Enhancing the Pan-Neurotrauma Data Commons (PANORAUMA) to a complete open data science tool by FAIR APIs

项目摘要 神经创伤(脊髓和大脑损伤)在美国影响着超过250万人，每年 医疗和社会经济成本800亿美元的经济影响。尽管改善了患者管理， 在过去的几十年里，促进神经康复的可行选择有限。脊髓损伤(SCI)和 创伤性脑损伤(TBI)导致跨越不同数据源和多个 分析的尺度。此外，这些损伤通常发生在中枢神经系统内的不同部位， 分级的严重程度产生具有不同结局轨迹的不同类型的损伤。弄明白这一点 复杂性要求跨多个伤害严重性、类型和分析范围汇集数据，范围包括 分子、解剖、生理和行为水平。大规模的数据资源和大数据工具 提供帮助的潜力。通过在单个级别汇集和协调不同的数据，可以 使神经创伤数据“可查找、可访问、可互操作和可重复使用”(FAIL)。公平的神经创伤数据 可以利用现代数据工作流和分析，引导新发现并加速 翻译。此外，公平的数据可以为广泛采用人工智能(AI)和 机器学习(ML)，它是美国国立卫生研究院数据科学战略计划和AI/ML准备倡议的核心 比如Bridge2A1和AIM-Ahead。 研究人员和数据科学家可以使用公平的神经创伤数据来推动新发现并建立强大的 可再现性和翻译工具，如数据处理软件和新的分析工作流程 管道。泛神经创伤数据共享母公司项目的总体目标是建立一个泛神经创伤数据共享中心 神经创伤(PANORAUMA)数据共享基础设施。该奖项旨在提高效率、质量、 社区驱动的脊髓损伤开放数据共享(ODC-SCI)和创伤数据共享的可持续性 通过集中运营和管理来减少脑损伤(ODC-TBI)。这个的nosi(非OD-22-068) 附录提供了一个机会，可以“从 从软件工程的角度来看。补充剂对PANORAUMA的可持续性和扩张至关重要 包括研究数据科学家和研究软件开发人员，以响应 美国国立卫生研究院的数据科学战略计划指出：可访问、组织良好、安全和高效 可操作的数据资源对现代科学研究至关重要。对于这一补充，我们建议：1)开发 PANORAUMA的应用程序编程接口(API)可更好地支持 云化并优化数据管道的可重用性、互操作性和可持续性；2)整合SmartAPI 公平的标准，最大限度地提高API的公平性和文档记录；3)增强PANORAUMA-API 与最先进的开源数据科学编码语言R和Python接口；以及4)构建 开发人员和数据科学家之间的社区伙伴关系。

Project Summary Neurotrauma (trauma to the spinal cord and brain) affects over 2.5 million individuals in the US, with an annual economic impact of $80 billion in medical and socioeconomic costs. Despite improved patient management in the last decades, there are limited viable options to promote neurological recovery. Spinal cord injury (SCI) and traumatic brain injury (TBI) result in multifaceted syndromes spanning heterogeneous data sources and multiple scales of analysis. In addition, these injuries often occur at various sites within the central nervous system, with graded severities producing heterogeneous injuries with diverse outcome trajectories. Making sense of this complexity requires pooling data across multiple injury severities, types, and scales of analysis ranging from molecular, anatomical, physiological, and behavioral levels. Large-scale data resources and big-data tools have the potential to help. By pooling and harmonizing diverse data at the individual level, it becomes possible to make neurotrauma data “Findable, Accessible, Interoperable, and Reusable” (FAIR). FAIR neurotrauma data can be harnessed using modern data workflows and analytics, directing novel discovery and accelerating translation. Moreover, FAIR data can set the stage for widespread adoption of artificial intelligence (AI) and machine learning (ML), and it is at the core of NIH Strategic Plan for Data Science and AI/ML-readiness initiatives like Bridge2A1 and AIM-AHEAD. Researchers and data scientists can use FAIR neurotrauma data to drive novel discoveries and build robust reproducibility and translation tools, such as data processing software and new analytical workflows and pipelines. The overarching objective of the Pan-Neurotrauma data commons parent project is to build a Pan- Neurotrauma (PANORAUMA) data commons infrastructure. The award aims at improving the efficiency, quality, and sustainability of the community-driven Open Data Commons for Spinal Cord Injury (odc-sci) and Traumatic Brain Injury (odc-tbi) by centralizing their operations and governance. The NOSI (NOT-OD-22-068) for this supplement provides an opportunity for “improving the quality and sustainability of research software from a software engineering perspective.” The supplement is vital for PANORAUMA sustainability and the expansion of the community of users to include research data scientists and research software developers in response to NIH’s strategic plan for data science which states that “accessible, well-organized, secure, and efficiently operated data resources are critical to modern scientific inquiry.” For this supplement, we propose to: 1) develop the Application Programming Interface (API) of PANORAUMA to better support data science activities in the cloud and optimize reusability, interoperability, and sustainability of data pipelines; 2) incorporate the SmartAPI FAIR standards to maximize the API’s FAIRness and documentation; 3) enhance the PANORAUMA-API interface with state-of-the-art, open-source data science coding languages R and Python; and 4) build community partnerships between developers and data scientists.