Reconstruction and Analysis of Within-host Longitudinal HIV-1 Evolution by a Distance-based Sequential-linking Algorithm

In this study we develop a new phylogenetic method to describe within-host (patient) viral evolution. Our method accommodates ordinary phylogenetic tree with viral evolution. Viral genomic data often arise from longitudinal sampling and at each observation time we have many types of viral variants. Our method establishes the phylogenetic relation between these viral variants observed at consecutive time points. Our method also can incorporate both neutral evolution and Darwinian selective evolution. In several viral genomic sites, the Darwinian evolutions are seen, so that the phylogenetic method for the viral evolution should deal with both neutral and selective modes of evolution. We apply Yang's codon-based method to specify when and how neutral and adaptive modes of evolution take place in the course of viral evolution. In our previous study, we proposed the preliminary formulation of our longitudinal phylogenetic tree method, based on the maximum likelihood method. But calculation of likelihood takes much time and number of possible tree topologies increase exponentially according to the increase of number of variants. In this study, we propose another new algorithm, distance-based sequential-linking algorithm using neighbor-joining (NJ) method. We applied this new algorithm to a data of the V3 region of the HIV-1 envelope genes sequenced at different years after the infection of a single patient. The results suggest that this algorithm successfully reconstruct a longitudinal phylogenetic tree that describes the within-host viral evolution.