Covid19 Literature Bio-curation, Text-mining And Semantic Web Technologies (COVlit)

The world wide scientific response to the COVID-19 pandemic is reflected in the ever-growing scientific literature. Enriching our current knowledge base (https://biokb.lcsb.uni.lu) with these publications requires joint efforts at each stage of the chain of process involved in the text-mining pipeline. This pipeline comprises several challenging tasks such as part-of-speech tagging, entity recognition and normalisation, or event extraction, which are essential to discover relevant knowledge in the form of entities, relations and events. Such knowledge is then made available to the public via semantic web technologies and curated through collaborative curation interfaces. This literature growth calls for updated ontologies to cope with the new terms and biological entities, more robust event extraction and named entity recognition, as well as further development of our collaborative curation interface and search tools. In this project we are building a knowledge base with relevant cross-domain events and relationships from available COVID19 publications can help in-silico and in-vitro researchers navigate the growing COVID-19 literature corpus.