Exploring the potential of Oxford Nanopore Technologies sequencing for Mycobacterium tuberculosis sequencing: An assessment of R10 flowcells and V14 chemistry

• Published in: PloS one 2024-06, Vol.19 (6), p.e0303938
• Main Authors: Dippenaar, Anzaan, Costa Conceição, Emilyn, Wells, Felicia, Loubser, Johannes, Mann, Brendon, De Diego Fuertes, Miguel, Rennie, Vincent, Warren, Robin Mark, Van Rie, Annelies
• Format: Article
• Language: English
• Subjects: Accuracy; Automation; Bar codes; Biology and Life Sciences; Capital costs; Chemistry; Data analysis; Deoxyribonucleic acid; DNA; DNA sequencing; DNA, Bacterial - genetics; Drug resistance; Gene Library; Gene sequencing; Genetic testing; Genome, Bacterial; Genomes; Genomics; High-Throughput Nucleotide Sequencing - methods; Humans; Impurities; Libraries; Medicine and Health Sciences; Mycobacterium tuberculosis; Mycobacterium tuberculosis - genetics; Nanopore Sequencing - methods; Nanopores; Nucleotide sequence; Nucleotide sequencing; Performance assessment; Performance evaluation; Research and analysis methods; Research ethics; Sequence Analysis, DNA - methods; Technology assessment; Tuberculosis; Tuberculosis - diagnosis; Tuberculosis - microbiology; Whole genome sequencing; Whole Genome Sequencing - methods
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0303938

Oxford Nanopore Technologies (ONT) sequencing is a promising technology. We assessed the performance of the new ONT R10 flowcells and V14 rapid sequencing chemistry for Mtb whole genome sequencing of Mycobacterium tuberculosis (Mtb) DNA extracted from clinical primary liquid cultures (CPLCs). Using the recommended protocols for MinION Mk1C, R10.4.1 MinION flowcells, and the ONT Rapid Sequencing Kit V14 on six CPLC samples, we obtained a pooled library yield of 10.9 ng/μl, generated 1.94 Gb of sequenced bases and 214k reads after 48h in a first sequencing run. Only half (49%) of all generated reads met the Phred Quality score threshold (>8). To assess if the low data output and sequence quality were due to impurities present in DNA extracted directly from CPLCs, we added a pre-library preparation bead-clean-up step and included purified DNA obtained from an Mtb subculture as a control sample in a second sequencing run. The library yield for DNA extracted from four CPLCs and one Mtb subculture (control) was similar (10.0 ng/μl), 2.38 Gb of bases and 822k reads were produced. The quality was slightly better with 66% of the produced reads having a Phred Quality >8. A third run of DNA from six CPLCs with bead clean-up pre-processing produced a low library yield (±1 Gb of bases, 166k reads) of low quality (51% of reads with a Phred Quality score >8). A median depth of coverage above 10× was only achieved for five of 17 (29%) sequenced libraries. Compared to Illumina WGS of the same samples, accurate lineage predictions and full drug resistance profiles from the generated ONT data could not be determined by TBProfiler. Further optimization of the V14 ONT rapid sequencing chemistry and library preparation protocol is needed for clinical Mtb WGS applications.