COVID-19 Literature Knowledge Graph Construction and Drug Repurposing Report Generation

To combat COVID-19, both clinicians and scientists need to digest vast amounts of relevant biomedical knowledge in scientific literature to understand the disease mechanism and related biological functions. We have developed a novel and comprehensive knowledge discovery framework, COVID-KG to extract fine-grained multimedia knowledge elements (entities and their visual chemical structures, relations and events) from scientific literature. We then exploit the constructed multimedia knowledge graphs (KGs) for question answering and report generation, using drug repurposing as a case study. Our framework also provides detailed contextual sentences, subfigures, and knowledge subgraphs as evidence. All of the data, KGs, reports(1), resources, and shared services are publicly available(2).

One Earth, One Humanity Versus the Virus: Examining the Global COVID-19 Social Partnership Communication Networks on Social Media

The COVID-19 pandemic has posed severe challenges that require collaborative efforts from multi-sector organizations. Guided by an institutional theory framework that considers how both organizational fields and national level contexts affect organizations' social partnership communication, the current study examines the COVID-19-related social partnership communication network on social media. The cross-national study using semantic network analysis and exponential random graph models (ERGMs) first maps the meaning of COVID-19 social partnership network, and then investigates the role of organizational fields and a country's political system, economic system, educational system, and cultural system on the formation of interorganizational communication ties surrounding the relief efforts of COVID-19. Results reveal the importance of the political system-such as the presence of populist government, economic disparity, and uncertainty avoidance cultural orientation in shaping the social media-based social partnership communication network. In addition, NGOs from multiple issue areas are actively engaged in the network, whereas corporations from manufacturing and financial industries are active players.

Mask-Wearing as a Partisan Issue: Social Identity and Communication of Party Norms on Social Media Among Political Elites

This study draws on the social identity approach (SIA), to examine how political elites (i.e., members of the 116th United States Congress) communicated norms about mask-wearing on social media during the COVID-19 pandemic. Using Twitter data collected in 2020, we found that Republican members of Congress were significantly less likely to promote mask-wearing than Democratic members. We also observed some variations in norm-conforming behaviors among the members of each party. For Republicans, increased loyalty to the Trump leadership was significantly associated with a lower level of mask promotion. For Democrats, we found some evidence that loyalty to the party predicted higher levels of mask promotion. On the other hand, interactions with out-group members decreased adherence to party norms for both Republican and Democratic members of Congress. These findings allow us to better understand the social-psychological effects of party membership among political elites as well as the importance of leader-follower relationships and intergroup interactions.

Analysis on the Development Trend of COVID-19 Outbreak in Beijing Based on the Cluster Analysis and SEIR Model

[Purposes[ Through the cluster analysis and the SEIR Model established, to study the trend of COVID-19 outbreak in all the regions of Beijing and analyze the prevention and control effects to provide basis for its prevention and control. [Methods] We collected the data related to COVID-19 outbreak in Beijing from January 21 to March 31, 2020, used SPSS19.0 and MetaboAnalysis software to conduct a cluster analysis on the outbreak data of the cases in all the regions of Beijing, the cases from others places to Beijing and the imported cases and built a SEW Model through Matlab to simulate the development trend of COVID-19 outbreak in Beijing. [Results] According to the cluster analysis, the high incidence was mainly distributed in Chaoyang District, Haidian District, Changping District, Fengtai District, Dazing District, Xicheng District and among the people from other places to Beijing;according to the SEIR Model, in mid-to-late February, the inflection point of the local outbreak appeared but there were still the enormous risks of imported cases and the prevention and control was still in a severe situation. [Conclusion] At present, the outbreak situation has tended to be gentle but the comprehensive and strict prevention and control shall still be strengthened to strictly prevent the imported cases and resolutely curb the spread of the outbreak.