Although tuberculosis remains a substantial global threat, the mechanisms that enable mycobacterial persistence and replication within the human host are ill defined. This study represents the first genome-wide expression analysis of Mycobacterium tuberculosis from clinical lung samples, which has enabled the identification of M. tuberculosis genes actively expressed during pulmonary tuberculosis. To obtain optimal information from our DNA array analyses, we analyzed the differentially expressed genes within the context of computationally inferred protein networks. Protein networks were constructed using functional linkages established by the Rosetta stone, phylogenetic profile, conserved gene neighbor, and operon computational methods. This combined approach revealed that during pulmonary tuberculosis, M. tuberculosis actively transcribes a number of genes involved in active fortification and evasion from host defense systems. These genes may provide targets for novel intervention strategies.