Mycobacterium tuberculosis is the causative agent of tuberculosis, a disease with a global impact of >8 million new cases and 2 million deaths each year. M.tuberculosis also latently infects a third of the world’s population and can reactivate to cause disease at 5% life-time risk or 10% per year for HIV co-infections. Treatment requires prolonged antibiotic therapy for 6 months and treatment failures are leading to the emergence of multiple and extreme drug resistance, threatening the world with untreatable TB. Genome sequence information reveals much about the biology of this pathogen and its metabolic potential, including a complex regulatory potential for gene expression control which implies pathogenicity is driven by differential gene expression in response to multiple and changing environments within the host. Post-genomic technologies such as whole genome DNA microarrays have been used to measure changes in transcriptional responses of M.tuberculosis to a range of different environments, such as low oxygen, low iron or exposure to drugs. Expression profiles of mRNA at a whole genome level are called the Transcriptome and can be used to reveal the nature of the environment of the bacterium as well as reveal the physiological state of the bacteria. Studying the M.tuberculosis transcriptome during association with the host and in models of intracellular infection such as macrophage cultures [1], reveals how M.tuberculosis interacts with the host and responds to in vivo signals. Applying these powerful technologies to models of infection and naturally infected tissue is the next challenge and microarrays are now being used to reveal the metabolic process that underpins the clinically and biologically important state of latency. Advances in molecular biology that allow amplification of mRNA populations have been linked with microarray analysis to derive the transcriptome from low numbers of bacilli likely to be found using in vivo tissue or samples. Considerations of mRNA stability and differential harvesting are also important in order to allow in vivo transcriptome profiling. Slow growing or persistent organisms in natural infections are drug tolerant and sub-populations of bacilli in pulmonary tuberculosis account for the prolonged 6 month antibiotic therapy regimes. Identifying metabolic processes associated with drug tolerant persister populations in human sputum samples may therefore allow drug target identification for new, more rapidly sterilising antibiotics. Microarrays therefore provide an important experimental approach to dissect out the host-pathogen interactions in tuberculosis that hopefully will facilitate the quest for better drugs and vaccines for the treatment and prevention of tuberculosis.