Zachary Bickel

A DNA Sequencer in Every Pocket

A biotech company is building devices that will allow people to decipher genes in remote jungles, at sea, or even in space—and they say they’re just getting started.

Aboard the International Space Station, six people are currently orbiting the planet at 17,000 miles per hour, taking in fifteen sunrises and sunsets every day. The view is unbeatable; the floating sensation, sublime.

But good luck to them if they get sick.

There’s nothing on board the ISS that can definitively diagnose a disease, or identify the microbes behind it. Instead, sick astronauts have to settle for describing their symptoms to medical staff on the ground. They have no way of knowing for sure if their disease is bacterial, viral, or something else, or if raiding the station’s finite supply of antibiotics would do them any good.

If an astronaut could decipher the full genetic code of whatever’s plaguing her, she could identify the offending bug and work out if it’s vulnerable to any drugs. But until recently, this scenario would have been laughably impractical. Sequencers were all the size and weight of microwaves and fridges. They’d be impossible to cart aboard a space station, and probably wouldn’t have survived the trip.

Thanks to a British company called Oxford Nanopore Technologies, that’s no longer true.

In the spring of 2014, the company released a USB-powered sequencer called the MinION (pronounced “min-eye-on” not “min-ee-un,” and neither yellow nor cute). Four inches long, one inch wide, and 87 grams in weight, it’s smaller than most chocolate bars and smartphones. Earlier this year, I clutched one in my hand, with room for several more. One scientist describes it as “the DNA sequencer you can forget in your jacket pocket, which I’ve done once.”

“Finally, we have a sequencer that’s small enough that we can send it up into space,” says NASA engineer Kristen John. Having tested the MinION on an Earthbound flight simulator, she and her colleagues will be sending one to the ISS in June, along with some DNA samples to test. If it performs as well in microgravity as it does on the ground, astronauts could finally monitor their health in real-time, which could be crucial for future, ambitious missions. On a voyage to Mars, “we’ll lose the ability to resupply antibiotics,” says microbiologist Sarah Castro. “What we take with us is what we’ll be limited to. We’ll need to know what’s causing an infection to know how to treat it appropriately.”

With the MinION, astronauts could also do experiments to see how bacteria respond to microgravity, without first having to dunk their samples into fixatives and bring them back to Earth. And they could study microbes in the space station’s air, water, and food. “Currently, we’re telling the crew what they were eating, breathing, and drinking six months after the fact,” says Sarah Castro.

While one MinION is heading off-world, others have already traveled around the world. These tiny machines and their companion devices are set to revolutionize and democratize the world of genomics, unmooring it from well-equipped institutions and laboratories and releasing it into society at large. If Oxford Nanopore gets its way, people will be able to sequence DNA in hospitals and jungles, yachts and security checkpoints, classrooms and living rooms. But as history has shown, getting its way has never been easy.

* * *

A nanopore is exactly what it sounds like: a small hole. Typically, it’s a tiny peg-shaped protein with a hollow tube at its core, just a few billionths of a meter wide. In Oxford Nanopore’s devices, the protein sits in a synthetic membrane, submerged in liquid. When a voltage is applied across the membrane, ions flow through the pore, creating an electric current. But if something blocks the pore—say, a strand of DNA—the ions are impeded and the current drops.

The four building blocks (or bases) of DNA—A, C, G, and T—each change the current through the nanopore in different ways. By measuring that current, you can decipher the sequence of a DNA strand as it threads through the pore like a piece of ticker-tape.

This is dramatically different from traditional sequencing, where scientists have to amplify DNA molecules to create many identical copies, break those copies into small pieces, sequence the pieces individually, and finally assemble the fragmented sequences into a cohesive whole. It’s like reading a book by transcribing it, shredding it, and taping it back together. By contrast, nanopore sequencing is like reading the undamaged text from cover to cover. DNA can be sent through the hole without amplification or fragmentation, and sequenced in a long, continuous run.

Legend has it, David Deamer from the University of California, Santa Cruz, came up with the idea in 1989, while driving down California’s Interstate 5; he was reputedly so struck by it that he had to pull over to jot it down. It took a decade for him and his colleagues to show that they could capture DNA, funnel it through a nanopore, and differentiate between the various bases. And it took Oxford Nanopore almost the same amount to time to create a rugged, workable sequencer based on this technology. (Deamer and other nanopore pioneers sit on its technology advisory board.)

Founded in 2005, the company’s original plan was to produce a device very much like other sequencers—a large bulky box called the GridION that, in the words of one blogger, was “rocking the VCR-machine-circa-1992 look.” The idea of shrinking it down came from chief technology officer Clive Brown, a fidgety and outspoken man who was once described as “the most honest guy in all of next-gen sequencing.” When I ask him about the MinION’s origins, he says, “You can thank Illumina,” referring to the San Diego-based company that leads the sequencing market. “I’m desperately thinking of ways of bringing them down.”

In an earlier job, Brown helped to develop a DNA sequencer for a company called Solexa that, against his wishes, was sold off to Illumina in 2007. The sale helped transform Illumina into a sequencing juggernaut, whose machines drive almost every large sequencing center in the world, giving the company an almost unshakeable monopoly on the industry. In 2009, this colossus invested $18 million in Oxford Nanopore for the right to commercialize the upstart company’s technology, and their CEO got a chance to observe ONT’s board meetings. “He poo-pooed everything I said,” says Brown. “So I’d go away and think: what is the simplest thing I can make that works, and that doesn’t look like an Illumina box.” (Illumina declined to comment for this story.)

Brown revealed the MinION to the world in February 2012, at a conference in Florida. He spent most of his time talking about the traditional GridION, only mentioning the smaller device in his last two slides. “Almost as a punchline,” he says. It was a “megaton announcement,” said one scientist. Another tweeted: “I felt a great disturbance in the force, as if a million Illumina investors cried out in pain.” One enthusiastic fellow tried to break into Brown’s hotel room.

Crucially, the MinION was an Oxford Nanopore product, through and through. It used a variation on nanopore sequencing that wasn’t covered by the company’s deal with Illumina, and the two severed their financial ties in 2013. But, meanwhile, Oxford Nanopore was also inadvertently cutting itself off from the scientific community. At Florida, Brown had claimed that the MinION would be out by the end of 2012. It wasn’t. The company was beset by manufacturing problems that led to long delays. Worse still, they went silent, costing them both credibility and support. Many alienated scientists wrote off the MinION as vaporware.

Brown was unapologetic, and reportedly took note of the critics. “Clive has a list. A list of people that says he’ll see to it won’t get a MinION when it comes out,” wrote microbiologist Nick Loman in 2013. “I can’t tell if he’s joking.” Another geneticist, who did not want to be named, told me, “There are a lot of us who are concerned about criticizing the company because our access to the technology could be arbitrarily revoked.”

* * *

In 2015, the largest Ebola outbreak in history was entering its second year. More than 10,000 people had died and many more had been infected. Scientists and health workers were making solid progress at containing the epidemic, but one crucial element was missing: Ebola genomes.

By sequencing viruses in an outbreak, scientists can more effectively develop diagnostic tests and vaccines. By comparing viruses from different places and times, they can work out how many strains are at play and whether they’re mutating, and plan control measures accordingly. By comparing sequences from different patients, they can work out who is infecting whom, and curtail those routes of transmission.

But over the course of the Ebola outbreak, only a small number of viral genomes had been sequenced. The virus had struck remote regions, so samples had to be shipped to distant labs to be analyzed. Understandable, thought Nick Loman, but intolerable. Rather than waiting for outbreak samples to arrive at sequencing facilities, why not take the facility to the outbreak?

When Oxford Nanopore finally released the MinION via an early access program in February 2014, Loman was one of hundreds of scientists who signed up. In exchange for a $1,000 deposit, they got a MinION and a regular supply of flow cells—the disposable wafers that drive the sequencer, each containing 512 nanopores. These early versions were plagued by shipping problems, unreliable reagents, and technical difficulties,forcing several scientists to abandon them in frustration. But Loman persisted.

In June 2014, he used the MinION, now debugged and refined, to successfully study a Salmonella outbreak at a Birmingham hospital. And in April 2015, his student Joshua Quick travelled to Guinea with three MinIONs, inexplicably called Ribz, Chicken, and Brisket. He also brought three laptops, some chemical reagents, a centrifuge, and a thermocycler stolen from a lab mate—a mobile diagnostic laboratory that fit into just two suitcases and unfolded onto two small desks. Within two days of arriving, Quick started sequencing Ebola.

During the outbreak, the MinION proved its worth. Traditionally, scientists have to collect hundreds of samples, send them off, and wait for results to return after days or weeks—a slow process, ill-suited to the immediacy of an epidemic. But the MinION churns out results quickly; in one instance, Quick went from sample to sequence in under 24 hours. And those sequences emerge in real-time, almost as soon as the DNA strands cross the nanopores. “As that data is generated, it’s analyzable,” says Loman. “It’s a more interactive approach to sequencing.”

Over six months, the team sequenced 142 Ebola genomes, which their colleagues used to monitor the last vestiges of the outbreak. As Lauren Cowley, who took over the pop-up lab from Quick, told me last year: “I saw, first-hand, epidemiologists being able to accurately track transmission routes in real time and then intercept the chain to prevent further transmission of the virus.”

The World Health Organization declared the epidemic over in February 2016. But just as Ebola waned, another adversary ascended. This year’s big threat is the mosquito-borne Zika virus, which has been linked to a birth defect called microcephaly and other conditions. Zika has spread throughout the Americas and the Pacific, and once again, geneticists are playing catch-up. “There’s probably about 10 or 15 publicly available sequences,” says Loman, none of which are from northeast Brazil where the outbreak began. “It’s logistically quite difficult to access samples, and very difficult to ship them out of Brazil for sequencing,” he adds.

So once again, Loman is taking sequencers to the samples. In a month or so, his team will load a caravan with MinIONs and other laboratory equipment, and drive it around coastal Brazil, from Belem in the north to Salvador in the east. He hopes that this road trip will reveal more about Zika’s origins, how often it has entered the Americas, how it interacts with the immune system, how many strains there are, and whether it interacts with other viruses. “We have no real feel for that at the moment,” he says. “With Zika, there’s so little known.”

The MinION isn’t the only option for a project like this. In 2013, one group of scientists sailed a more traditional microwave-sized sequencer around the Southern Line Islands, some of the most remote landmasses on the planet. But the MinION is so lightweight, robust, and responsive that “it just makes things easier,” says Loman. One team took it into the Tanzanian rainforest to identify frogs. Another is planning to take it aboard a marine research vessel in the Indian Ocean this summer. John and Castro are sending it into space.

Despite its strengths, the MinION still doesn’t come close to competing with Illumina’s top-of-the-line sequencers in either cost or power. It can easily sequence the tiny genomes of viruses and bacteria but it’s too slow to handle the much larger genomes of animals or plants. (For comparison, the human genome is a thousand times bigger than that of the bacterium E. coli, and the wheat genome is five times bigger still.)

That may change as Oxford Nanopore rolls out Fast Mode—a hardware and software upgrade that will rev up the MinION’s speed by 4 to 7 times. The company is also launching the PromethION—the MinION’s bigger, badder cousin. Built for large-scale sequencing, its 144,000 nanopores could conceivably churn out 120 gigabytes of data every day, equivalent to 40 human genomes. The first of these beasts has already shipped, and more are set to follow.

Quality matters as much as quantity though, and concerns over low accuracy have plagued the MinION since its release. An Illumina sequencer has an error rate of just 0.1 percent. By contrast, Loman got error rates of around 10 percent in his Ebola work (although those plummeted when he read each genome several times and combined the results). “The error rate is now in single figure percentages in our hands, and we’re not at the limit,” adds Brown.

The technology also needs to be more robust. For now, the flow cells last for a couple of months—good enough for many applications but not, say, NASA’s extraterrestrial ambitions. They are costly, too: $500 each if you buy them in bulk. “If they were a fiver, that would be awesome,” says Loman. A drop of that magnitude isn’t unfeasible, but Oxford Nanopore has other financial concerns to worry about. In February, Illumina filed a lawsuit against them, claiming that they are using a particular nanopore that infringes upon Illumina’s patents—a claim that CEO Sanghera has snarkily denied. (“It is gratifying to have the commercial relevance of Oxford Nanopore products so publicly acknowledged by the market monopolist,” he said.)

Whether this means a rival nanopore sequencer is in the works is anyone’s guess. Even if Oxford Nanopore remains the only company producing such tech, and even if they can surmount their technical and legal challenges, they’ll face one final obstacle: Their little-sequencer-that-could is still reliant on the trappings of laboratory science. You can’t just drip a spot of blood or water into the flow cells; you need to prepare the sample first. That process requires chemicals and equipment like centrifuges, thermocyclers, and pipettes, not to mention training in molecular biology. And once the sequences are ready, you need specialized software and expertise to interpret the strings of As, Cs, Gs, and Ts.

Which means that the MinION can go anywhere, but it can’t be used by anyone. Or, at least, not yet.

* * *

“What does VolTRAX mean?” I ask Clive Brown. “It doesn’t mean anything; I just made it up,” he answers. “The way it works here is I say we need a name, and there’s silence throughout the company. I say, ‘What about this?,’ and they all poo-poo it. I say, ‘So what are your suggestions?,’ and there’s nothing. So we go with my name, which is how we ended up with MinION.”

The etymology of VolTRAX may be farcical, but its purpose is not. This domino-sized add-on for the MinION is designed to prepare a biological sample—say bodily fluids, or swabs of soil—for sequencing. It moves liquid through a network of fine channels, bombards it with chemical reagents to extract DNA, and loads that DNA into the MinION. “We spend a lot more time in the lab preparing samples than we do sequencing,” says Loman. “If, and it’s a big if, you could drop the clinical sample onto the chip that does it all for you, it would be hugely advantageous.”

VolTRAX is set to go out to early users this summer. Jon Wetton, a geneticist and forensic scientist at the University of Leicester, wants to use it to fight illegal wildlife trafficking, by sequencing telltale genes that act as identity badges for different species. Conservationists have already used this technique, known as DNA barcoding, to track sources of elephant ivory or identify whale meat posing as sushi. But samples must be shipped for analysis, and “you’re looking at weeks or months to get the results back,” says Wetton. “That can’t be done on anything perishable, or if you’ve got a suspect in custody. But with an on-the-spot test, you could confiscate, arrest, or do something about it.”

With a nanopore sequencer, inspectors could tell the difference between a cut of beef or bushmeat from threatened apes and monkeys. They could analyze the blood on a suspected poacher’s tools to reveal the identity of the last animals it cut. They could work out if seized caviar belong to legitimate fish species or endangered sturgeon. “You could answer a whole slew of serious wildlife crime issues with the same test,” says Wetton.

That still leaves the significant problem of parsing the data, but Oxford Nanopore has a solution for that, too: an online hub called Metrichor, where people can connect to ready-made apps for analyzing DNA sequences. One such app, developed by Oxford Nanopore itself, is called “What’s In My Pot?” or WIMP. It takes sequences and identifies the organisms they belong to. The team have already field-tested it on microbes from a sewage-contaminated river behind their own building, and on unpasteurized milk from the back of a New York lorry.

Third-parties can develop Metrichor apps, too. “It’s the classic iPhone approach; you give a developer kit to anyone,” says Dan Turner, the director of applications at Oxford Nanopore. “Veterinary companies are developing an app that can check your dog’s pedigree. And we had someone who wanted to sequence poo to identify the defouling dog in their Californian neighborhood.”

Since the service is cloud-based, anyone with an internet connection and a MinION, as Oxford Nanopore hopes, could, “sequence anything, anywhere.” Without either laboratories or laboratory skills, a dairy farmer could monitor the quality of their milk. An astronaut could scrutinize their air and water. A supermarket chain could check for dangerous bacteria in its food chain.

Students could try their hands at sequencing, too. At Columbia University, Sophie Zaaijer recently developed a genomics course where undergraduate and masters students sequenced samples of food taken from New York restaurants, supermarkets, and Zaaijer’s lunchbox. They then identified any microbes within using WIMP. Normally, such students would only learn theory and manipulate data. This time, “they could really connect to the technology,” says Zaaijer. “They held this device, and they could see DNA being read through the pores in real-time.”

Similar plans are afoot all over the world. In a survey of educators, carried out by geneticists Karen James, respondents dreamt about using nanopore sequencers to get undergraduates, school students, and members of the public to study everything from feathers in Maori cloaks, to poop from animals in safari parks, to their own snot. “If every school had a MinION and was streaming data to the internet … the implications are absolutely mind-boggling,” says Joe Parker from the Royal Botanic Gardens at Kew. “But to be attractive to a citizen scientist, the price needs to come down.”

He is optimistic, though. “Ten years ago, there was some sci-fi stuff floating around about this nanopore thing. Five years ago, no one had heard from Oxford Nanopore, and many people thought they had gone away. Now, we know it definitely works. The ball’s in their court to show that they can lower their prices and scale up.”

* * *

Kris Griffin was 32 when he went to his doctor with a bad back, and came away with a diagnosis of chronic myeloid leukaemia. Thankfully, two drugs—first imatinib, and now dasatinib—have kept his cancer under control with minimal side effects. Eight years on, Griffin is doing well. He’s an education consultant based in Kidderminster, England; husband to a partner he married just after his diagnosis; and father to a four-year-old boy. “I live a normal life,” he says.

He isn’t cured, though. His disease is caused by the abnormal merger of two chromosomes, creating a chimeric gene called BCR-ABL that makes his blood cells divide uncontrollably. That’s what dasatinib inhibits. To check that the drug still works, Griffin has to visit a hospital several times a year, so his doctors can measure the number of cells that carry the fused gene. The trips eat into his days, and the results can take weeks to arrive. “And there’s no bigger reminder to someone that they’re doing poorly than walking through those doors,” Griffin says.

He has always dreamed of carrying out the tests himself in the comfort of his own home, in the same way that people with diabetes can monitor their own blood sugar levels. A year ago at a London conference, he “saw this chap on stage with this little device,” he recalls. That was Clive Brown.

Brown spoke about using nanopore sequencing on people, to analyze the bits of DNA that are released into our bloodstreams by our dying cells. To cancer researchers, this circulating DNA acts as a liquid biopsy, which can reveal whether tumors are progressing, responding to treatments, or evolving resistance to drugs. Many companies, Illumina included, are getting in on the action, and developing blood-based tools for cancer screening.

But Brown thinks that if MinION and VolTRAX become cheap and accurate enough, people could monitor their circulating DNA themselves. “We need to get the price down by an order of magnitude, but there’s no reason why you couldn’t take a daily snapshot of the contents of your blood,” he tells me. He wants to bridge the worlds of DNA sequencing and the quantified self. “My intention is to give people a tool where they can understand their own biology and make their own inferences about it.”

Griffin lit up when he heard Brown’s vision. Maybe he could eventually monitor his own BCR-ABL levels and just upload the data to his doctors. “The power it could give to patients ... Psychologically, it feels so important,” he says. He has been liaising with Oxford Nanopore ever since, and even though they’ve assured him that the technology still needs work, he is undeterred. “I want to be the guinea pig—the first person with CML to monitor my blood at home. I think this will mean everything to so many people.”

Daily monitoring might also reveal signs of an infection before symptoms occur. And it might reveal answers to questions that haven’t been asked yet. “No one has systematically inventoried circulating DNA over a long period, even in just one person,” says Brown. “What’s the baseline? It’s unknown at the minute. But we can get the data.”

There’s still the challenge of getting DNA out of your bloodstream and into a VolTRAX. “You want a simple, idiot-proof sampling device—some kind of pen with a consumable tip that contains all the gubbins for the nanopores,” says Brown. “You touch it to something that’s already wet—a drop of spit or a bit of food—and it does the rest for you.” It could gently prick the skin, or simply sample the fluids that leak out of capillaries and circulate between skin cells; blood sugar monitors, which Sanghera helped to pioneer, already use both methods.

A sampling device could also be used to check for viruses in air or bacteria in a food-production line. It might not even need a human operator; just program it to collect regularly, and upload the data to the cloud. That would make what Brown calls the Internet of Living Things: a network of sensors, sequencing the world.

“What if you can track every beef burger? What is it, where did it come from, and what’s growing on it?” he asks. “What if every package that goes through an airport had a swab taken off it? Or the air supply in a hospital? We have interest from plant breeders who want to track what’s happening to their crops. I spoke to someone in the defense industry who wanted to detect pathogens in real-time on the London Underground.”

For now, these ideas seem far-off, and perhaps far-fetched. Then again, so was the concept of a portable sequencer five years ago, or before that, the very idea of reading DNA by forcing it through a tiny hole. “Nobody believed it would work at all to start with,” says Brown. Whether Oxford Nanopore succeeds or not, Joe Parker says that their efforts will force their competitors to up their game, injecting fresh blood into a market that risks stagnation.

Whatever the outcome, he foresees that we will enter a “second age of genomics,” one where sequencers will become like telescopes: a formerly boutique scientific instrument that you can now buy from a toy store. That would not only make sequencing ubiquitous, but it would vastly increase the amount of publicly available genomic information. “It’s the difference between doing astronomy with only a handful of telescopes versus everyone having one,” he says. “It changes the amount of sky you can look at.”