Profile Hidden Markov Model for Detection and Prediction of Hepatitis C Virus Mutation

Hepatitis C virus (HCV) is a widely spread disease all over the world. HCV has very high mutation rate that makes it resistant to antibodies. Modeling HCV to identify the virus mutation process is essential to its detection and predicting its evolution. This paper presents a model of HCV based on profile hidden Markov model (PHMM) architecture. An iterative model learning procedure is proposed and applied to both full-length sequence of virus and its very high variation (mutation) zone called NS5A. A pilot study on HCV dataset of type 4 is conducted which is of special concern in Egypt