This story has been updated to include additional information from Singular Genomics.
NEW YORK – Singular Genomics on Thursday officially launched the G4 sequencer, a sequencing-by-synthesis platform that could challenge Illumina's near-monopoly in the short-read sequencing market.
The new instrument, the company's first, is a "Swiss Army knife that can meet whatever needs the customer has," said Singular cofounder and CEO Drew Spaventa. "The target market is broad. Anyone running an [Illumina] NextSeq, HiSeq, or NovaSeq … will benefit from the G4 in terms of faster turnaround times and more flexibility and versatility."
San Diego-based Singular is accepting orders now for second quarter 2022 delivery of the $350,000 instrument and its first generation of kits. The firm's initial F2 flow cells can generate approximately 150 million reads and have a data output of up to 50 Gb. The 100-, 200-, and 300-cycle F2 kits can be used for applications such as RNA gene expression, single-cell RNA sequencing, exome sequencing, and other targeted sequencing. Human whole-genome sequencing will be unlocked by the end of next year, the firm said, with its F3 flow cells, which will provide up to 330 million reads and up to 100 Gb output. Costs for sequencing kits will range from about $16 to $45 per Gb, with volume discounts available that the company hopes will make the platform cost-competitive with the lower end of the NovaSeq range.
With single-read accuracy of 99.6 percent to 99.9 percent and quality scores above Q30 for 75 to 90 percent of bases across all kits, Spaventa believes Singular has met the "table stakes" to get in the game. For comparison, the Illumina NovaSeq and NextSeq platforms boast more than 85 percent of bases at higher than Q30 in 2x150 bp mode.
Moreover, the company has built a platform that it thinks will appeal to the many needs of researchers. "We really started from a blank slate and tried to figure out what people needed," Singular CSO Eli Glezer said. "At the beginning, I said, 'We're only about 300 good design decisions away from a great product.'"
Singular believes that modularity is chief among those design decisions. "A lot of people still have a MiSeq, not because they don't have a higher-throughput machine, but because they need to keep their lanes separate," Spaventa said. Each G4 flow cell has four independent lanes that can be loaded separately. Using the G4, a researcher could keep samples separate with the added benefit of lower costs enabled by mid-throughput sequencing.
"This is really exciting from a core lab and service center perspective," said Scott Tighe, technical director of the University of Vermont advanced genomics lab, who reviewed some of Singular's promotional materials. "This is a level of flexibility we haven't seen in a while, since HiSeq was discontinued. They've got the output; they're just trying to clean up the accuracy and Q scores. Clearly, they're going to get there. I don't think anybody that sees this will doubt that they'll hit Q30, given enough time."
"The data look very good. We're excited to see this enter the marketplace," he said. "They are something to reckon with, right now. They're going to make Illumina very nervous."
According to Spaventa, the G4 is the result of some networking done by former Illumina VP and CSO David Barker, now a Singular cofounder and board member, "and a lot of luck."
Several years ago, Barker introduced him to Glezer over lunch in San Diego, setting the duo on the path to reinvent sequencing by synthesis. Spaventa, an entrepreneur who still carries the competitive drive he held as a member of the University of California, San Diego basketball team, had already invested in several genomics firms, including Omniome and Edico Genome. Illumina bought Edico in 2018 for $100 million.
Glezer, now also an entrepreneur, spent nearly 20 years at Meso Scale Diagnostics, where he led immunoassay development as chief technology officer. "I was looking to do something more on my own," he said. "Something to have an impact."
From the company's founding in 2016, Spaventa and Glezer put special emphasis on the company's sequencing chemistry. "We both agreed that the firm needed to overinvest in the core chemistry," Spaventa said, and that was all the company worked on for its first year. Though several of Illumina's core SBS patents are set to expire by 2023, the company decided to build its own chemistry from the ground up, with novel polymerases, custom reversible terminators, linkers, dyes, and its own imaging method. "We built our R&D so that we don't need any of those [Illumina] patents to expire," Spaventa said. "We've taken a thoughtful approach to stay away from that landmine."
"Kudos to them, if they managed to develop new dyes," said Chris Mason, a sequencing expert at Weill-Cornell Medicine. "Most dyes have been tried. I'd be impressed if that were true."
The result is a four-color sequencing chemistry that fits in with established sample preparation and bioinformatics workflows. Singular has its own sequencing adapters, but other than that, it should be a "drop-in replacement" for existing methods, Spaventa said. Performing imaging in parallel with the chemistry "allows a really fast cycle time."
At launch, the claimed cycle time is four minutes. Some early-access partners have seen cycle times faster than that, Spaventa said, and he expects to get below three minutes by the time the system ships. Ultimately, he thinks the cycle time can be as low as two and a half minutes.
Featuring flow cell and imaging technologies developed in-house, Singular beta-tested its platform at three sites: Fate Therapeutics, Sanford Burnham Prebys Medical Discovery Institute, and a spatial genomics lab at Beth Israel Deaconess Medical Center. Researchers at those labs did not respond to repeated requests for comment.
The firm has also secured a handful of undisclosed early-access partners.
In addition to instrument and kit specs, Singular released a technical report this week that includes data generated by the G4. Both Tighe and Mason noted that the data suggested a drop in coverage in higher GC content regions. "It's not awful, but it's less coverage than what you'd expect," Mason said. "There seems to be a drop above 45 to 50 percent [GC]. Maybe there's a little bit of work to be done in the basecalling algorithms or chemistry to eliminate those dips."
Tighe suggested that researchers could simply sequence those regions more deeply. "People don't worry about sequencing deeper because we're all fighting with duplicates anyway," he said.
The indel precision and sensitivity was less than 99 percent, Mason noted. "They'll need to work on that."
Mason added that he would have liked to see a transition to transversion ratio, a quality indicator for inferring SNPs in NGS data, as well as data on structural variant and copy number variant analysis. "Based on metrics here, it's probably going to be comparable to Illumina," he said. Singular Genomics did provide a transition to transversion ratio, which was 1.99 for targeted sequencing on an F2 flow cell at 30X and 40X coverage and 2.00 for 20X coverage.
He also wondered what the performance was with bisulfite sequencing for methylation analysis or with custom adapters for various multiplexing schemes. Overall, he called the G4 platform "compelling," especially because of its flexibility. "It looks like it's competitive," he said.
Among the group of companies that are chasing Illumina — which also includes BGI's MGI, PacBio's Omniome, Genapsys, Element Biosciences, and others — the G4 launch will help Singular pull a bit ahead in the early going. But they won't be in this position for long. A recent jury verdict invalidated one of Illumina's core SBS patents that was set to expire in 2023, meaning that companies like BGI could enter the US market as early as next year.
Still, the company that poses the biggest threat to Singular is Illumina. "I don't know if 'scare' is the right word, but they've set the bar pretty high," Spaventa said. "They've provided the only solution for a lot of applications, and no one has been able to meet them yet, in terms of cycle time, throughput, or accuracy. I think anyone else that could be mentioned is not really a thought for us."
He also said he wouldn't consider selling to Illumina. "Our vision is to build a really massive company. There's a lot of growth in the space and room to create better clinical offerings."
One of those offerings, coming at the end of 2022, is a kit for high-accuracy rare variant detection and sequencing cell-free DNA. "It will push accuracy two orders of magnitude higher, up to Q50," Spaventa said, by using a hairpin and adapter to preserve association between complementary strands of DNA.
And there's always the PX system, still in development, for spatial, single-cell, and multiomics.
"I think we're in a unique position in that we're able to leverage a tremendous amount of technology in the G4 that will give us a dangerous and compelling head start in the multiomics space," Spaventa said.