BioThesaurus

[…] A+ indexes concepts from the abstracts including genes, proteins, diseases, symptoms, drugs, enzymes and simple chemical compounds, identified using biological databases and thesauri such as UniProt, BioThesaurus, Unified Medical Language System (UMLS), KEGG and DrugBank. The user interacts with the system by issuing queries in the form of a word, a concept identifier or any Boolean combination of […]

[…] d drugs), and presents them in a tabular format ranked based on co-occurrence statistics. The concept IDs and their names and synonyms are collected from several biomedical databases such as UniProt, BioThesaurus, UMLS, and DrugBank.We queried FACTA with all human protein-coding genes and stored for each query gene the set of related gene IDsc, UMLS and DrugBank terms.GoPubMed[]—developed by the T […]

[…] tatistical machine learning method. The dictionary consists of 266,000 entries for general English words extracted from WordNet [], together with 1.3 million entries for protein names, extracted from BioThesaurus []. The tool is available at http://text0.mib.man.ac.uk/~sasaki/bootstrep/nemine.html.Text mining techniques are typically evaluated against 'gold standards' []. We evaluated the NER tool […]

[…] ext and the synonyms in the dictionaries. It is flexible because the mention and the synonyms are previously pre-processed by dividing the token according to punctuations, numbers, Greek letters, and BioThesaurus terms, and finally ordering the parts of the token alphabetically. The initial lists of synonyms for the four organisms were available in the two editions of the BioCreative challenge: Bi […]

[…] The dictionary was built from two resources: BioThesaurus 2.0 [] and ABGene lexicon []. To improve the dictionary coverage, we converted all the dictionary entries to lowercases, removed hyphens and tokenized the terms from non-letter and non-di […]

[…] e, we enriched this retrieval by additional references that are contained in the BioLexicon (work in progress). The BioLexicon is a new data resource that combines the protein term repository called "BioThesaurus" with other terminological resources (e.g., NCBI taxonomy, ChEBI) and adds linguistically relevant information []. We used terms from the BioLexicon to retrieve additional references to M […]