Evolutionary_and_epidemiological_dynamics_of_multistrain_bacterial_pathogens

Population genomics demonstrates many important bacterial species are subdivided into genetically, and phenotypically, distinct strains. Evolution in response to interventions, such as vaccine-associated population dynamics and increases in antibiotic-resistant pathogens, is often primarily driven by strains changing in prevalence, or new strains emerging. Using data from Streptococcus pneumoniae, the recently-developed 'multilocus negative frequency-dependent selection' model promises to explain stable strain co-existence, quantifiably reproduce strain-level responses to perturbations, and forecast population dynamics. This project would develop such ecological modelling across larger datasets, initially focussed on four additional pathogens with divergent patterns of antibiotic resistance that represent current, or intended, vaccine targets. Model assumptions would be tested using genomics, epidemiological modelling and experimentation, and their structure refined accordingly. New open-source methods for fitting these models would be implemented, with the intention of integrating batches of surveillance data, allowing for near real-time analysis and trend prediction. Strains' life cycles, from emergence to replacement, would also be modelled through integrating genomic data with experimentally-measured properties of recombination. This work would identify what evolutionary and ecological pressures affect bacterial epidemiology, establish the predictability of routine population dynamics and new strain emergence or invasion, and inform optimal intervention design for improving disease prevention.