Evaluation of Nanopore sequencing for Mycobacterium tuberculosis drug susceptibility testing and outbreak investigation: a genomic analysis

Abstract

Background: Mycobacterium tuberculosis whole-genome sequencing (WGS) has been widely used for genotypic drug susceptibility testing (DST) and outbreak investigation. For both applications, Illumina technology is used by most public health laboratories; however, Nanopore technology developed by Oxford Nanopore Technologies has not been thoroughly evaluated. The aim of this study was to determine whether Nanopore sequencing data can provide equivalent information to Illumina for transmission clustering and genotypic DST for M tuberculosis.

Methods: In this genomic analysis, we analysed 151 M tuberculosis isolates from Madagascar, South Africa, and England, which were collected between 2011 and 2018, using phenotypic DST and matched Illumina and Nanopore data. Illumina sequencing was done with the MiSeq, HiSeq 2500, or NextSeq500 platforms and Nanopore sequencing was done on the MinION or GridION platforms. Using highly reliable PacBio sequencing assemblies and pairwise distance correlation between Nanopore and Illumina data, we optimise Nanopore variant filters for detecting single-nucleotide polymorphisms (SNPs; using BCFtools software). We then used those SNPs to compare transmission clusters identified by Nanopore with the currently used UK Health Security Agency Illumina pipeline (COMPASS). We compared Illumina and Nanopore WGS-based DST predictions using the Mykrobe software and mutation catalogue.

Findings: The Nanopore BCFtools pipeline identified SNPs with a median precision of 99·3% (IQR 99·1-99·6) and recall of 90·2% (88·1-94·2) compared with a precision of 99·6% (99·4-99·7) and recall of 91·9% (87·6-98·6) using the Illumina COMPASS pipeline. Using a threshold of 12 SNPs for putative transmission clusters, Illumina identified 98 isolates as unrelated and 53 as belonging to 19 distinct clusters (size range 2-7). Nanopore reproduced 15 out of 19 clusters perfectly; two clusters were merged into one cluster, one cluster had a single sample missing, and one cluster had an additional sample adjoined. Illumina-based clusters were also closely replicated using a five SNP threshold and clustering accuracy was maintained using mixed Illumina and Nanopore datasets. Genotyping resistance variants with Nanopore was highly concordant with Illumina, having zero discordant SNPs across more than 3000 SNPs and four insertions or deletions (indels), across 60 000 indels.

Interpretation: Illumina and Nanopore technologies can be used independently or together by public health laboratories performing M tuberculosis genotypic DST and outbreak investigations. As a result, clinical and public health institutions making decisions on which sequencing technology to adopt for tuberculosis can base the choice on cost (which varies by country), batching, and turnaround time.

Funding: Academy for Medical Sciences, Oxford Wellcome Institutional Strategic Support Fund, and the Swiss South Africa Joint Research Award (Swiss National Science Foundation and South African National Research Foundation).

Figure 1

Recall and precision of SNPs from the Illumina COMPASS pipeline, and the Nanopore BCFtools pipeline with a cumulative selection of filters Each point represents a single isolate with a PacBio assembly. The midline in each box plot is the median, the upper and lower bounds of each box indicates the span of the quartiles of the data (ie, IQR), and the whiskers extend 1·5 times the IQR. #nofilter is BCFtools with no filtering of variants. Moving right from #nofilter, each box accumulates a new filter plus the previous ones. Each filter describes the criterion for removing an SNP. -QUAL<25 removes SNPs with a quality score less than 25; -FRS<90% removes SNPs whereby less than 90% of reads support the called allele; -FED<20% removes SNPs with read depth below 20% of the isolate's median depth; -DP<5 removes SNPs with less than five reads at the position; -SR<1% removes SNPs with less than 1% of read depth on either strand; -MQ<30 removes SNPs with a mapping quality below 30; -VDB<1e–5 removes SNPs with a variant distance bias less than 0·00001. SNP=single-nucleotide polymorphism.

Figure 2

Pairwise SNP distance relationship between Illumina (COMPASS) and Nanopore (BCFtools) data Each point represents the SNP distance between two isolates. The dashed line shows the identity line (ie, y=x). The isolate pairs shown are all pairs whereby the COMPASS distance is 20 or less. The red area and points indicate pairs with a Nanopore distance of more than 12 but an Illumina distance of 12 or less. These pairs are deemed false negative connections. The red area with stripes indicates pairs that are false negative connections at an Illumina threshold of five (Nanopore threshold of six), but not when the threshold is expanded to 12. These pairs are shown as square points. The grey area and points are the inverse—ie, false positive connections. Thus, the grey striped area shows pairs of samples that are false positive connections at an Illumina threshold of five (Nanopore threshold six), but not when the threshold is expanded to 12. SNP=single-nucleotide polymorphism.

Figure 3

Agreement of Illumina and Nanopore transmission clustering with thresholds of five for Illumina and six for Nanopore (A) and thresholds of 12 for both technologies (B) The expected (Illumina COMPASS) clusters are shown on the left and the Nanopore BCFtools clustering is shown on the right. The title of each panel indicates the SNP threshold used for clustering. Nodes are coloured and numbered according to their Illumina cluster membership. Isolates clustered by Nanopore and not clustered (singletons) by Illumina are represented as boxes and are named S. Clusters are horizontally aligned and connected with black lines; however, the order of nodes and the length of edges have no significance. Each Nanopore panel shows the SACR, SACP, and XCR value (with the raw numbers in parentheses) with respect to the Illumina clustering. SACP=sample-averaged cluster precision. SACR=sample-averaged cluster recall. SNP=single-nucleotide polymorphism. XCR=excess clustering rate.

Figure 4

Simulating heterogeneous datasets with varying proportions of Nanopore and Illumina genomic data The different thresholds indicate the cutoff for defining isolates as part of a cluster. The y-axis depicts the SACP, SACR, or 1–XCR distributions over all simulation runs. For each ratio and threshold combination we ran 1000 simulations whereby the Nanopore and Illumina data were randomly split into the relevant ratio (eg, 1:9 means one Nanopore isolate for every nine Illumina isolates) and clusters were defined based on the relevant threshold. The titles for each subplot indicate the SNP threshold used when comparing Illumina, Nanopore, or mixed-technology isolate pairs. Dashed horizontal lines show the median and quartiles. SACP=sample-averaged cluster precision. SACR=sample-averaged cluster recall. SNP=single-nucleotide polymorphism. XCR=excess clustering rate.

Figure 5

Number of resistant and susceptible phenotypes correctly predicted by Mykrobe from Illumina and Nanopore whole-genome sequencing data For resistant phenotypes (left plot) the bars show for each drug the breakdown of resistance predictions in samples that are phenotypically resistant; false negatives (wrongly calling a sample as susceptible) are coloured red, and the rest of the bar (true positives) is coloured to show the technology (Nanopore or Illumina). For susceptible phenotypes (right plot) the bars show for each drug the breakdown of resistance predictions in samples that are phenotypically susceptible; false positives (wrongly calling a sample as resistant) are coloured purple, and the rest of the bar (true negatives) is coloured to show the technology (Nanopore or Illumina).