Tuberculosis, caused by Mycobacterium tuberculosis, is the most common mycobacterial disease in the world and remains a leading public health problem. Numerous other mycobacterial species are present in the environment and are most often termed atypical or nontuberculous mycobacteria. Like the attenuated vaccine Bacille Calmette-Guérin (BCG) they are generally poorly virulent, even so they can be at the origin of severe infections if the host immune response is impaired. It has been clearly demonstrated that the intrinsic virulence of a mycobacterial species is not the only factor determining disease severity, which is illustrated by the observation that the majority of individuals infected with M. tuberculosis do not develop clinical disease. Numerous arguments suggest that disease severity depends largely on susceptibility/resistance determined by the host genetic make up. In the following review we will discuss the studies on the genes implicated in complex predisposition to tuberculosis and Mendelian predisposition to disease caused by less virulent mycobacteria, proposing a continuous spectrum between those two types of predisposition.