WGS ONT pipeline for TB diagnostics

Background: Drug resistant tuberculosis is a serious concern and delays in correct diagnosis can hinder an already arduous drug treatment regimen. Whole genome sequencing provides more detail than standard molecular tests (e.g. GeneXpert) and phenotypic testing, but commonly used platforms (such as the Illumina MiSeq) are expensive to set up. The Oxford Nanopore Technologies (ONT) MinION is a potentially more cost effective, user-friendly option. Aims: We aimed to optimise each step in a tuberculosis whole genome sequencing diagnostic pipeline using the ONT MinION, balancing user-friendliness, cost-effectiveness and time to results, whilst ensuring accuracy. Methods: Culture times, media, nucleic acid precipitation and spin column CTAB DNA extraction protocols were compared to optimise DNA yield. ONT library preparation kits were compared, to identify the most cost- and time-effective pipeline components. Bioinformatic tools for basecalling and analysis were evaluated to find the most accurate resistance SNP and lineage predictor. Results: A column CTAB DNA extraction method was combined with the Rapid Barcoding 96 library preparation kit, high accuracy basecalling and data analysis using TB-Profiler. Our pipeline was 76% concordant with phenotypic antibiotic susceptibility test results. The pipeline was overall 88% (15/17) concordant with Illumina data. It was concordant for 16/17 (94%) isolates for lineage (one isolate was mixed lineage) and for 16/17 (94%) isolates for resistance SNPs (a SNP was picked up by ONT that was missed by Illumina). The cost per sample was less than 130 USD and time to diagnosis was around four weeks (including culture). Discussion: The optimised TB sequencing pipeline developed does not require bioinformatics experience, takes less time than phenotypic drug sensitivity testing and the cost per sample is comparable with Illumina and culture. These features will make it an important tool to be incorporated into routine drug-resistant TB diagnostic pipelines and larger scale TB drug resistance surveillance.