VirusFinder: a Web-based virus identification system using viral nucleotide signatures

VirusFinder was developed as a Web-based virus identification system using viral signatures. Nucleotide signature-based detection has been proven to be a rapid and accurate pathogen identification approach in natural outbreaks or acts of bioterrorism. Species-specific signatures of selected families of HFVs including arenaviruses, bunyaviruses, filoviruses, and flaviviruses have been identified by comparative genomics. The unique signatures for each individual species of HFVs were derived from the multiple sequence alignments and were then saved into the signature database after validation. VirusFinder was implemented using J2EE technology with an intelligent signature recognition algorithm. As such, species-level differentiation and identification of HFVs was automatically achieved given an input of unknown nucleotide sequences. Results show that VirusFinder is able to identify HFVs at species level with 100% accuracy.