Despite the infectious agent that causes tuberculosis having been discovered in 1882, many aspects of the natural history and transmission dynamics of TB are still not fully understood. This is reflected in differences in the structures of mathematical models of TB, which in turn produce differences in the predicted impacts of interventions. Gaining a greater understanding of TB transmission dynamics requires further empirical laboratory and field work, mathematical modelling and interaction between them. Modelling can be used to quantify uncertainty due to different gaps in our knowledge to help identify research priorities. Fortunately, the present moment is an exciting time for TB epidemiology, with rapid progress being made in applying new mathematical modelling techniques, new tools for TB diagnosis and genetic analysis and a growing interest in developing more-effective public-health interventions.