Convincing evidence exists that host genes influence the outcome of infection in human tuberculosis. We are employing two complementary strategies to find the genes involved, a linkage‐based, comprehensive genome screen and an association‐based candidate gene study. In a genome screen of 282 markers on 92 affected sib pairs we have found evidence of co‐segregation of disease with five markers, but further studies are required to replicate these results. The absence of a single strongly linked marker demonstrates that susceptibility to human tuberculosis is not controlled by a single major gene. Using a candidate gene approach investigating over 400 tuberculosis cases and 400 ethnically matched healthy controls we have found evidence that NRAMP1 and vitamin D receptor gene (VDR) polymorphisms are associated with tuberculosis. It is hoped that by identifying the genes which account for why only a minority of those exposed to tuberculosis develop disease, we will develop new insights into potential therapeutic and preventative strategies.