深センにおける非結核性抗酸菌同定における標的ナノポアシーケンシングの性能評価：比較研究

This study aims to analyze the performance differences between targeted nanopore sequencing, Sanger sequencing, and metagenomic sequencing in comparatively identifying non-tuberculous mycobacteria (NTM) species. Additionally, it explores the clinical application potential of targeted nanopore sequencing for identifying NTM clinical isolates in the Shenzhen region.

This retrospective study collected a total of 50 suspected NTM isolates from drug-resistant tuberculosis surveillance across 10 districts in Shenzhen, China, between December 2024 and June 2025. The species of the NTM isolates were initially identified using fluorescence PCR probe melting curve analysis. Genomic DNA was extracted from all 50 isolates, and species identification was performed using targeted nanopore sequencing (tNS), metagenomic sequencing (mNGS), and Sanger sequencing. The Jaccard similarity index, Kappa coefficient for classification consistency, and F1 score for model performance were calculated to evaluate the concordance among the three sequencing methods and assess the detection performance of targeted nanopore sequencing in NTM species identification.

The most frequently detected NTM species by tNS, mNGS, and Sanger sequencing were M.abscessus and M.fortuitum, while M. tuberculosis was predominantly identified through mNGS results. Among the 50 suspected NTM samples, 18 (36%) showed complete concordance between tNS, mNGS, and Sanger sequencing, with the highest agreement observed between mNGS and tNS (28 samples, 56%). The final species identification reference results for the 50 samples were confirmed through a comprehensive evaluation using the Jaccard similarity coefficient, precision, and recall. Based on reference results, the F1 scores for tNS, mNGS, and Sanger sequencing were 0.927, 0.896, and 0.543, respectively. The tNS exhibited the highest concordance with the reference results, outperforming the other two methods.

tNS represents a preferred auxiliary methodology for clinical identification of NTM isolates in Shenzhen, China, with identification results optimally validated through integration with mNGS findings. This study provides strong support for the application of tNS technology for NTM species identification.