Viral sequence data has great potential for answering questions about the epidemiological dynamics and evolution of viruses. Classical approaches have sought amino acid changes that alter pathogenesis or transmissibility by influencing a virus's ability to enter or replicate within cells. However, this approach rarely recognises the fundamental impact of heterogeneous host contact structures and existing immunological responses on viral transmission. This thesis draws heavily on ecological and immunological concepts to explore the epidemiological dynamics, diversity and evolution of viruses using molecular sequence data. A number of different research approaches and study systems are used in this thesis. I begin by describing a novel polyomavirus in a European badger, and apply phylogenetic techniques to analyze the evolutionary history of the Polyomaviridae. I subsequently describe a large metaviromic study in a population of wild mute swans, for which host demographic data are available. I describe nine new viral species and test whether age and season are associated with differences in abundance and prevalence of different viral taxonomic groups. The study highlights the potential of metaviromics for investigating viral epidemiological dynamics in natural populations. Influenza A viruses of avian origin (AIV) threaten human and animal health. Using phylogeographic methods, I reconstruct the spatial spread of an H5N8 virus at a regional scale, and investigate how bird density and migration shaped this dispersal. Despite the importance of acquisition of humoral immunity to different strains throughout the lifespan of wild birds for epidemiological dynamics, this topic is poorly understood. I assess the accumulation of immune responses to AIV with age in mute swans. I consider how ecological factors, including age-structured immunity, might have affected the epidemiology of an H5N8 outbreak in the population.