High-resolution characterization of recent tuberculosis transmission in Botswana using geospatial and genomic data – the Kopanyo Study

Abstract Introduction Combining genomic and geospatial data can be useful for understanding Mycobacterium tuberculosis (Mtb) transmission in high tuberculosis burden settings. Methods We performed whole genome sequencing (WGS) on Mtb DNA extracted from sputum cultures from a population-based tuberculosis study conducted in 2012–2016 in Gaborone, Botswana. We used kernel density estimation, spatial K -functions, and created spatial distributions of phylogenetic trees. WGS-based clusters of isolates ≤5 single nucleotide polymorphisms were considered recent transmission, and large WGS-based clusters (≥10 members) were considered outbreaks. Results We analyzed data from 1449 participants with culture-confirmed TB. Among these, 946 (65%) participants had both molecular and geospatial data. A total of 62 belonged to five large outbreaks (10–19 participants each). Geospatial clustering was detected in two of the five large outbreaks, suggesting heterogeneous spatial patterns within the community. Conclusions Integration of genomic and geospatial data identified distinct patterns of tuberculosis transmission in a high-tuberculosis burden setting. Targeted interventions in these smaller geographies may interrupt on-going transmission.