A Wavelet Packet Based Approach for the Research of the Avian Influenza Virus Cross-species Infection

Since 2003, the continuously outbreaks of Influenza Virus in Asia has aroused considerable public concern. Qinghai Lake, as a main place where migrating birds propagate, also had the highly pathogenic H5N1 virus broken out. The reason for avian flu cross-species infection is unknown at the present. Aimed at wild bird epidemic monitoring (an application of Qinghai Lake E-science project), we present an approach, which is based on wavelet packet decomposition, clustering and classification, to identify whether an Avian influenza A virus has the ability to cross species infect human. We also developed an intuitive and easy to-use web interface to serve the scientists.