First insight of characteristics and prediction of Mycobacterium tuberculosis drug resistance by whole genome sequencing in Fujian Province, China

To delineate the molecular characteristics and assess the drug resistance prediction value of whole-genome sequencing (WGS) for Mycobacterium tuberculosis (MTB) in Fujian Province, China. Representative MTB surveillance isolates underwent phenotypic drug susceptibility testing (pDST), followed by WGS(gDST), with resistance mutation analysis conducted on the SAM-TB platform. The rate of mono-resistance, poly-resistance, and multidrug-resistance (MDR) identified by pDST were 38.67%, 12.00% and 10.00% respectively. Lineage analysis revealed that L2 (60.0%) and L4 (38.7%) predominated. No statistically significant difference in drug-resistance rates among lineages (χ² = 4.85, P = 0.183). Total drug resistance rate by gDST was 67.03%, harboring 126 resistance-associated mutations. The most frequent mutations were KatG S315T (22.22%), rpoB S450L (10.32%), rpsL K43R (12.70%), embB M306I (4.76%), and gyrA D94G (7.94%). Using pDST as the reference standard, the sensitivity, specificity and concordance rate of gDST to RFP, INH, FQs, EMB, SM was 78.95%, 78.05%, 76.47%, 53.85%, 45.71%, and 98.47%, 94.50%, 93.98%, 97.08%, 95.65%, and 0.81, 0.74, 0.64, 0.55, 0.48, respectively. WGS provided rapid, comprehensive genotyping of MTB isolates, revealing the predominant lineages L2 and L4 in Fujian Province and reliably predicting resistance to RFP, INH, and FQ.