HuBMAP HIVE Tools Component Supplement

联系PD/PI：Gehlenborg，Nils 项目摘要 我们建议贡献一套基于云的数据可视化、探索和分析工具， HuBMAP数据公平，并将在社区内广泛重新使用这一宝贵的资源。我们设想， 支持高级用户的批处理和交互式分析工具，以及基于任务特定UI的分析工具 对于没有深入的计算方法知识的最终用户。我们的目标是为用户提供 通过可视化界面与分析出处进行交互的能力。这些将基于 我们开发该技术是为了与工作流衍生的数据来源进行交互，以进行可重复的生物医学研究。 我们的可视化工具将是交互式的，并为数据探索和通信而设计。基于 我们的Vistories方法用于可再现的视觉分析，我们将在这两个不同的任务之间创建一个桥梁。 这也将使其能够与远程合作者分享观察和发现。理论分析和 可视化工具将嵌入到一个框架中，使我们能够整合和部署由 HIVE、HuBMAP和更广泛社区的成员。只要有可能，我们将加强现有的工具， 并将它们包装到基于容器的、可扩展的、与云无关的部署框架中。若要启用搜索和 探索将由HuBMAP生成的巨大数据空间，我们建议实现一组 这些工具能够系统地导航和检索组学和成像数据，而不是基于文本的搜索。我们将 通过类似于SATORI系统的接口，重点关注元数据和本体的使用 用于生物医学数据存储库的本体引导的可视化探索。最后，我们将支持合作 在HuBMAP和HIVE中，通过可视化分析工具监控整个项目进度 和数据质量，包括计划实验的时间表，提交的数据等。 第6页 项目概要/摘要

Contact PD/PI: Gehlenborg, Nils PROJECT SUMMARY We propose to contribute a suite of cloud-based data visualization, exploration, and analysis tools that will make HuBMAP data FAIR and will enable broad re-use of this valuable resource within the community. We envision to support both batch and interactive analysis tools for power users, as well as task-specific UI-based analysis tools for end users without deep knowledge of computational approaches. Our goal is to also provide users with the ability to interact with analysis provenance in through visual interfaces. These will be based on the approaches that we developed for interaction with workflow-derived data provenance for reproducible biomedical research. Our visualization tools will be interactive and designed for both data exploration and communication. Based on our Vistories approach for reproducible visual analysis, we will create a bridge between these two distinct tasks. This will also make it possible to share observations and discoveries with remote collaborators. Both analysis and visualization tools will be embedded in a framework that will enable us to incorporate and deploy tools created by members of the HIVE, HuBMAP, and the broader community. Whenever possible, we will enhance existing tools and wrap them into a container-based, scalable, and cloud-agnostic deployment framework. To enable search and exploration of the enormous data space that will be generated by HuBMAP, we propose to implement a set of tools that enable systematic navigation and retrieval of omics and imaging data beyond text-based search. We will focus on the use of metadata and ontologies to accomplish this through interfaces similar to our SATORI system for ontology-guided visual exploration of biomedical data repositories. Finally, we will support collaboration within HuBMAP and the HIVE through visual analysis tools that enable monitoring of the overall project progress and data quality, including timelines for planned experiments, submitted data, and others. Page 6 Project Summary/Abstract