Tricks to translating TB transcriptomics

Transcriptomics and other high-throughput methods are increasingly applied to questions relating to tuberculosis (TB) pathogenesis. Whole blood transcriptomics has repeatedly been applied to define correlates of TB risk and has produced new insight into the late stage of disease pathogenesis. In a novel approach, authors of a recently published study in Science Translational Medicine applied complex data analysis of existing TB transcriptomic datasets, and in vitro models, in an attempt to identify correlates of protection in TB, which are crucially required for the development of novel TB diagnostics and therapeutics to halt this global epidemic. Utilizing latent TB infection (LTBI) as a surrogate of protection, they identified IL-32 as a mediator of interferon gamma (IFNγ)-vitamin D dependent antimicrobial immunity and a marker of LTBI. Here, we provide a review of all TB whole-blood transcriptomic studies to date in the context of identifying correlates of protection, discuss potential pitfalls of combining complex analyses originating from such studies, the importance of detailed metadata to interpret differential patient classification algorithms, the effect of differing circulating cell populations between patient groups on the interpretation of resulting biomarkers and we decipher weighted gene co-expression network analysis (WGCNA), a recently developed systems biology tool which holds promise of identifying novel pathway interactions in disease pathogenesis. In conclusion, we propose the development of an integrated OMICS platform and open access to detailed metadata, in order for the TB research community to leverage the vast array of OMICS data being generated with the aim of unraveling the holy grail of TB research: correlates of protection.