Host Transcriptomics as a Tool to Identify Diagnostic and Mechanistic Immune Signatures of Tuberculosis

Front Immunol. 2019 Feb 19:10:221. doi: 10.3389/fimmu.2019.00221. eCollection 2019.


Tuberculosis (TB) is a major infectious disease worldwide, and is associated with several challenges for control and eradication. First, more accurate diagnostic tools that better represent the spectrum of infection states are required; in particular, identify the latent TB infected individuals with high risk of developing active TB. Second, we need to better understand, from a mechanistic point of view, why the immune system is unsuccessful in some cases for control and elimination of the pathogen. Host transcriptomics is a powerful approach to identify both diagnostic and mechanistic immune signatures of diseases. We have recently reported that optimal study design for these two purposes should be guided by different sets of criteria. Here, based on already published transcriptomics signatures of tuberculosis, we further develop these guidelines and identify additional factors to consider for obtaining diagnostic vs. mechanistic signatures in terms of cohorts, samples, data generation and analysis. Diagnostic studies should aim to identify small disease signatures with high discriminatory power across all affected populations, and against similar pathologies to TB. Specific focus should be made on improving the diagnosis of infected individuals at risk of developing active disease. Conversely, mechanistic studies should focus on tissues biopsies, immune relevant cell subsets, state of the art transcriptomic techniques and bioinformatics tools to understand the biological meaning of identified gene signatures that could facilitate therapeutic interventions. Finally, investigators should ensure their data are made publicly available along with complete annotations to facilitate metadata and cross-study analyses.

Keywords: host immune response; human immunology; infectious diseases; transcriptomics; tuberculosis.

Publication types

  • Research Support, N.I.H., Extramural
  • Review

MeSH terms

  • Animals
  • Computational Biology / methods
  • Gene Expression Profiling / methods
  • Humans
  • Transcriptome / genetics*
  • Transcriptome / immunology*
  • Tuberculosis / genetics*
  • Tuberculosis / immunology*