← Go back


COVID19DS based on CORD-19 datasets

About the demo


The rapid generation of large amounts of information about the novel coronavirus SARS-CoV-2 and the disease COVID-19 makes it increasingly difficult to gain a comprehensive overview of current insights related to the disease. This holds especially for scientific research, where a growing number of publications provide insights that might support the development of a cure or vaccine. With this work, we aim to support the rapid access to a comprehensive data source on COVID-19 targeted especially at researchers. Our dataset, Covid-19-DS, an RDF knowledge graph of scientific publications, abides by the Linked Data and FAIR principles. The base dataset for the extraction is CORD-19, a dataset of COVID-19-related publications, which is regularly updated. Consequently, Covid-19-DS is updated periodically. Our generation pipeline applies named entity recognition, entity linking and link discovery approaches to the original data. The current version of the resulting dataset contains 69,434,763 triples and is linked to 9 other datasets by over 1 million links. In a case study, we demonstrate the usefulness of our knowledg graph for different applications. Covid-19-DS can be accessed as an RDF dump, a SPARQL endpoint, and via an HTML endpoint. All data we generated is available under the CC BY-NC license. 3 The software developed for the extraction is available under the GPL 3.0 license.

The number of papers pertaining to SARS-CoV-2 and COVID-19 has surged over the last few months, making it hard to keep track of the latest research findings on the subject matter. Hence, the Allen Institute initiated a growing corpus of publications about COVID-19 called CORD-19, which is updated on a regular basis. we present Covid-19-DS, a comprehensive RDF knowledge graph of COVID-19 based on CORD-19. We provide a detailed representation of the COVID-19 publications in RDF including properties like publication title, authors names and their institutions, paper sections (e.g., abstract, introduction, body, discussion etc.) and annotated references (e.g., references to figures). Resources such as authors and named entities augment the original data and make it easier to process for the sake of question answering and machine learning. All resources in the dataset are dereferenceable HTTP IRIs, which can be accessed via LodView 5 or via the dataset’s SPARQL endpoint. 6 In addition, we link our dataset to the biomedical entities in other relevant datasets (e.g., DrugBank, Sider, Kegg).

RDF Data Model


We use the following namespaces in COVID19-DS

@prefix cvdr: <https://covid-19ds.data.dice-research.org/resource/> .
@prefix cvdo: <https://covid-19ds.data.dice-research.org/ontology/> .
@prefix bibo: <http://purl.org/ontology/bibo/> .
@prefix bibtex: <http://purl.org/net/nknouf/ns/bibtex#> .
@prefix dcterms: <http://purl.org/dc/terms/> .
@prefix fabio: <http://purl.org/spar/fabio/> .
@prefix foaf: <http://xmlns.com/foaf/0.1/> .
@prefix its: <http://www.w3.org/2005/11/its/rdf#> .
@prefix nif: <http://persistence.uni-leipzig.org/nlp2rdf/ontologies/nif-core#> .
@prefix prov: <http://www.w3.org/ns/prov#> .
@prefix rdf: <http://www.w3.org/1999/02/22-rdf-syntax-ns#> .
@prefix rdfs: <http://www.w3.org/2000/01/rdf-schema#> .
@prefix schema: <http://schema.org/> .
@prefix sdo: <http://salt.semanticauthoring.org/ontologies/sdo#> .
@prefix swc: <http://data.semanticweb.org/ns/swc/ontology#> .
@prefix vcard: <http://www.w3.org/2006/vcard/ns#> .
@prefix xml: <http://www.w3.org/XML/1998/namespace> .
@prefix xsd: <http://www.w3.org/2001/XMLSchema#> .
@prefix inria: <http://ns.inria.fr/covid19/> .
@prefix ncbi: <https://www.ncbi.nlm.nih.gov/pmc/articles/> .
@prefix pubnt: <http://pubannotation.org/docs/sourcedb/CORD-19/sourceid/> .
@prefix ldf: <https://data.linkeddatafragments.org/> .
@prefix fccc: <https://fhircat.org/cord-19/fhir/Commercial/Composition/> .
@prefix makg: <http://ma-graph.org/property/> .

RDF Example Resource

cvdr:pmc1616946     a swc:Paper,
                    schema:ScholarlyArticle ;
    dcterms:license "cc-by-nc" ;
    dcterms:title "Antisense-induced ribosomal frameshifting" ;
    bibtex:hasAuthor    cvdr:christineAnderson,
                        cvdr:michaelHoward ;
    bibtex:hasJournal "Nucleic Acids Res" ;
    bibo:doi "10.1093/nar/gkl531" ;
    bibo:pmid "16920740" ;
    fabio:hasPubMedCentralId "PMC1616946" ;
    schema:url ncbi:PMC1616946 ;
    fccc:c1ad13d83e926979dbf2bbe52e4944082f28dfea.json ;
    owl:sameAs inria:c1ad13d83e926979dbf2bbe52e4944082f28dfea,
        ncbi:pmc1616946 ;
    prov:hadPrimarySource cvdr:cord19Dataset ;
    foaf:sha1 "c1ad13d83e926979dbf2bbe52e4944082f28dfea" ;
    cvdo:cordUid "xgwbl8em" ;
    cvdo:hasBody cvdr:pmc1616946_Body ;
    cvdo:hasDiscussion  cvdr:pmc1616946_Discussion ;
    cvdo:hasIntroduction    cvdr:pmc1616946_Introduction ;
    cvdo:publishTime    "2006-08-18" ;
    cvdo:sourceX    "PMC" .

cvdr:pmc1616946_Introduction a cvdo:PaperIntroduction ;
    cvdo:hasSectionv cvdr:pmc1616946_Section1, cvdr:pmc1616946_Section2,
        cvdr:pmc1616946_Section3, cvdr:pmc1616946_Section4,
        cvdr:pmc1616946_Section5, cvdr:pmc1616946_Section6 .

Example section representation

cvdr:pmc1616946_Section1 a sdo:Section ;
    nif:isString "The standard triplet 
    readout of the genetic code can be 
    reprogrammed by signals in the mRNA 
    to induce ribosomal frameshifting [reviewed in (1-3)]." ;
    bibtex:hasTitle "INTRODUCTION" .

Example of a reference to a paper and its associated bibtex entry

cvdr:PMC1616946_Section1_B1_1 a nif:Phrase ;
    nif:anchorOf "1" ;
    nif:beginIndex "140"^^xsd:nonNegativeInteger ;
    nif:endIndex "141"^^xsd:nonNegativeInteger ;
    nif:referenceContext cvdr:PMC1616946_Section1 ;
    its:taIdentRef cvdr:PMC1616946_B1_1 .

cvdr:PMC1616946_B1_1 a bibtex:Entry ;
    bibtex:Inbook "159-183" ;
    bibtex:hasAuthor cvdr:DMDunn, cvdr:JFAtkins,
                       cvdr:RBWeiss, cvdr:RFGesteland ;
    bibtex:hasTitle "Ribosomal frameshifting from -2 to +50 nucleotides";
    bibtex:hasVolume "39" ;
    bibtex:hasYear 1990 ;
    schema:EventVenue "Prog. Nucleic Acid Res. Mol. Biol." .

Provenance information for the non-commercial dataset

cvdr:cord19Dataset a prov:Entity ;
    "2020-05-21T02:52:02+00:00"^^xsd:dateTime ;
    aws.com/latest/document_parses.tar.gz" .

Sparql endpoint

We publicly serve our final RDF dataset via our SPARQL endpoint link. Some example queries are performed below.

  • List all paper URIs written by the author “Ian Mackay.”
      ?paper bibtex:hasAuthor ?author .
      ?author  foaf:firstName "Ian" .
      ?author foaf:lastName "Mackay" .
  • List the top 10 papers-URIs with the most number of authors.
  SELECT ?author count( * ) as ?cnt
      ?paper a swc:Paper .
      ?paper bibtex:hasAuthor ?author .
  LIMIT 10
  • List all papers and sections mentioning “folic acid.”
  SELECT DISTINCT  ?paper ?section
    WHERE {
        ?s nif:anchorOf "folic acid" .
        ?s nif:referenceContext ?section .
        ?body cvdo:hasSection ?section .
        ?paper cvdo:hasBody ?body .
  • SPARQL example for retieving more data via interlinking with MAKG
  SELECT DISTINCT ?name ?paperCount ?citationCount WHERE {
      ?paper a swc:Paper.
      ?paper bibtex:hasAuthor ?author .
      ?author owl:sameAs ?maAuthor.
      SERVICE <http://ma-graph.org/sparql> {
          ?maAuthor makg:paperCount ?paperCount .
          ?maAuthor makg:citationCount ?citationCount.
          ?maAuthor foaf:name ?name.
  LIMIT 100