QuantTB - a method to classify mixed Mycobacterium tuberculosis infections within whole genome sequencing data

Area of Science:

• Genomics
• Microbiology
• Bioinformatics

Background:

• Tuberculosis (TB) diagnosis and treatment are complicated by mixed infections of Mycobacterium tuberculosis and antibiotic heteroresistance.
• Current detection methods for mixed TB infections lack sensitivity and resolution.
• Whole genome sequencing (WGS) offers a more sensitive approach to analyze genetic differences within M. tuberculosis samples.

Purpose of the Study:

• To present QuantTB, a novel method for identifying and quantifying individual M. tuberculosis strains in WGS data.
• To enable accurate estimation of TB infection multiplicity and detection of heteroresistance.
• To differentiate between TB relapse and reinfection events.

Main Methods:

• QuantTB utilizes single nucleotide polymorphism (SNP) markers to identify the combination of strains present in a sample.
• The method analyzes allelic variation in WGS data to determine strain composition.
• Outputs include identified strains, their relative abundances, and predicted drug resistance-conferring mutations.

Main Results:

• QuantTB demonstrates high resolution, differentiating communities with as few as 25 SNPs and detecting strains at 1× coverage.
• Simulated data showed QuantTB outperformed existing metagenomic tools in strain detection and multiplicity quantification.
• In a clinical study, QuantTB accurately detected mixed infections and reinfections in 50 paired isolates.

Conclusions:

• QuantTB accurately determines infection multiplicity and identifies heteroresistance patterns in TB.
• The method aids in differentiating relapse from reinfection and clarifying transmission events.
• QuantTB is a valuable resource for clinicians and researchers, outperforming existing tools even with low-coverage samples.