eGIFT: Mining Gene Information from the Literature

摘要: With the biomedical literature continually expanding, searching PubMed for information about specific genes becomes increasingly difficult. Not only can thousands of results be returned, but gene name ambiguity leads to many irrelevant hits. As a result, it is difficult life scientists and curators rapidly get an overall picture from documents that mention its names synonyms. In this paper, we present eGIFT ( http://biotm.cis.udel.edu/eGIFT ), web-based tool associates informative terms, called i Terms, sentences containing them, with genes. To associate Terms gene, ranks based on score which compares frequency occurrence term in gene's general. retrieve (Medline abstracts), considers all names, aliases, Since ambiguous, applies disambiguation step remove matches do not correspond gene. Another additional filtering process applied retain those abstracts focus rather than passing. eGIFT's pre-computed users search by using or EntrezGene identifier. are grouped into different categories facilitate quick inspection. also links Term mentioning allow see relation between We evaluated precision recall 40 genes; 88% 94% were marked as salient our evaluators, UniProtKB keywords these identified Terms. Our evaluations suggest capture highly-relevant aspects Furthermore, showing provide description helps survey high-throughput experiments, annotators find articles describing functions.