In March 2020, SARS-CoV-2 was spreading around the globe and a first wave of infections hit Australia. COVID-19 had been declared a pandemic and our team was eager to use our expertise in genomics to contribute to the global response.

The first genome sequence of SARS-CoV-2, isolated from an early case in Wuhan, was published in January 2020. As the virus spread, different labs across the world began sequencing the genome of the virus in local cases from their areas. Whole genome sequencing (WGS) can be a powerful tool in virus surveillance and response, as it helps to define transmission networks and identify the sources of new outbreaks. In assisting with the current pandemic, this information could be used to help identify behaviours that spread COVID-19 and ultimately shape public policy.

Our study compares two sequencing platforms that can be used to sequence the SARS-CoV-2 genome: Illumina and Oxford Nanopore Technologies (ONT), which offer short and long sequencing reads of DNA, respectively. Illumina is the dominant platform in the field and has been preferred by most researchers and, especially, public health initiatives during the COVID-19 pandemic, due to its high per-read accuracy. However, ONT has a number of advantages for viral genomics that should not be overlooked. ONT devices are cheap, portable, scalable and require minimal laboratory infrastructure, potentially enabling access to users in remote or under-resourced areas. Further, results can be obtained rapidly (less than 24 h), which is vital to ensure the effectiveness of contact tracing.

Portable ONT devices: Flongle (left) and MinION (right)

Nonetheless, there has a been a reasonable reluctance in adopting ONT sequencing for SARS-CoV-2 surveillance due to concerns around sequencing accuracy. This is rational, as accurate sequence determination is critical for defining outbreaks, especially in a virus with a low mutation rate like SARS-CoV-2, where any errors can have a large confounding impact. The ability to accurately define disease clusters using genomics has been critical to the success of Australia and other countries in controlling the spread of SARS-CoV-2. Given this, it was important to us to undertake rigorous validation experiments to assess the suitability of ONT for use in SARS-CoV-2 surveillance.

Live monitoring sequencing coverage through RAMPART

This work was performed as part of a fruitful collaboration between academic research (the Garvan Institue of Medical Research and the Kirby Institute) and public health laboratories (NSW Health Pathology), as has so much of the essential COVID-19 research carried out around the world. To evaluate the analytical validity of ONT, we performed viral WGS of 157 SARS-CoV-2 clinical specimens in parallel on both Illumina and ONT platforms, with the former following a gold-standard, clinically accredited workflow for pathogen sequencing.

Our study demonstrated that ONT sequencing is suitable for SARS-CoV-2 genomic surveillance, as it accurately detects variants in SARS-CoV-2 specimens with >99% sensitivity and precision, above a minimum ~60-fold sequencing coverage. Detection of short insertions and deletions was challenging, however, ONT was well suited for detection of larger deletions, of which we found a surprising variety in the specimens analysed.

We hope that our study will help to remove an important barrier preventing widespread adoption of ONT sequencing for SARS-CoV-2 WGS. The advantages ONT can deliver in cost, portability and turnaround time may allow it to serve as a useful tool in ongoing COVID-19 response strategies.

Read the full paper at https://www.nature.com/articles/s41467-020-20075-6