Supporting Biomedical Discovery with the ROBOKOP Graph Knowledgebase.

使用 ROBOKOP 图知识库支持生物医学发现。

• 批准号: 10494653
• 负责人: Christopher Bizon
• 金额: $ 81.9万
• 依托单位: UNIV OF NORTH CAROLINA CHAPEL HILL
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-05 至 2027-06-30
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/10494653
• 关键词: Academia; Address; Age; Algorithmic Software; Algorithms; Area; Asthma; Biochemical; Biomedical Research; Case Study; Chemical Exposure; Clinical; Collection; Communities; Complex; Coronavirus Infections; Coupled; Data; Data Set; Data Store; Databases; Development; Disease; Drug Exposure; Ensure; FAIR principles; Generations; Genes; Goals; Graph; Growth; Immune; Individual; Industry; Informatics; Ingestion; International; Knowledge; Knowledge Discovery; Link; Lung diseases; Maintenance; Mediating; Methodology; Mining; Outcome; Pathway interactions; Patients; Pharmaceutical Preparations; Primary Ciliary Dyskinesias; Privatization; Public Domains; Publications; Research; Resource Development; Review Literature; Science; Scientific Advances and Accomplishments; Semantics; Services; Source; Structure; Support System; System; Technology; Testing; Translational Research; adverse outcome; asthma exacerbation; austin; base; community engagement; data infrastructure; design; drug induced liver injury; flexibility; hackathon; high throughput technology; improved; innovation; knowledge graph; knowledgebase; light weight; method development; network models; novel; pollutant; prototype; repository; research and development; success; tool; trend; user-friendly; virtual

The proliferation of high-throughput technologies has led to previously unimaginable growth in biomedical research data sets and knowledgebases. Nearly all these data and knowledge sources address specialized areas of biomedical research, leading to natural diversity but also growing disintegration between individual knowledgebases. This trend generates downstream inefficiencies when applying analytics to enable actionable knowledge discovery from databases. Growing efforts, both in academia and industry, are focused on the development of methods and tools to enable semantic integration and concurrent exploration of disparate biomedical knowledge sources. Recent innovations include the development of biomedical `knowledge graphs' (KGs) that support knowledge discovery through the application of querying and reasoning algorithms and tools. Our team has contributed to these efforts by developing a KG-based question-answering system termed Reasoning Over Biomedical Objects linked in Knowledge-Oriented Pathways (ROBOKOP). Herein, we propose synergistic research and development efforts that aim to significantly advance the ROBOKOP graph knowledgebase capabilities to contribute to high-impact applications across diverse biomedical research domains. Our overarching goal is to equip users with a unique and comprehensive knowledgebase system that supports the rapid generation of mechanistic hypotheses that can explain, validate, or predict biomedical phenomena. We will achieve our objectives by executing studies planned under the following Specific Aims: Aim 1. Enrich and Enhance the ROBOKOP graph knowledgebase. We will enhance the data and infrastructure of the ROBOKOP KB. Aim 2. Provide tools to explore the ROBOKOP graph knowledgebase. We will enhance the ROBOKOP KG by developing and employing novel reasoning tools for KG mining and edge inference. Aim 3. Prove utility and promote use of the ROBOKOP graph knowledgebase through impactful use cases. We will conduct several collaborative proof-of-concept research applications in diverse biomedical domains and diseases. We will actively promote community engagement, user acceptance, and broader impact of ROBOKOP. We expect that our diverse, cutting-edge approach to research, development, and community engagement, coupled with our high-impact biomedical applications, will lead to the formation of a core group of regular users, promote long-term sustainability, and generate impactful new scientific knowledge and mechanistic hypotheses for subsequent testing.

高通量技术的普及导致了生物医学领域以前难以想象的增长 研究数据集和知识库。几乎所有这些数据和知识来源都涉及专门的 生物医学研究领域，导致自然多样性，但也导致个体之间日益分裂 知识库。当应用分析来实现可操作性时，这种趋势会导致下游效率低下 从数据库中发现知识。学术界和工业界越来越多的努力集中在 开发方法和工具以实现语义集成和并发探索不同的 生物医学知识来源。最近的创新包括生物医学“知识图”的开发 （知识图谱）通过应用查询和推理算法和工具来支持知识发现。 我们的团队通过开发一个基于知识图谱的问答系统为这些努力做出了贡献 对知识导向路径（ROBOKOP）中链接的生物医学对象进行推理。在此，我们建议 旨在显着推进 ROBOKOP 图的协同研发工作 知识库能力有助于跨不同生物医学研究的高影响力应用 域。我们的首要目标是为用户提供一个独特且全面的知识库系统， 支持快速生成可以解释、验证或预测生物医学的机制假设 现象。我们将通过执行以下具体目标下计划的研究来实现我们的目标： 1.丰富和增强ROBOKOP图知识库。我们将加强数据和基础设施 ROBOKOP 知识库的。目标 2. 提供探索 ROBOKOP 图知识库的工具。我们将 通过开发和采用用于知识图谱挖掘和边缘的新颖推理工具来增强 ROBOKOP 知识图谱 推理。目标 3. 通过有影响力的方式证明 ROBOKOP 图知识库的实用性并促进其使用 用例。我们将在不同的生物医学领域进行多项合作概念验证研究应用 领域和疾病。我们将积极促进社区参与、用户接受度以及更广泛的 ROBOKOP 的影响。我们期望我们多样化、尖端的研究、开发和 社区参与，加上我们高影响力的生物医学应用，将导致形成 经常用户的核心群体，促进长期可持续性，并产生有影响力的新科学知识 以及后续测试的机制假设。