Abstract
Recombination explains a considerable amount of genetic diversity in natural populations. In general genes located in genomic regions with low levels of recombination have relatively lower levels of polymorphism. Recombination especially affects DNA/RNA virus diversity, pathogenesis, and ability to evade immune pressures and antiviral agents. Thus the accurate detection of recombinant from DNA/RNA virus sequences is very important. Two basic methods ñ distance method and phylogenetic method are available now to detect recombination and look for breakpoints in highly variable sequences, and different approaches have been developed. Most approaches fall into a sequential testing trap, by first using the data to determine recombinant structure and then using the same data to assess significance conditional on the optimal solution. For overcoming this shortfall a new model is developed by using extended Bayesian multiple change-point model to infer the existence and locations cross-over point and parental subtypes. Bayes factor was introduced to test significance of recombination events. This new model has been used to test recombination in any other viruses except HIV. Here we used HBV recombination to test the Bayesian multiple change-point model.