The abstract is read at the publisher; the summary is JClub's.
catomatic-1, built from 39,358 Mycobacterium tuberculosis isolates, algorithmically classified 1,329 resistance-associated variants, with 440 for pyrazinamide and 5 for linezolid.
- Why it matters: WHO resistance catalogues support WGS-based susceptibility prediction but rely on data that are not fully released and are hard to interpret, limiting reproducible surveillance and diagnostics.
- What they did: The automated catomatic pipeline used a binomial test to link variants with resistance or susceptibility, then ran sensitivity analyses to optimize parameters per drug and benchmarked against WHOv1 on 14,380 independent isolates.
- The result: On the independent Validation Dataset, catomatic-1 matched WHOv1 performance across drugs, achieving sensitivities above 88% for first-line agents and improving sensitivity by detecting resistant variants with low read support.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.