Genomes Spot Tuberculosis Clusters Faster Than Traditional Methods

Tuberculosis is a disease public health teams still watch closely, even in countries with relatively few cases. In places like France, researchers have started using whole-genome sequencing (WGS) to follow Mycobacterium tuberculosis at a much finer scale than older tools allow. Charlotte Genestet and colleagues took that step further by creating an automated bioinformatics pipeline to make sense of the huge amounts of sequence data that routine WGS produces. Beginning in November 2016, every clinical M. tuberculosis isolate from eight hospitals in three cities of the Auvergne-Rhône-Alpes region underwent WGS. By July 2023 the team added an automated pipeline that combined anti-TB drug resistance prediction with an unbiased method to detect transmission clusters using single nucleotide polymorphism (SNP) distances. The goal was practical: enable clinical microbiologists to interpret WGS results directly and trigger rapid public health action rather than wait for retrospective studies. To test the system, the team collected epidemiological, microbiological and clinical data alongside the genomic data and classified close contacts as household, frequent, or occasional, creating a framework that could be used every day in routine care and surveillance.

The study linked rich clinical information to genomics. Epidemiological, microbiological and clinical data were gathered for patients diagnosed between 2016 and 2025, and index cases were classified by their level of extra-household transmission (EHT). Using the automated pipeline, which flagged resistance predictions and clustered samples based on SNP distances, the team screened 1,152 TB patients. They identified 75 clusters involving 247 patients (21.4%). WGS reliably detected resistance to first-line anti-TB drugs by leveraging the WHO mutation catalogue, allowing the lab to predict which treatments might fail before results from slower tests arrived. Because the pipeline ran in routine workflows, it generated real-time alerts for TB control centres and prompted expanded field investigations that uncovered community spillover, nosocomial transmissions, and even a school outbreak. Surprisingly, classic signs used to judge contagiousness—smear results, cavitary disease—were not linked to higher EHT. Instead, lower TB severity indices and longer duration of symptoms correlated with greater extra-household transmission, a finding only visible when genomic and clinical data were combined.

The work shows how systematic WGS, when paired with an automated, easy-to-use pipeline, can change the pace and precision of TB control. By giving clinical microbiologists direct, actionable results for both drug resistance prediction and transmission clustering, the approach supports faster field responses and more targeted investigations. The authors argue this is a time- and cost-effective all-in-one solution for routine TB management: it streamlines species identification, resistance detection, and outbreak detection into a single workflow. Because this study implemented the pipeline across multiple hospitals and used it to trigger real-time public health action, it moves WGS from a retrospective research tool to a proactive element of surveillance. The findings also suggest national and international guidelines should evolve to support broader integration of WGS-based pipelines. The project was supported by SHAPE-Med@Lyon, funded by the French National Research Agency under the France 2030 program (ANR-22-EXES-0012), indicating that coordinated investment can make routine genomic surveillance feasible.