Oxford Nanopore technology update: CTO Clive G Brown unveils latest sequencing chemistry with highest performance to date, Short Fragment Mode and latest methylation performance evaluations
Wed 30th March 2022
Clive G Brown, CTO, Oxford Nanopore Technologies, reviews some of the latest technology updates during a webinar today. Highlights include performance and base modification calling improvements, as well as updates on the recently-released Short Fragment Mode (SFM) and a new basecaller.
- Latest chemistry enables 99.3% raw read accuracy at high data yields; new Kit 14 and flow cells containing R10.4.1 pore will be provided as an upgrade to the current Kit 12 chemistry and R10.4 pore, and eventually replace the R9 chemistry versions
- Remora, a tool that offers real time methylation analysis is shown to outperform bisulfite sequencing
- Short Fragment Mode is now allowing users to choose fragments from 20 bases long to millions; methylation analysis of shorter fragments enabled
- New basecaller, Dorado, introduced to support high performance and broader hardware compatibility.
Updates in sequencing chemistry to enhance performance
Nanopore sequencing enables the analysis of native DNA/RNA and sequencing of any length fragment from short to ultra-long, enabling the elucidation of all types of genetic and epigenetic variants.
Oxford Nanopore continuously drives further performance enhancements by iterating its technology to deliver improving raw read accuracy. Since 2021, users have been able to achieve >99% raw read accuracy using the most recent Kit 12 chemistry and R10.4 pore, and today Clive Brown introduces chemistry updates that bring improved data output alongside delivery of >99% raw read accuracy.
The Oxford Nanopore team has successfully modified the pore, motor and run conditions to increase speed that DNA/RNA is passed through the nanopore, therefore increasing the data yield while retaining high-accuracy data. This latest update will be included in a new kit (Kit 14), to be used alongside a new pore (R10.4.1) which has been shown to achieve 99.3% raw read accuracy, whilst maintaining speeds of 420 bases per second — a similar speed to the current high-yield R9.4.1 flow cells.
Figure 1: Raw read accuracy and speed of DNA/RNA passing through the nanopore (bases per second). Faster translocation speeds enable higher data outputs, supporting cost effectiveness and experiments that may require higher data volumes.
Further to this, to provide users with everything they need in one kit, Kit 14 will allow users to tune their accuracy and output needs by modifying the run conditions, which will be programmable in the software. Early work has shown that the highest raw accuracy (99.6%) can be achieved when run at 28 degrees. At 35 degrees, it’s possible to maintain >Q20 raw accuracy (99.3%) whilst generating a higher yield.
The Duplex method, where template and complement strands are paired informatically, is also improved with the new chemistry, with users now able to achieve >Q30 modal accuracy following kit updates.
As the R10 pores and compatible chemistries continue to improve, Oxford Nanopore plans to discontinue the R9 pore. This is expected to be towards the end of 2023, ensuring parity of kit performance across all technology applications and allowing for a suitable transition period, through which users will be supported.
Real time methylation analysis
Nanopore sequencing of native DNA/RNA enables users to gain information about base modifications alongside nucleotide sequence. Clive reviews the recent release of Remora – Oxford Nanopore’s latest methylation analysis tool —which further enhances base-modification analysis on nanopore devices and will be integrated into MinKNOW soon, enabling much greater access to methylation data in real time.
Clive shared data demonstrating clear benefits of nanopore sequencing using Remora over bisulfite treatment with short read sequencing. The data showed that when tested on five oligos representing regions of the human genome, the modifications were called with 99.8 % accuracy using nanopore compared to 98.6 % with bisulfite, where methylation analysis with nanopore sequencing required no additional up-front processes or cost.
Additionally, nanopore sequencing can differentiate between modifications such as 5mC and 5hmC and novel modifications can be trained into the algorithms as these become fully characterised, ensuring a complete methylation picture from a single experiment.
Sequence any length fragment of DNA
Oxford Nanopore’s technology sequences DNA or RNA molecules of any length, from short to ultra-long. It is the only technology on the market capable of sequencing DNA lengths spanning five orders of magnitude in a single technology. However, software configurations have historically been chosen to avoid shorter strands to minimise “adapter only” reads.
Last week, Short Fragment Mode (SFM) was released into the operating software, MinKNOW, and is designed to allow reads as short as 20 bases. Oxford Nanopore have demonstrated over 250M native human reads, with an average read length ~ 200 bases, on a PromethION flow cell.
Following analysis of Clive’s own genome and cell-free DNA isolated from plasma — the ‘cf Cliveome’ — using SFM, the team uncovered interesting biology, including plausible tissue typing from methylation data and greater insight into nucleosome arrangements.
Clive also introduced Dorado — a new basecalling framework that will ultimately speed up access for users. Dorado is designed with support for Apple GPUs and new NVIDIA hardware. It is also projected to keep up with high accuracy (HAC) models on PromethION 48.
He also briefly touched on adaptive sampling – real time selection of regions of interest using system software - on the high-output PromethION device. This will be available soon as a beta feature and is already available for users of the benchtop GridION and palm sized Mk1C devices.
The PromethION device can currently turn around a keep/reject decision in the time it takes for 1000 bases to pass through a nanopore. It is expected to reach GridION levels of ~200 bases with further firmware development.
Further updates will be provided at our nanopore sequencing conference, London Calling, in May 2022.
This announcement contains certain forward-looking statements. Phrases such as "potential", “expect”, "intend", “believe we can”, “working to”, "anticipate", "when validated", and similar expressions of a future or forward-looking nature should also be considered forward-looking statements. Forward-looking statements address our expected future business, and by definition address matters that are, to different degrees, uncertain and may involve factors beyond our control.
Watch the update in full here: