eLwazi Open Science Data Platform for Africa

摘要 数据科学方法在大规模、多样化和复杂的生物医学领域的应用 新发现的数据集需要访问数据、工具、工作流程和计算 基础设施，沿着培训和支持。由于非洲缺乏许多这类设施， 执行技能，需要开发新的基础设施， 非洲背景并解决低收入和中等收入国家存在的障碍。我们 将为DS-I非洲联盟开发一个开放数据科学网关， eLwazi平台及相关培训、支持工具和资源。eLwazi，象征着 “知识”将是一个基于数据生物圈概念的模块化、可扩展平台， 使数据科学工具的应用能够推动健康方面的新发现。关于eLwazi 初学者和高级用户将能够：1）在数据存储库中搜索数据 和目录; 2）检索存储在本地或联合站点的相关数据集; 3）搜索和 检索相关的工具和工作流程; 4）在选择的计算上执行分析 环境;和5）可视化和分享他们的结果。用户将成为 发展进程，就其需求和反馈提供投入，并将得到支持 通过帮助台、用户组和全面的数据科学培训计划。我们将 建立透明的成本模型，并建立非洲的技能和基础设施，以确保长期 平台的可持续性。开放数据科学网关将促进开放， 合作研究，致力于在非洲及其他地区开发新的健康应用。 这一由非洲主导的倡议符合新兴非洲公开赛的愿景和要求 科学平台。一个积极的外联方案将把eLwazi的范围扩大到其他群体， 数据集和联盟，并促进学术界和工业界的伙伴关系。利用 自2012年以来，H3 ABioNet的专业知识和能力与国际 该领域领导人的专业知识，并利用现有的框架和标准， 确保及时实施成功的平台，并与通用 资助项目和其他国际倡议，以确保全球适用性。

Abstract The application of data science methodologies to large-scale diverse and complex biomedical datasets for novel discoveries requires access to data, tools, workflows and computing infrastructure, along with training and support. With many of these lacking in Africa, including the skills for execution, there is a need to develop new infrastructure that takes into account the African context and addresses barriers present in low and middle income countries. We will develop an Open Data Science Gateway for the DS-I Africa consortium that includes the eLwazi platform and associated training, supporting tools and resources. eLwazi, symbolizing “knowledge”, will be a modular, scalable platform based on the Data Biosphere concept, that enables the application of data science tools to drive novel discoveries for health. From eLwazi workspaces, novice and advanced users will be able to: 1) search for data in a data repository and catalogue; 2) retrieve relevant datasets stored locally or at federated sites; 3) search and retrieve relevant tools and workflows; 4) execute their analyses on a choice of computing environment; and 5) visualize and share their results. Users will be an integral part of the development process, providing input on their needs and feedback, and will be supported through a helpdesk, user groups and a comprehensive data science training program. We will develop a transparent cost model and build African skills and infrastructure to ensure long term sustainability of the platform. The Open Data Science Gateway will facilitate open, collaborative research that strives to develop novel health applications in Africa and beyond. This African led initiative aligns with the vision and requirements of the emerging African Open Science Platform. An active outreach program will extend the reach of eLwazi to other cohorts, datasets and consortia and promote partnerships in academia and industry. Leveraging H3ABioNet’s expertise and capacity developed since 2012 together with international expertise from the leaders in the field, and using existing frameworks and standards will ensure a successful platform is implemented timeously and is interoperable with Common Fund programs and other international initiatives to ensure global applicability.