UC San Diego

Mycobacterium Tuberculosis is an infectious disease and a serious public health concern due to the organism’s adaptive transcriptional response to environmental stresses via the transcriptional regulatory network (TRN). (Galagan et al., 2013) While many studies seek to better characterize specific portions of the M. tuberculosis TRN, a systems level characterization and analysis of interactions between the controlling transcription factors has yet to be done. Here, we utilize unsupervised machine learning to compartmentalize and describe the transcription factors and regulatory interactions of M. tuberculosis’s TRN, allowing us to create a model for how the bacterium responds to environmental stresses.(Boot et al., 2018; Serafini et al., 2019) By applying Independent Component Analysis (ICA) to over 650 transcriptomic samples, we obtained 80 independently regulated gene sets known as “I-modulons” that help to explain the variance in the organisms transcriptional response. This ICA structure helps to elucidate the function of previously undescribed regulons, as well as the transcriptional shifts that occur during environmental changes such as shifting carbon sources, oxidative stress, and virulence events. Additionally, this analysis has also uncovered an inherent cluster of transcriptional regulons that connects several important metabolic systems, including lipid catalysis, cholesterol catalysis, and sulfur metabolism. This system-wide analysis of the organism’s TRN can help inform future research on effective ways to study and manipulate the transcriptional regulation of M. Tuberculosis.