SAN DIEGO – Roswell Biotechnologies has unveiled the capabilities of its core electronic sensing technology and in doing so established itself as more than just a DNA sequencing firm.
"We always had in the back of our minds that you could do diagnostics this way," said Barry Merriman, the company's cofounder and CSO. "Sequencing is essentially the hardest diagnostic, maybe the highest value diagnostic. But when the pandemic hit, we said, 'Let's see if our tech can make a contribution here.'"
Now, having shown it can detect SARS-CoV-2 spike protein using its platform, Roswell is targeting diagnostics, especially for infectious diseases, or drug discovery as its first applications.
At an event on Monday to unveil its ME 1947 molecular electronics chip and the Trinity desktop instrument featuring it, the firm set grandiose expectations. Company officials said they chose the date to correspond to the 50th anniversary of the release of Intel's 4004 microprocessor, the first of its kind and a major step toward ubiquitous computing. The company also published full-page, color ads in the New York Times, San Francisco Chronicle, and San Diego Union-Tribune, proclaiming its arrival.
But at the very least, the firm has developed a new way to look at the binding kinetics of many biological molecules, not only between DNA polymerase and the natural bases but also between complementary nucleic acids, antibodies and their targets, proteins and their inhibitors, and even CRISPR complexes. The current generation of chips offers 16,000 features, each of which can be seeded with a molecule of choice, turning it into a current meter.
With this technology, researchers will "be communicating with biology on a level never seen before," said Jim Tour, a professor at Rice University who helped create the first molecular circuit nearly 30 years ago. Tour is an adviser to Roswell and a minor shareholder. "The amount of information we're going to be able to get from this is much more than we're seeing today. We have no idea how much information we're going to get from listening to what nature does in real time."
The event set the stage for the next phase of commercialization. The company showed data demonstrating several potential applications, including infectious disease diagnostics, drug discovery assays, and DNA and methylation sequencing. Now, it is looking for partners to put their molecule of interest onto the chips.
Founded in 2014, by sequencing industry veterans Merriman and Paul Mola, Roswell was an attempt to bring molecular electronics — the integration of biological molecules into semiconductor computing — to DNA sequencing. The pair had previously worked together on Thermo Fisher Scientific's SOLiD and Ion Torrent sequencing platforms, as well as Genia, a nanopore-based sequencing technology acquired by Roche in 2014.
Roswell has grown to nearly 50 employees and keeps getting bigger. "As we start to engage these development projects, we intend to grow. I see a path to at least double in the not-too-distant future and triple by some time in the next year," Mola said.
"As we move towards the market, we need to get on the fundraising wagon," he said, adding that the firm has plans for a Series C financing round. Roswell has already closed a Series B round, but Mola declined to say how much it raised and from whom. In 2019, it raised $32 million in Series A financing.
In May 2020, Roswell announced that it had partnered with Belgian semiconductor manufacturing firm Imec to create its chips, which essentially contain thousands of current meters. These circuits are completed when filled with a long, helical protein bridge that is linked to the molecule of interest. As that molecule binds to others, the chip can detect changes in current, which it records as current pulses over time.
While the primary readout is current level, the width of the pulses — dwell time — and the time between them, as well as maximum and minimum current levels, and even the distribution of interactions, can be used to recreate familiar datasets, such as DNA melt curves.
Among the molecules that Roswell has been able to seat in its chips are DNA and RNA oligos, aptamers, antibodies, other proteins, CRISPR guide RNAs, and DNA polymerase. The firm has also shown it can use electric fields to direct molecules into different rows of sensors on the chip, suggesting multiplexing capabilities. Ultimately, it aims to have each circuit be individually addressable.
The current, third generation chips each have 16,000 features, but Roswell is looking to increase that by several orders of magnitude. Its fourth generation chips, expected by the end of 2023, could have up to 1 million features, and the generation after that as many as 20 million.
That means whole-genome DNA sequencing is still some ways off. Mola suggested that genome sequencing would require at least 1 million sensors per chip and as many as 5 million. However, the firm has shown proof of concept of sequencing and still plans to pursue that application. "Discrimination of the four bases can approach over 90 percent, and in complex templates, 70 percent," Mola said. "These areas are all under active development."
"Like everyone else, we'd rely on consensus sequence for sequencing [accuracy]," Merriman added. Read lengths, theoretically, are only limited by the type of DNA polymerase attached to the circuit. The chip could also be used to do methylation sequencing without bisulfite conversion.
"We intend to be able to provide lower costs than other folks, too," Merriman said. "The whole point of putting these on scalable chips is that in the long run, we can beat anyone's costs. The scaling limits of our technology are insane. We don’t even bother to quantify them because chip costs get driven to essentially nothing."
"The point is, we're trying to get to the lowest cost per data point," Mola said.
Lower costs may help the company eventually achieve another application it has had in mind: at-home diagnostic testing. On display in a glass case at its San Diego offices is a concept device for at-home viral testing. Shaped like a hexagonal flying saucer and about the size of a 2000s-era personal CD player, it would take a cylindrical reagent cartridge with, say, a sputum receptacle. The company hasn't yet created a functional one yet, Merriman said, but "there's an actual plan, with parts and things. It’s not just a graphical imagination." Assays could be for infectious disease testing, like COVID-19 and other respiratory viruses, or even for other facets of health, such as vitamin and hormone levels, Mola suggested.
"We had a goal, where the target price would be such that you could run this weekly if you wanted to, daily, if needed," Mola said. "What's different now is that we need platforms like that. We had this at-home testing plan before the pandemic and, really, no one listened."
Now, the firm is looking for partners who want to insert their molecule of interest into the Roswell chips and expects to announce its first collaborators soon. It needs experience incorporating other types of molecules into its bridge proteins, which will help it as it moves to an early-access program, slated to begin around Q2 of 2022. "We've had success with pretty much everything we've tried," Merriman said, "but we need to understand how it's going to change to create a product that makes sense [to the partner]."