Drugs, Diseases, Genes and Proteins in the CORD-19 Corpus

The BioNER+BioNEN system described in the paper "<strong><em>An Overview of Drugs, Diseases, Genes and Proteins in the CORD-19 Corpus</em></strong>", <em>Badenes-Olmedo, Carlos et. al, (2022)</em> was used to identify and normalize the drugs, diseases and genetic-related terms mentioned in the CORD-19 corpus (January 2022 Edition). Entity recognition and normalization was done for each paragraph of the scientific article. A first group of labels is created to identify the medical terms as they appear in the text (i.e. <em>diseases_ss</em>, <em>chemicals_ss</em>, <em>genetics_ss</em>), and in a standardized way (i.e. <em>disease_terms_ss</em>, <em>chemical_terms_ss</em>, <em>genetic_terms_ss</em>). In the case of diseases and genes/proteins, a predefined category is also established during the normalization process (i.e. <em>disease_types_ss</em>, <em>genetic_types_ss</em> ). The following group of labels contains the codes for each of the classification systems described in Section 3 (i.e. <em>mesh_codes_ss</em>, <em>atc_codes_ss</em>, <em>cid_codes_ss</em>, <em>doid_codes_ss</em>, <em>cui_codes_ss</em>, <em>icd10_codes_ss</em>, <em>icd9_codes_ss</em>, <em>gard_codes_ss</em>, <em>snomed_codes_ss</em>, <em>nci_codes_ss</em>, <em>ncbi_codes_ss</em>, <em>uniprot_codes_ss</em>). The suffix <em>_ss</em> in all tags indicates that the format is a textual list (i.e. string sequence).