A likelihood method for the detection of selection and recombination using nucleotide sequences

Different regions along nucleotide sequences are often subject to differentevolutionary forces. Recombination will result in regions having differentevolutionary histories, while selection can cause regions to evolve atdifferent rates. This paper presents a statistical method based onlikelihood for detecting such processes by identifying the regions which donot fit with a single phylogenetic topology and nucleotide substitutionprocess along the entire sequence. Subsequent reanalysis of these anomalousregions may then be possible. The method is tested using simulations, andits application is demonstrated using the primate psi eta-globinpseudogene, the V3 region of the envelope gene of HIV-1, and argF sequencesfrom Neisseria bacteria. Reanalysis of anomalous regions is shown to revealpossible immune selection in HIV-1 and recombination in Neisseria. Acomputer program which implements the method is available.