool.