TB-Genaly: A multicenter evaluated whole genome sequencing pipeline for tuberculosis drug resistance assessment and...

Whole-genome sequencing (WGS) enables comprehensive characterization of Mycobacterium tuberculosis (MTB), but its routine clinical use is limited by the lack of standardized and multicenter-validated analytical workflows.

We developed TB-Genaly, a user-friendly and secure bioinformatics pipeline for MTB WGS analysis. Performance of drug-resistance prediction was evaluated using a multicenter, population-based cohort comprising 461 clinical MTB isolates from four centers, each with paired phenotypic drug susceptibility testing (pDST) results. Sensitivity and specificity were calculated to assess concordance between genotypic predictions and pDST.

TB-Genaly demonstrated high predictive performance for anti-tuberculosis drug resistance. For first-line drugs, it achieved sensitivity and specificity of 95.8% and 92.6% for rifampicin, 95.5% and 92.4% for isoniazid, 91.8% and 99.4% for pyrazinamide, and 87.3% and 92.7% for ethambutol. For second-line drugs, sensitivity and specificity ranged from 87.8% to 94.7% and 97.5% to 98.0%. The mean values of sensitivity and specificity for predicting MDR-TB were 95.4% (95%CI 92.3%-97.7%) and 93.5% (95%CI 91.1-95.5%), respectively. Receiver operating characteristic (ROC) curve analysis further confirmed excellent discriminatory power, with area under the curve (AUC) values of 0.999, 0.997, 0.995 and 0.969 for the four agents. Additionally, TB-Genaly supports automated generation of standardized clinical reports and transmission network visualizations.

TB-Genaly provides accurate, standardized WGS-based drug-resistance prediction across multiple centers. Its usability and comprehensive reporting facilitate integration into clinical diagnostics and public health surveillance, highlighting its potential to support clinical decision-making and tuberculosis control efforts.