Using genomic and human mobility data to elucidate the long-range migration of streptococcus pneumoniae

Abstract

Streptococcus pneumoniae is a globally endemic, human obligate bacteria, and a leading cause child mortality by pneumonia and meningitis worldwide. Despite the enormous public health burden, the extent and mechanisms of pneumococcal spread across spatial scales remains largely unknown. We analysed 6910 geolocated pneumococcal genome sequences from patients, collected from 2000-2014 in South Africa. We develop mechanistic phylogeographic models to reconstruct pathogen spread, incorporating the generation time distribution, human mobility data provided by Meta, population size, and underlying heterogeneities in sampling. We found that pairs from the same province had 2.43 [95%CI 2.02-3.17] times the relative risk of having a recent common ancestor (within 5 years) than distal pairs (>1000km apart). As the evolutionary time between isolates increases the relative probability of being from the same province decreases, however, it is only after around 50 years that the pneumococcus appeared to be well mixed throughout the country We were able to accurately recover this observed pattern of spatialspread using our mechanistic model estimating that most transmission events were local (within municipality 87.6% [95%CI 80.2-88.7%]), with occasional long-distance transmissions. Our framework also showed the key role of