Reconstructing the Evolutionary History of RNA Viruses using Relaxed Molecular Clocks

Abstract

Teasing apart the evolutionary forces responsible for biological phenomena is difficult in the absence of a detailed evolutionary history, especially if this history is lacking a temporal component. RNA viruses, due to their rapid rate of molecular and phenotypic evolution, provide a unique biological system in which to study the temporal aspects of evolutionary processes. These types of studies are possible because of relaxed molecular clock dating techniques, which allow the rate of evolution to vary across a phylogenetic tree. The primary focus of the research presented here concerns the age of the simian immunodeficiency virus (SIV), the primate precursor to HIV. SIV has long been thought to be an ancient infection in non-human African primates, and it has been hypothesized that codivergence with its primate hosts has shaped the SIV phylogeny and resulted in a virus capable of apathogenic infection. The codivergence theory was tested by comparing the phylogeny of a group of monkeys thought to be exemplary of SIV-host codivergence to the phylogeny of their SIVs (Appendix A). These phylogenies were incongruent, suggesting that SIV may have infected these monkeys after their common ancestor speciated. The codivergence theory was investigated further by estimating the time of most recent common ancestor for the SIV lineages that directly gave rise to HIV, found in sooty mangabeys and chimpanzees (Appendix B). The temporal estimates suggest that these SIV lineages are only of hundreds of years old, much younger than expected under the codivergence hypothesis. Next, the same dating techniques were employed to elucidate the evolutionary history of an emerging RNA virus of shrimp, Taura syndrome virus (Appendix C). This analysis provided phylogenetic confirmation that Taura syndrome virus emerged out of the Americas and spread rapidly around the world. Finally, because all of these studies utilized relaxed molecular clocks, a simulation study was performed to test the hypothesis that relaxed molecular clocks provide higher quality phylogenetic inference compared with traditional time-free phylogenetic inference (Appendix D). This simulation found no difference in the overall quality of phylogenetic inference between these methods.