BioHCVKD: a bioinformatics knowledge discovery system for HCV drug discovery - identifying proteins, ligands and active residues, in biological literature.

Hepatitis C Virus (HCV) causes significant morbidity worldwide with restricted treatment options and lack of a universal cure which necessitate design of novel drugs. Researchers face an enormous growth of literature with very small portions of HCV knowledge accessible in structured way. This paper proposes the BioHCVKD that helps researchers to annotate relevant HCV information targeted to accelerate HCV drug discovery. BioHCVKD combines the dictionary based filtering and conditional random field (CRF) based gene mention tagger. BioHCVKD is supported by two modules, the Abstract Insertion module, and the Protein Insertion module. BioHCVKD achieves a recall of 73.25%, a precision of 70.5% and F-score of 71.85%, which improves the performance of the name entity tagger.