RAPID: Dashboard for COVID-19 Scientific Development

科学发现有赖于知识的积累。关于任何特定主题的文章数以千计，但没有一个人能读完所有的文章。这一限制在新冠肺炎时代更为重要，在这个时代，可靠的知识可能意味着生死之别。该项目致力于改进与科学知识合成相关的方法，开发一个可视化仪表板来总结新冠肺炎相关研究成果。主要目标是将白宫目前的新冠肺炎文献数据集与PubMed的知识图谱和Data2Discovery开发的药物发现知识图谱整合在一起。这将使“抗击新冠肺炎仪表盘”的创建成为可能，这是一个可视化工具，将集中和可视化与CoVID相关的关键、最新数据和科学信息。这个仪表板将帮助科学家和临床医生访问和可视化关于COVID的最新信息。这些信息对于挖掘出版物产生研究假设以及确定科学交流中的合作和创新模式以阻止COVID的传播也至关重要。私人投资机构将向公众开放他们的数据和构建仪表盘的代码，以便于未来的努力，并通过透明的方式增强公众对科学的信任。该项目开发了对抗新冠肺炎的数据集和可视化仪表盘，以推动信息科学，帮助对抗新冠肺炎。这是通过将白宫当前的新冠肺炎文献数据库与PubMed的知识图谱以及将临床前药物发现中数十个公开可用的数据库互联在一起的药物发现知识图谱整合在一起实现的。这个仪表板将显示与新冠肺炎相关的信息，包括：1)PubMed和临床试验中当前提到最多的生物实体(例如，药物、疾病、疫苗、基因)；2)根据PubMed文献和临床试验得出的相关生物实体的演变；3)相关生物实体的网络连接；4)活跃的科学家、团队、机构及其研究课题的列表；5)科学团队的合作，以建立网络并激发潜在的合作，以对抗新冠肺炎。这项研究将推动文本分析方法的发展，超越关键字分析，提高对关键字所指对象的理解，并推动文本方法向基于知识图的分析迈进。它还为最近对与生物实体相关的新冠肺炎科学研究的演变路径的研究增加了一个纵向因素，并以新的和潜在的创新方式将科学与相关研究领域联系起来。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Scientific discovery depends on the accumulation of knowledge. There are thousands of articles on any given topic, but no one person can read them all. This limitation is more important in the COVID-19 era, where dependable knowledge can mean the difference between life and death. This project works to improve methods related to the synthesis of scientific knowledge by developing a visual dashboard to summarize COVID-19 related research efforts. The main goal is to integrate a current COVID-19 literature dataset from the Whitehouse with a knowledge graph from PubMed and a drug discovery knowledge graph developed by Data2Discovery. This would enable the creation of the “Fight COVID-19 Dashboard,” a visualization tool that would centralize and visualize crucial, up to date data and scientific information related to COVID. This dashboard will help scientists and clinicians access and visualize the most recent information about COVID. Such information is also crucial for mining publications to generate research hypotheses and for identifying patterns of collaboration and innovation in scientific communication working to stop the spread of COVID. The PIs will make their data and the codes for constructing the dashboard open to the public to enable future efforts and enhance public trust in science through transparency.This project develops the Fight COVID-19 Dataset and visual dashboard to advance information science and aid in the fight against COVID-19. This is accomplished by integrating a current COVID-19 literature dataset from the White House with a knowledge graph from PubMed and a drug discovery knowledge graph interlinking dozens of publicly available databases in pre-clinical drug discovery. This dashboard will display COVID-19 related information, including: 1) the currently most mentioned biological entities (e.g., drugs, diseases, vaccines, genes) in PubMed and clinical trials; 2) the evolution of related biological entities according to PubMed literature and clinical trials; 3) the network connections of related biological entities; 4) the lists of active scientists, teams, and institutions and their research topics; and 5) collaborations of scientific teams to enable networking and inspire potential collaborations to fight against COVID-19. This research will advance textual analysis methods by moving beyond keyword analysis towards advanced understanding of the objects that the keywords indicate and propelling textual methods towards knowledge graph-based analysis. It also adds a longitudinal element to recent investigations of the evolving pathway of COVID-19 scientific studies related to bio entities and links the science of science to related research domains in new and potentially innovative ways.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.