Rapid sequencing of RNA virus genomes

Enteroviruses (EV) are single‑stranded RNA viruses that cause infections in millions of people worldwide each year. The clinical manifestations of infection are broad and can range from benign sore throats to more serious conditions such as bronchitis and pneumonia1. PCR‑based detection is the most commonly used method for the routine identification of viruses of clinical importance. Yet, point mutations and recombination events, which occur frequently in viral genomes, can potentially lead to false negative results when PCR‑based techniques are used. In addition, current detection techniques require culture of the virus, which typically takes 5–10 days — significantly delaying time to result1. To overcome these challenges, Dr. Alban Ramette and his team at the University of Bern, Switzerland, assessed the capabilities of nanopore cDNA and direct RNA sequencing to obtain complete genome sequences of Enteroviruses from clinical research samples1.

Using viral cDNA derived from a cultured sample, it was possible to obtain a sequence consensus accuracy of 98.8% within just a few minutes of starting the sequencing run. Polishing the sequence using the nanopolish2 tool subsequently increased the accuracy to 99.8% (Figure 8).

Direct RNA sequencing, which does not require reverse transcription or amplification, is particularly useful in time‑ critical applications such as pathogen characterisation. The sample preparation method employed at the University of Bern for direct RNA sequencing took just 5.5 hours, compared with 23 hours for the cDNA approach1.

In an effort to further streamline the procedure, the direct RNA sequencing sample was obtained from a stool sample with no prior viral culturing. Although the sample preparation methodology yielded only 140 ng of RNA — some way short of the recommended 500 ng starting amount — the team were able to sequence and assemble a complete Coxsackievirus genome based on 11 reads, whose lengths were generally greater than 1000 bases. Furthermore, almost the entire genome was shown to be covered in a single read of 7208 bases (i.e. only 25 and 109 bases were missing at the 5’ and 3’ end of the sequence as compared to the reference Coxsackievirus genome sequence.

Summarising his work, Alban commented:

"cDNA sequencing provides sensitivity equal to that of PCR assay, while direct RNA sequencing offers the fastest sample to answer turnaround time —  with no amplification or reverse transcription biases"1.

Figure: Nanopore sequencing of cDNA obtained from a cultured Coxsackievirus isolate delivered a 98.8% consensus accuracy within a few minutes of sequencing, and this consensus accuracy value did not improve significantly with up to 16 hours of further sequencing. The use of nanopolish on the first 5‑10,000 reads, obtained within 10 minutes of sequencing, increased this accuracy to 99.8%. Image courtesy of Dr. Alban Ramette, University of Bern, Switzerland.

This case study is taken from the Microbiology white paper.

  1. Ramette, A. Applications of nanopore sequencing to whole genome sequencing of human viruses in the clinical setting. [online] Available at: https://nanoporetech.com/resource-centre/applications-nanopore-sequencing-technologies-whole-genome-sequencing-human-viruses [Accessed: 13 February 2018].
  2. GitHub. Nanopolish. Available at: https://github.com/jts/nanopolish [Accessed: 25 February 2018].