The majority of isoniazid and rifampin resistant strains had mutations in katG315 (71.9%) and rpoB531 (50%). However, rpoB mutations in codons 511, 516 and 533 were also detected in five rifampin susceptible strains. MIC determinations revealed low-level rifampin resistance for those strains. Thus, the sensitivity and specificity of sequencing of katG for detection of drug resistance were 86.7% and 100% and for sequencing of rpoB 100% and 93.8%, respectively.Strikingly, none of the streptomycin resistant strains had mutations in rrs, but 47.5% harboured mutations in rpsL. Further changes were detected in gidB. Among ethambutol resistant strains 46.7% had mutations at embB306. Pyrazinamide resistant strains displayed a variety of mutations throughout pncA. The specificities of sequencing of rpsL, embB and pncA for resistance detection were high (96-100%), whereas sensitivities were lower (48.8%, 73.3%, 70%).Our study reveals a good correlation between data from molecular and phenotypic resistance testing in this high-incidence setting. However, the fact that particular mutations in rpoB are not linked to high-level resistance is challenging and demonstrates that careful interpretation of molecular resistance assays is mandatory. In addition, certain variations, especially in gidB, appear to be phylogenetically informative polymorphisms rather than markers for drug resistance.