DATA AND SPECIMEN HUB (DASH)

数据和样本中心 (DASH)

• 批准号: 10517101
• 负责人: DEBRA VAVRECK
• 金额: $ 341.54万
• 依托单位: BOOZ ALLEN HAMILTON
• 依托单位国家: 美国
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-12-01 至 2021-11-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10517101
• 关键词: AIDS/HIV problem; Acquired Immunodeficiency Syndrome; Address; Agreement; Archives; Catalogs; Childhood; Clinical Trials; Communicable Diseases; Communities; Data; Data Set; Deposition; Ensure; FAIR principles; Faculty; Future; HIV; Individual; Life; Link; Meta-Analysis; Metadata; Morphologic artifacts; National Institute of Child Health and Human Development; Policies; Process; Publications; Publishing; Research; Research Design; Research Personnel; Research Training; Semantics; Specimen; Statistical Methods; Structure; System; Testing; United States National Institutes of Health; archive data; archived data; career development; data hub; data lake; data repository; data sharing; epidemiology study; prediction algorithm; public repository; repository; secondary analysis; web site

The benefits of data sharing are well-established. A growing body of evidence shows that when data are made accessible, they are frequently reused by other researchers to the benefit of all. For example, two recent studies found that between 15 and 20 percent of datasets archived in a public repository are reused (Navar et al., 2016; Piwowar et al., 2013) and that publications where the data were made available through a public repository received more citations than similar studies where the data were not made available (Piwowar et al., 2013; Piwowar et al., 2007). A recent analysis of the reuse of clinical trial data made available through a centralized data repository highlights that data sharing can benefit the scientific community in a variety of ways including testing new research questions or hypotheses, testing statistical methods, performing meta-analyses, and developing predictive algorithms (Coady et al., 2017). Towards these ends the Maternal and Pediatric Infectious Disease Branch at NICHD has made one of its priorities the archiving of data, and specimens when available, from HIV/AIDS studies supported by NICHD, in the NICHD DASH system (https://dash.nichd.nih.gov/). Launched in 2015, DASH addresses many of the challenges that have historically limited data sharing. DASH provides a platform to share data from completed studies and it is ready to accept data as soon as the data are ready for dissemination. In doing so it provides investigators with another mechanism to meet the data sharing requirements laid out in the NIH Data Sharing Policy (Final NIH Statement on Sharing Research Data, 2003). DASH acts as a data lake, accepting data in a variety of structures and formats. In accordance with FAIR principles (Wilkinson et al., 2016), all studies archived in DASH are annotated with metadata tags to maximize their findability. Users can locate studies and study artifacts archived in DASH using a keyword semantic search or through the study catalog which allows users to browse by topic, study type or life stage. Each study archived in DASH contains a study overview page that includes links to related studies, such as precursor studies or sub-analyses, with available datasets archived in DASH or another external repository. DASH also ensures data are accessible by making the process for obtaining data clear, fair and transparent. Towards this end, DASH publishes the NICHD DASH Data Use Agreement on its website along with a tutorial on the process of requesting data from DASH. To date, exactly half of the 42 studies in DASH are HIV-related. These range from epidemiologic studies (NISDI), studies from the ATN network, and data from individual trials (HPTN 040). In addition, numerous planned submissions in the future will be from other HIV-related studies (ATN, PHACS, PROMOTE) so that approximately 50% of the activity of DASH should continue to be HIV-related for the foreseeable future. The studies in DASH already and planned for deposit in the future all cover the high priorities for AIDS research as defined by OAR. Importantly, because secondary analyses of such studies are often a major part of career development for trainees and junior faculty, this effort also addresses OAR’s research training priority.

数据共享的好处是众所周知的。越来越多的证据表明，当数据可以访问时，它们经常被其他研究人员重新使用，以造福所有人。例如，最近的两项研究发现，在公共存储库中存档的数据集有15%到20%被重用 （Navar等人，2016; Piwowar等人，2013年），并且通过公共存储库提供数据的出版物比没有提供数据的类似研究获得了更多的引用（Piwowar等人，2013; Piwowar等人，2007年）。最近对通过集中式数据存储库提供的临床试验数据的重复使用的分析强调，数据共享可以以各种方式使科学界受益，包括测试新的研究问题或假设、测试统计方法、执行荟萃分析和开发预测算法（Coady等人，2017年）。 为此，NICHD的妇幼传染病分支将其优先事项之一定为将NICHD支持的艾滋病毒/艾滋病研究的数据和样本（如有）存档到NICHD DASH系统（https：//dash.nichd.nih.gov/）。DASH于2015年推出，解决了历史上限制数据共享的许多挑战。DASH提供了一个共享已完成研究数据的平台，一旦数据准备好发布，它就准备好接受数据。在这样做的过程中，它为研究者提供了另一种机制，以满足NIH数据共享政策（最终NIH共享研究数据声明，2003年）中规定的数据共享要求。DASH充当数据湖，接受各种结构和格式的数据。 根据FAIR原则（威尔金森等人，2016），DASH中存档的所有研究都使用元数据标签进行注释，以最大限度地提高其可查找性。用户可以使用关键字语义搜索或通过研究目录（允许用户按主题、研究类型或生活阶段浏览）来定位DASH中存档的研究和研究工件。DASH中存档的每个研究都包含一个研究概述页面，其中包括相关研究的链接，例如前驱研究或子分析，以及DASH或其他外部存储库中存档的可用数据集。 DASH还通过使获取数据的过程清晰、公平和透明来确保数据的可访问性。为此，DASH在其网站上发布了NICHD DASH数据使用协议，沿着了有关从DASH请求数据的过程的教程。 到目前为止，DASH的42项研究中有一半与艾滋病毒有关。这些数据包括流行病学研究（NISDI）、ATN网络研究和个体试验数据（HPTN 040）。此外，许多计划在未来提交将来自其他艾滋病毒相关的研究（ATN，PHACS，PROMOTE），使大约50%的DASH活动应继续在可预见的未来与艾滋病毒相关。DASH中已经进行的和计划在未来存款的研究都涵盖了OAR定义的艾滋病研究的高度优先事项。重要的是，因为这些研究的二次分析往往是学员和初级教师职业发展的重要组成部分，这项工作也解决了OAR的研究培训优先事项。