Viral infections are common and are particularly problematic in immunocompromised individuals. However, other than for HIV, Hepatitis B, Hepatitis C, Influenza, and more recently SARS-CoV-2, there have been few approved drugs available for treating viral infections. Instead, repurposed drugs are often used, especially at the beginning of the current pandemic, for treating SARS-CoV-2. It remains unclear how these repurposed drugs act on the viral population and whether the suppression of viral load we observe is attributed to the drug or the immune response or a combination of both.

The research presented in this thesis primarily focuses on the study of two RNA viruses, SARS-CoV-2 and Norovirus. A mixture of viral load data and viral genomic data were analysed to understand the course of infection within individuals. First, we presented a meta-analysis on SARS-CoV-2 viral load dynamics where we investigated the changes of viral dynamics over time, with and without the presence of antiviral drugs. Then, we presented an evolutionary model used for reconstructing haplotypes in mixed infections. Finally, we demonstrated the use of viral deep sequencing to study the within-host evolution of RNA viruses. We identified mutagenic signatures and consensus level changes associated with antiviral treatments. We developed unique methods to analyse viral sequences which allow us to understand the within-host genomic variations and hence inform our understanding of the heterogeneous efficacy of a drug between patients.

Overall, this thesis provides insights into how the efficacy of a drug can be evaluated by monitoring the within-host viral dynamics and evolution.