Bacterial identification in clinical samples – comparing Nanopore and Illumina sequencing of the 16S rRNA and rpoB genes

Abstract Background Recent improvements in the sequencing accuracy of the Oxford Nanopore Technologies (ONT) platform have increased its suitability for clinical microbiology applications, including broad-range amplification and sequencing of the 16S ribosomal RNA (rRNA) gene directly from clinical samples. This approach has a strong potential to improve diagnosis of bacterial infections. However, even full-length 16S rRNA sequencing may lack sufficient resolution to distinguish closely related species, a limitation exacerbated by the remaining ONT error rate. The rpoB gene represents a high-resolution bacterial phylogenetic marker that offers improved species-level discrimination. We evaluated a strategy combining ONT sequencing of the V1-V3 segment of the 16S rRNA gene and a segment of the rpoB gene. In addition, we assessed two novel pipelines designed specifically for the analysis of ONT amplicon data in clinical microbiology, where knowing the associated confidence level for a given identification is crucial, as is the ability to detect organisms that may not be represented in reference databases. Methods A staggered mock community and 25 abdominal abscess samples were investigated using ONT sequencing of the 16S rRNA and rpoB genes, with Illumina sequencing of the same genes used as a comparator method to validate the analytical accuracy of the approach. Results rpoB ONT sequencing achieved species-level identification in 91.2% of detections, as compared to 62.8% with 16S rRNA ONT sequencing. However, 16S rRNA ONT sequencing obtained a higher sensitivity of 93%, as compared to 83.7% for rpoB ONT sequencing. Combining 16S rRNA and rpoB ONT sequencing yielded a sensitivity of 98.8% with 89% species-level discrimination. For both gene targets, ONT sequencing paralleled Illumina sequencing. Conclusions Combining ONT sequencing with high-sensitivity broad-range amplification of the 16S rRNA gene and amplification of the rpoB gene for enhanced taxonomic resolution enables accurate detection and species-level identification of bacteria directly from clinical samples. This approach represents a rapid, flexible and cost-efficient tool for culture-independent diagnostics in clinical microbiology.