Computes similarities between an input text and already curated instances contained in a knowledge base to infer gene ontology concepts. The main limit of GOCat, both observed by reviewers and mentioned in our papers, was the difficulty to integrate it in a curation workflow: it is stated that GOCat proposes more accurate (Gene Ontology) GO concepts, but these concepts are inferred from the whole abstract, then the curators still have to locate the function in the publication and to link the correct GO concept with a gene product.
BiTeM group, University of Applied Sciences, HEG – Library and Information Sciences, Geneva, Switzerland; Division of Medical Information Sciences, University and Hospitals of Geneva, Geneva, Switzerland; SIBtex group, SIB Swiss Institute of Bioinformatics, Geneva, Switzerland
GOCat funding source(s)
This work was funded by the Swiss National Fund for Scientific Research [BiND project 3252B0-105755] and by the FP7 program [Khresmoi project FP7–257528].