PlagueKG: A plague knowledge graph based on biomedical literature mining

Abstract BackgroundPlague is an extremely horrific infectious disease caused by Yersinia pestis. Its mortality rate is exceedingly high that threatens the lives of human and animals seriously. There is a vast amount of knowledge related to plague in the literature at present. It is particularly important to obtain useful information from an enormous amount of literature and intuitively represent it in the form of knowledge graph. This will provide help to researchers to quickly understand plague’s complex pathogenesis and potential therapeutic approaches and provide the potential action mechanism of corresponding drug candidates. Consequently, the speed of vaccine research and development could be accelerated.ResultsThis paper identifies and extracts plague-related entities and relations based on 5388 abstracts that obtained from PubMed biomedical literature library automatically. Then we construct the plague knowledge graph called PlagueKG, which contains 9633 nodes of 33 types such as disease, gene, protein, species, symptom, treatment, geographic location and so on, and 9466 association relations such as disease-gene, gene-protein, disease-species and so forth. The Neo4j graph database is used to store the relational data in the form of triple. Finally, a multi-factor correlation knowledge graph centered on plague is constructed, as well as a plague knowledge base platform.ConclusionsWe extracted and integrated knowledge from existing plague-related literature using text mining techniques, and constructed a plague knowledge graph, which shows detailed plague-related knowledge in an intuitive and clear way. To the best of our knowledge, it is the first plague knowledge graph that is built using literature mining techniques. In addition, our plague knowledge base platform successfully managed and visualized a large amount of structured data related to plague. Researchers can acquire integrated plague information more conveniently by using this platform. It provides more direct and comprehensive knowledge of the disease.