Escherichia coli is a leading cause of infections in hospitals and the community. The increasing global prevalence of E. coli producing extended-spectrum beta-lactamases (ESBL-EC) represents a major public health challenge. ESBL-EC infections are associated with excess morbidity, mortality, longer hospital stay and higher healthcare costs. I aim to define the source(s) of ESBL-EC acquisition in a defined hospital population, in which I consider other patients, the hospital environment, food and water as potential sources. I will test the hypothesis that genome sequencing of ESBL-EC from these sources will define the major reservoirs for these strains and transmission pathways between them. I will place this genome data into a clinical context through a prospective longitudinal cohort study of ESBL-EC in hospitalised haematology patients. The most probable source(s) of new ESBL-EC acquisition in this cohort will be determined using epidemiological and whole genome sequence data on ESBL- EC from patients entering and leaving hospitals (n=975), the hospital environment (n=150), livestock (n=425), wastewater (n=425), a nearby nursing home (n=427), and bloodstream infections in the UK and Ireland (n=1519). Using mathematical modelling, I will infer transmission routes and predict the likely impact of different interventions on antimicrobial resistance reservoirs and transmission events The bacterium Escherichia coli is a major cause of infections in people in hospitals and the community. E. coli can become resistant to commonly used antibiotics through the acquisition of resistance-encoding genes (abbreviated as ESBL-EC, for extended-spectrum beta-lactamase E. coli). When people are infected with these strains, their infection can be difficult to treat and they are at a higher risk of death. Vulnerable patients in hospital often acquire these in their gut flora from which they can then become infected. The central question that we want to address is to determine the routes by which they are acquired. I aim to use genome sequence data of ESBL-EC from hospitalised haematology patients (n=975), the hospital environment (n=150), livestock (n=425), wastewater (n=425), nursing home residents (n=427), and bloodstream infections in the UK and Ireland (n=1519). I will analyse these to determine the genetic relatedness of isolates from different reservoirs, which will highligh t shared bacterial genotypes and routes of spread between them. Having defined the most likely reservoirs and routes of spread by which patients acquire ESBL-EC, mathematically modelling will also be used to infer transmission, test hypotheses and to determine the likely impact of interventions to reduce transmission of ESBL-EC.