Using Intrahost Genetic Diversity to Understand Rna Virus Evolution and Transmission

Abstract

RNA viruses generate genetically diverse populations during acute infections within human hosts. Studying viral population dynamics in natural infections is important for understanding how new viral variants arise and spread. However, these dynamics are poorly understood in most viruses. In my thesis, I have comprehensively analyzed viral within-host evolution across three unique taxa. In the first study, I defined the within-host diversity of influenza B viruses in a household cohort in southeastern Michigan. Using an experimentally validated next-generation sequencing approach, I found that influenza B viruses accumulate less genetic diversity compared to influenza A viruses. Similar to influenza A, I found that influenza B virus faces a stringent transmission bottleneck. These results suggest a complex relationship between viral mutation rates, intrahost diversity, and global rates of viral evolution. In the second study, I investigated the early evolution of the live-attenuated oral polio vaccine (OPV) by sequencing samples from vaccine recipients and their close contacts in a field trial of polio vaccines. In contrast to endemic viruses, I found that OPV exhibits a significant amount of parallel evolution within primary vaccine recipients. I identified 19 sites under positive selection, most of which were previously thought to evolve neutrally. Between hosts, a tight transmission bottleneck limited the spread of adaptive mutations. These results demonstrate the distinct within-host dynamics of live-attenuated vaccines, highlight the role of transmission bottlenecks in constraining virus evolution, and offer valuable information for interpreting genetic surveillance data in the ongoing effort to eradicate polio. In the third study, I defined the within-host variation of SARS-CoV-2 in hospitalized patients and infected healthcare workers during the first months of the pandemic. I sequenced known virus mixtures to show how viral load impacts the accuracy of variant identification. In contrast to several early reports, I found that intrahost diversity is low over the course of infections. I demonstrated that variants arise in parallel across individuals from separate transmission networks, which complicates the use of intrahost variants in transmission inference. These findings clarify conflicting results from previous work on SARS-CoV-2 variant generation and spread. More broadly, these studies demonstrate the challenges of accurately detecting intrahost viral variants, illustrate how within-host studies can resolve ambiguities observed in global evolutionary dynamics, and provide important context for using intrahost variation in sequence-based transmission inference.