Simplified Point-of-Care Full SARS-CoV-2 Genome Sequencing Using Nanopore Technology

Abstract

The scale of the ongoing SARS-CoV-2 pandemic warrants the urgent establishment of a global decentralized surveillance system to recognize local outbreaks and the emergence of novel variants of concern. Among available deep-sequencing technologies, nanopore-sequencing could be an important cornerstone, as it is mobile, scalable, and cost-effective. Therefore, streamlined nanopore-sequencing protocols need to be developed and optimized for SARS-CoV-2 variants identification. We adapted and simplified existing workflows using the 'midnight' 1200 bp amplicon split primer sets for PCR, which produce tiled overlapping amplicons covering almost the entire SARS-CoV-2 genome. Subsequently, we applied Oxford Nanopore Rapid Barcoding and the portable MinION Mk1C sequencer combined with the interARTIC bioinformatics pipeline. We tested a simplified and less time-consuming workflow using SARS-CoV-2-positive specimens from clinical routine and identified the CT value as a useful pre-analytical parameter, which may help to decrease sequencing failures rates. Complete pipeline duration was approx. 7 h for one specimen and approx. 11 h for 12 multiplexed barcoded specimens. The adapted protocol contains fewer processing steps and can be completely conducted within one working day. Diagnostic CT values deduced from qPCR standardization experiments can act as principal criteria for specimen selection. As a guideline, SARS-CoV-2 genome copy numbers lower than 4 × 10⁶ were associated with a coverage threshold below 20-fold and incompletely assembled SARS-CoV-2 genomes. Thus, based on the described thermocycler/chemistry combination, we recommend CT values of ~26 or lower to achieve full and high-quality SARS-CoV-2 (+)RNA genome coverage.

Keywords: (+)RNA genome sequencing; COVID-19 surveillance; variant-of-concern (VOC); variant-of-interest (VOI).

Figure 1

Nanopore sequencing options (2–6) compatible with diagnostic qPCR pipelines (1). The pyrosequencing option in the path (6) requires biotin-tagged primers for qPCR [10].

Figure 2

Surveillance using nanopore sequencing (A), and effects of specimens dilution on the SARS-CoV-2 genome coverage (B), whereby viral copy numbers were 1.28 × 10⁶ (B); 5.12 × 10⁸ (C); 2.56 × 10⁸ (D); 4 × 10⁶ (E); 2 × 10⁶ (F); 1 × 10⁶ (G); 5 × 10⁵ (H). In (A) increasing barcode numbers (X-axis) correspond to decreasing viral titers.

Figure 3

Comparison of the consensus sequences returned by ‘Nanopolish’ and ‘Medaka’ algorithm and the merged consensus for a nanopore sequencing run corresponding to a viral load of 1.28 × 10⁶. (A) Alignment of the ‘Nanopolish’ and ‘Medaka’ consensus as well as the merged consensus on the SARS-CoV-2 reference genome by the Nextstrain program. The merged consensus conserved all of the high-quality mutations that mapped to the known variant 20I (shown in color or light grey), while most non-matching mutations (in dark grey, likely sequencing errors) of the ‘Nanopolish’ or ‘Medaka’ consensus were lost; (B) Comparison of variant solving by ‘Nanopolish’, ‘Medaka’ and our code. The multiple alignments were performed by MAFFT online tool with the three consensus sequences and the SARS-CoV-2 reference genome (MN908947.3). Contrary to the ‘Nanopolish’ and ‘Medaka’ consensus, the merged consensus solved the entire region and led to accurate variant calling; (C) Clade mapping and phylogenetic distances calculated by Nextstrain for the three consensus sequences.

Figure 4

Comparison on the evolutionary distance calculation, when data from the same specimen was processed by 2 different bioinformatics pipelines. Phylogenetic tree visualization was done using the Nextstrain open-source platform for pathogen genome data analyses [14]. We used serial dilutions of identical specimens (A), or a selection of samples from different individuals from clinical routine (B) for phylogenetic analyses. Below the trees, a magnification of clade 20I is shown, wherein most specimens are grouped (A,B).

Figure 5

Comparison of a rapid SARS-CoV-2 whole (+)RNA genome nanopore sequencing pipeline [11] with the novel simplified workflow, whose main achievements are: 1. Purified RNA from SARS-CoV-2-positive patients can be directly taken from residual diagnostic specimens in 96-deepwell-plates; 2. cDNA synthesis and successive multiplex PCR with 2 primer pools can be performed in single tube reactions. Since cDNA synthesis is primed by SARS-CoV-2-specific primers for 1200 bp amplicon amplification, there is no need for use of unspecific hexanucleotide priming; 3. Onboard Guppy basecalling with the Oxford Nanopore MinION Mk1C device and implementation of the interARTIC led to a further reduction of working steps and hands-on time.