An RNA-seq Based Machine Learning Approach Identifies Latent Tuberculosis Patients With an Active Tuberculosis...

Area of Science:

• Immunology
• Genomics
• Infectious Diseases

Background:

• Accurate identification of active tuberculosis (TB) and latent TB infection (LTBI) is crucial.
• Predicting LTBI individuals at high risk for progression to active TB requires better understanding of host immune responses.

Purpose of the Study:

• To build a class-prediction model using gene expression profiles to identify TB infection.
• To identify LTBI individuals with an increased risk of developing active TB.

Main Methods:

• Gene expression profiling of active TB patients and their LTBI and uninfected contacts.
• Development of a class-prediction model for TB infection identification.
• Combination of the prediction model with immunological parameter measurements.

Main Results:

• Identification of key genes and metabolic pathways involved in the immune response to TB.
• Discovery of a 'TB-like' subset of LTBI individuals with transcriptional and immunological profiles indicating higher active TB risk.

Conclusions:

• Gene expression profiling combined with immunological data can identify individuals at higher risk of active TB.
• Further longitudinal studies in endemic areas are needed to validate this novel approach for risk stratification.