NEW YORK (GenomeWeb) – Oxford Nanopore Technologies plans to make several upgrades and changes to its existing platforms and kits that will improve sequencing throughput and accuracy while reducing costs. The firm is also working on a new product, code-named Plongle, that will enable multiplexed low-throughput sequencing in a 96-well plate format.
During a presentation at Oxford Nanopore's annual user meeting in London yesterday, a video of which was released on the firm's website today, CTO Clive Brown also outlined the concept of a new DNA data storage technology that will use nanopores to encode information by making modifications on existing DNA strands.
The company recently released the 48-flow cell version of the PromethIon, P48, which has been able to generate 7.3 Tb of data in its best internal run at the company so far, and more than 5 Tb of data in the best run at a customer site. The best internal run for the smaller P24 version to date has generated 3.8 Tb of data. Brown said there is still headroom to improve the platform further, enabling it to generate 20 Tb of data in a run, though he did not provide a timeline.
Oxford Nanopore is also working on an update to the GridIon X5 that will start shipping in September. Brown said the company has had technical issues with the power supply of the sequencer. The new version will have a different GPU, allowing it to be run with a smaller power supply, which he said will "make it more reliable."
According to Rosemary Dokos, senior director of product management, the GridIon upgrade will be available free of charge to GridIon customers with an active warranty and open flow cell orders.
The Flongle, a flow cell adapter for the MinIon or GridIon with disposable flow cells that Oxford Nanopore launched in March, has seen "enormous interest", Brown said, though he conceded that it has been a challenge to ship Flongle flow cells in large numbers. He said the company is "working hard on scaling manufacturing" and just built a new factory for Flongle in Oxfordshire. The current customer record for output per Flongle run of 1.9 Gb beats the firm's internal record of 1.5 Gb, he said, and the customer median output has been 524 Mb per run. There is still room for doubling or quadrupling the output of Flongle, he said.
The company is also dropping the price for the cheapest Flongle package to $1,860 for a bundle that contains one adapter and 12 flow cells and allows users to purchase additional adapters for $750 and additional flow cells for $90. Dokos said customers had complained about the steep entry prices of $5,250 for a package with one adapter and 48 flow cells.
In addition, Oxford Nanopore will make its new R10 flow cells, which contain a new type of nanopore with a longer barrel and a dual reader head that promises better accuracy, widely available to customers soon. The new flow cells will start shipping to MinIon and GridIon customers in June, PromethIon customers in July, and Flongle customers in September. Older R9.4.1 flow cells will be available and supported across all Oxford Nanopore devices and R9.5.1 flow cells for the MinIon and GridIon only.
The MinIon Mk1C, which integrates MinIon sequencing and real-time computing in a single device, will start shipping in July, and customers can trade in their MinIT compute units and used flow cells towards an upgrade.
Brown also outlined several new products that are still in development. He said the company has built a new application-specific integrated circuit (ASIC) that uses less power, has a smaller footprint and up to 400 channels, and will be disposable because of its low fabrication costs.
While the new ASIC could become part of the planned SmidgIon mini-sequencer that runs off a mobile phone, Oxford Nanopore plans to use it first as part of a multiplexed sequencing platform that is code-named Plongle for plate-flongle.
Plongle will enable nanopore sequencing in a 96-well plate format, where each well on the plate will have one of the new ASICs and 70 channels. The expected initial output is 0.5 Gb to 1 Gb of sequencing data per well, at a price of $25 to $50 per well. The target launch date for Plongle is the second half of 2020.
Oxford Nanopore is also still working on the Voltrax-based droplet sequencing technology that Brown introduced a year ago. "We're still working on this, it's moving along very nicely," he said, though a lot of design work still needs to be done. The firm has shown proof of concept for the sequencing chemistry and the technology "may appear in some form next year" on the market, he said.
Finally, Brown unveiled a new DNA data storage technology the company has been exploring. It will use protein nanopores to "write on DNA" and "treat it like a tape," he said, thus encoding information on it in the form of DNA modifications. For example, a modified cytosine might encode a 1 and an unmodified cytosine a 0. The DNA modifications would take place inside the nanopore and could be triggered chemically or by light, and the nanopore would also serve as an in-built proofreading device. Nanopores could also be used to decode, or read, the stored information.
Brown said this way of storing information in DNA would be more scalable than traditional DNA synthesis, although the information density would be about tenfold less. "Whether it's commercially viable, I don't know," he said. "I think it might be, if we get it working well enough."
It is unclear whether this work is related to that of Oxford Nanopore spinoff Genome Foundry, which Brown first mentioned in 2016 and which he said then was working on a proprietary, non-phosphoramidite-based DNA synthesis chemistry with the goal of generating long DNA molecules in parallel.