NEW YORK – INanoBio, a nanopore sequencing startup, is working on speeding up the analysis of DNA.
Earlier this week, the Arizona State University spinout said it can make sensors that read electric signals at a rate of 100 megahertz — about 100 times faster than existing nanopore technologies.
"We've built the world's fastest nanopore sensor," said Bharath Takulapalli, the company's CEO and inventor of the field effect nanopore transistor, or FENT, the firm is using as a sensor. It's an important internal milestone for the Scottsdale, Arizona-based company, which is ultimately seeking to develop a long-read, high-throughput sequencing platform. "The transistor is working better than I expected. From a sequencing point of view, we don’t need anything faster," he said.
The firm has already paired its FENTs with solid-state nanopores 50 nm in diameter and said it has found a way to manufacture pores less than 5 nm wide, as it would need to do for a sequencing device. Now, it must find a way to combine its fast FENTs with smaller pores, at production scale.
"The technology indeed has a reported higher nanopore sampling rate than any other known competitor, which is compelling," said Chris Mason, a sequencing technology expert at Weill Cornell Medicine. Oxford Nanopore's sequencing technology is just one of the platforms used in Mason's lab. "But in-depth benchmarking is the next logical step, such as with titrated human and microbial controls samples," such as standards from the Genome in a Bottle Consortium and the International Microbiome and Multi-Omics Standards Alliance, he said.
"With iNanoBio, the whole intention is to go as quickly as possible," said George Church, professor at the Wyss Institute for Biologically Inspired Engineering and Harvard Medical School and a member of the company's scientific advisory board.
Oxford Nanopore's technology, the only nanopore-based sequencing method on the market so far, must use a polymerase as a ratcheting mechanism to slow the passage of DNA through the protein pore and give the device enough time to make the electrical measurements for each DNA base.
By detecting changes to the surface charge of the molecule, iNanoBio can theoretically call bases without slowing them down, as long as the device has a high enough sampling rate. With the FENT working at 100 MHz, iNanoBio now has that device, one that could even go as fast as 1 GHz, Takulapalli noted.
Spun out of Arizona State University in 2007, iNanoBio has spent the last few years progressing on its solid-state concept. In 2019, the company received a $5.4 million grant from the Defense Advances Research Projects Agency to develop nanopore tech for sequencing epigenetic modifications. That grant was part of a $27.8 million project led by Mount Sinai's Icahn School of Medicine to develop a device for rapid analysis of epigenetic markers using blood samples, with potential applications in detecting exposure to weapons of mass destruction.
The firm has now grown to 17 employees, up from eight in 2014, with goals to more than double in size over the next year.
Takulapalli also said he's working on securing Series A financing but declined to provide further details. In 2019, he told GenomeWeb that he wanted to raise more than $40 million in private investment by the end of 2020.
The company is now working toward a DNA sequencing platform, Takulapalli said, but how long it might take to get there is unclear.
The next technical milestone is to generate chips with FENTs paired with 20 nm to 30 nm pores, a process which could take up to a year. Takulapalli noted that the firm's manufacturing methods are done with the type of silicon wafers actually used in foundries, "so not just a prototype in a small lab." But the firm has often provided overly optimistic timeframes: In 2019, Takulapalli told GenomeWeb the company was aiming to have its proof-of-concept device ready by the end of 2020, and he said the same in 2014.
Church said the "wishful thinking academic" in him hoped to get an iNanoBio prototype in his lab "within a year," but he cautioned he was "in no way speaking on behalf of the company."
The firm has not yet publicly released data on its technology. "We know we can get a solid publication," Takulapalli said, "but we want to get a little bit closer to commercial launch and then publish." He said the firm is targeting read lengths between 100 kb and 1 Mb and wants to enter the market with "greater than 99 percent raw read accuracy" and "at least 99.9 percent consensus accuracy."
"Getting read qualities above Q20 [or 99.9 percent accuracy] would be ideal," Mason said, "but even Q15 can be useful if the reads are long for certain applications like assembly, phasing, or structural variation mapping."
"Very often there's no upper limit to how much accuracy you can get if you throw enough reads at it," Church added.
In parallel to its work on manufacturing, the company is working on a base calling algorithm. "We've already started on doing very traditional analysis and also machine learning-based analysis for base signature detection," Takulapalli said.
In addition to seeking funding partners, he said the firm is "open to working with others to commercialize" its technology. The firm will target the research market first, but "dares" to go after the diagnostics market, he said, including point-of-care diagnostics.
The nanopore tech could even be used for even more than just DNA sequencing. "It will be able to do both protein detection and protein sequencing," Takulapalli said. "However, these are not our immediate areas of focus. Later on, once we develop a DNA sequencer, we will start work on protein sequencing."