Knowledge graphs provide a model of entities and relationships in a particular domain. These graphs can be used to represent background knowledge and can also be used to infer or discover new relationships through reasoning. Several COVID-19 knowledge graphs have been constructed by combining relations detected in the literature with other ontologies and databases of structured relationships. The CovidGraph(https://covidgraph.org/) is perhaps the largest of these, combining literature, case statistics and genomic and molecular data. Another project, the Knowledge Graph Toolkit [36], integrates the CORD-19 corpus with gene, chemical, disease and taxonomic information from Wikidata (https://www.wikidata.org/) and the Comparative Toxigenomics Database (http://ctdbase.org/), as well as the Blender Lab COVID-KG (http://blender.cs.illinois.edu/covid19/) [99], another COVID-19 knowledge graph focused on drug repurposing. These knowledge graphs are used by several systems in Section on “Text mining systems” to provide entity- or relation-based exploration of the literature or as a way to visualize data. Knowledge graphs can also support automated reasoning and inference and the potential discovery of novel relationships.