Nanopore sequencing

Definition

Nanopore sequencing makes it possible to sequence nucleic acids - DNA (deoxyribonucleic acid) or RNA (ribonucleic acid) - directly from biological samples in real time. Approximately a quarter of all SARS-CoV-2 virus genomes sequenced worldwide to date have been done on a nanopore device. One of the latest generation in sequencing technologies, the technique determines the order of nucleotides in DNA or in RNA by measuring fluctuations in an electric current as the molecule passes through a nanopore. The nanopore, a tiny hole one billionth of a meter in diameter, is embedded in a membrane that separates two chambers containing electrolyte solutions. When a small voltage is applied, an enzyme steadily ratchets the molecule through the nanopore along with an ionic current. Specialised software works out its sequence based on how much short sequences of individual nucleotides block the flow of ions and tiny changes in electrical current. Both DNA and RNA contain adenosine (A), cytosine (C) and guanine (G) nucleotides. They also differ by one nucleotide. DNA has thymine (T), while RNA has uracil (U).

Image adapted from Oxford Nanopore Technology. It shows a double-strand piece of DNA being unzipped and a single strand passing through a nanopore sensor. The pore sends an electrical signal to show how much of the current running through the pore is blocked by individual nucleotides (the building blocks of nucleic acid - DNA and RNA). Specialised software is used to decode the signal to read the sequence.

Importance

Nanopore sequencing is rapidly becoming one of the fastest and least expensive methods for deciphering genetic variations at the molecular level. First released on to the market in 2014, the method has radically transformed the way sequencing is carried out. Unlike previous forms of sequencing nanopore sequencing does not require a sample to be initially amplified to carry out the sequencing. Moreover, it can sequence a complete nucleic acid strand in one go, sometimes over a million bases in length, so sequences no longer need to be assembled computationally from fragments just a few hundred bases in length. It also has the benefit that it can read a sequence directly from a biological sample in real-time.

Now available in highly portable devices which can be plugged into laptops to upload data to the cloud, the technology makes it possible to carry out sequencing for the first time in remote areas with limited laboratory resources, no internet access or electricity supply. It also does not need any specially trained staff. This makes nanopore sequencing an invaluable tool for the rapid identification and monitoring of pathogens responsible for new disease outbreaks as well as tracking chains of transmission. Pocket-sized nanopore sequencers, for example, enabled viral genome sequences to be rapidly obtained from blood samples taken from patients during the outbreaks of the Ebola virus in remote areas of West Africa and the Zika virus in the hard to reach forested regions of Brazil. More recently, the technology was used in China to sequence and identify SARS-CoV-2, the virus causing the COVID-19 pandemic, and its variants.

Virus outbreaks are just the start of where researchers have begun to demonstrate the utility of nanopore sequencing. What is particularly attractive about nanopore sequencing is it offers results in real-time and it can sequence both DNA and RNA. The possibility of sequencing RNA was first flagged up by in 2012. One of the advantages of using the technology to sequence RNA is that it circumvents the need for reverse transcription or amplification steps which other methods require (Garalde et al).

Aside from viruses, nanopore sequencing provides a major tool for the detection and control of bacterial infections resistant to antibiotics, which is a major growing threat to global public health. Able to rapidly identify a microbe, nanopore sequencing could soon supplant the need for the time-consuming process of culturing bacteria for identification and testing their susceptibility to antibiotics, thereby enabling physicians to shift away from the use of broad spectrum antibiotics to more specific treatment plans. Nanopore sequencers could thus become essential to the good stewardship of antibiotics and reducing the over-prescription or futile use of such drugs.

In addition to improving the diagnosis and treatment of patients, nanopore sequencing provides a means to map out the evolution of many different pathogenic microbes. It can also help track their transmission routes, which is vital to the surveillance of emerging infectious diseases and bringing epidemics under control. Nanopore sequencing has many more applications, including in fundamental biological research, ancestry analysis, and forensic science. The technology is now being used for example to explore microbes found in the most extreme and inaccessible environments on Earth, such as in the Arctic and Antarctic (Edwards et al 2019; Johnson et al) and a deep mine in South Wales (Edwards et al 2017).

When nanopore sequencing was first released to researchers in 2014, it was used to read bacterial and viral genomes, or for interrogating small targeted regions within the human genome. Since then a number of advances have been made which has enabled the reading of much larger genomes, opening up its potential use for human genomes (Bowden). The first human genome sequenced with nanopore sequencing was reported in January 2018. This was a major turning point for the technology because the human genome is used as a yardstick to assess the 'performance of DNA sequencing instruments' and is now an important tool in medicine, such as in cancer research and diagnosis (Jain et al).

Discovery

Many of the scientists behind the development of nanopore sequencing point out that its emergence was gradual and rested on the coming together of several techniques developed by different players based in both academia and industry. Nanopore sequencing therefore cannot be seen as a sudden eureka breakthrough by one individual working alone or as a linear process. Nor was its development simple and quick. Instead it was a gradual process which involved multiple academic and industrial actors, each contributing to important aspects of its development from different directions and disciplines. Each of the participants behind its evolution have their own story to tell.

The history of nanopore sequencing is also an important reminder of the crucial role of collaboration between academia and industry. What contributions each made can be traced in different ways. While publications help reveal the path taken by academics, patents provide a map of the work undertaken by the commercial scientists (Bayley interview; Brown emails; Brown, Clarke, Willcocks interview).

One of the major foundations for nanopore sequencing was the development of single-channel recording, a technique developed by Erwin Neher and Bert Sakmann in 1976, respectively based at Yale University School of Medicine and the Max Planck Institute for Biophysical Chemistry in Göttingen. Awarded the Nobel Prize in 1996, Neher and Sakman developed a means to study ion permeation mechanisms in biological membranes. They did this by clamping a micropipette, or patch pipette, to the membrane filled with an electrolyte solution, connected to an electrode and placing another electrode in a bath surrounding the cell or tissue to form an electrical circuit. Refined over the years, Neher and Sakman’s method provided a means to record and measure the amount of current flowing through a single ion channel when it opens and closes (Neher, Sakmann; Bayley, Martin).

Nanopore sequencing begins as an idea

David Deamer, an American biologist and biomolecular engineer was one of the first people to propose using a nanopore to sequence DNA. In 1989, while based at the University of California, Davis (UC Davis), Deamer sketched out a plan to use an electric field to drive a single-stranded DNA molecule through a protein nanopore incorporated into a very thin membrane made up of two layers of lipid molecules, known as a bilayer (Bayley, 2015). Knowing that individual nucleotides, the small units of DNA, differed in size, Deamer hypothesised that it would be possible to work out their sequence by measuring the extent to which they blocked an ionic current as they passed through the nanopore (Deamer, Akeson, 2000. ).

Being able to synthesise RNA from scratch was an important component of Deamer’s wider research to understand how the first forms of primitive life came to exist on earth. Deamer had first become interested in how life began in 1975 during a sabbatical he spent with Alec Bangham at the Animal Physiology Institute in Babraham, a few miles south of Cambridge University. Bangham, a British biophysicist and haematologist who had a strong interest in the physicochemical properties of membranes and cell surfaces, was best known for his discovery of liposomes in 1961 (Watts). Inspired by his conversations with Bangham, Deamer decided to focus his research on investigating the role of membranes in the origins of cellular life (Damer).

As well as attempting to synthesise RNA in a liposome, Deamer had another line of research that helped him formulate his nanopore sequencing idea. This was directed towards understanding the molecular basis of transport in biological membranes, the process by which a substance gets transferred across a cell membrane (Deamer, 1992). He was doing this work with Mark Akeson, a postdoctoral researcher in his laboratory. Their project was focused on investigating different types of channels that could enable solutes like potassium ions and hydrogen ions to pass through lipid bilayer membranes. One of these was gramicidin, an antibiotic peptide produced by the bacteria species Bacillus brevis. Gramicidin forms very small ion channel-like pores in the cell membrane. From this work Deamer knew that it was possible to get substances to travel across membranes by using a protein to create a channel in the membrane (Deamer interviews).

Initially, Deamer viewed the ion channel (nanopore) created by gramicidin as a means to deliver ATP into a liposome to synthesise RNA. It did not take long for him to realise, however, that if he was able to thread ATP, a large molecule, through a nanopore in a membrane with the help of an electric field, it might also be possible to do the same with a string of nucleotides in a piece of DNA (Deamer interviews).

Deamer envisioned nanopore sequencing working along the same lines as the Coulter counter, an instrument routinely used to count blood cells and other particles (Deamer interviews). Invented in 1954-55, the Coulter counter consists of a tube with a very small hole in its wall which is immersed in a beaker with particles suspended in an electrolyte solution. The hole acts as a sensing zone. Electrodes are placed both inside and outside the tube to form an electrical circuit. The number of cells and particles are counted by measuring transient changes in the electrical current as they pass through the hole (Beckman Coulter).

The nanopore sequencing method proposed by Deamer was radically different from the DNA sequencing approach first described by Fred Sanger, in 1977, for which he went on to win a Nobel Prize. Widely commercialised and used by laboratories around the world, Sanger’s method works by using a polymerase, an enzyme, to synthesize DNA and the addition of tags to distinguish DNA fragments with different nucleobase endings. These fragments are then placed on acrylamide gels in different lanes and an electric current is applied which separates the fragments according to their size. The sequence is deduced from black bands read from an x-ray film which captures the position of the nucleotides on each of the fragments. By the late 1980s the Sanger technique had become an automated process, but it remained slow and expensive for determining the sequence of large and complex genomes like those of humans. This was because it required DNA in biological samples to be purified, amplified, fragmented into small pieces that were chemically labeled and then separated to read the dye labels. Before the sequence of the molecule could be worked out the fragments had to be reassembled by computational techniques. In principle, nanopore sequencing offered a much simpler and less expensive method (Branton, Deamer ; Deamer interviews).

Deamer could not pursue his idea immediately because he lacked a nanopore large enough to accommodate DNA. By 1991, however, he had learned about the research of John Kasianowicz, a physical scientist based at the National Institute of Standards and Technology in Gaithersburg, Maryland. Kasianowicz was experimenting with alpha-haemolysin, a protein toxin secreted by the bacteria Staphylococcus aureus (Branton, Deamer). The protein binds to the outer membrane of target cells in mammals, including humans, to form a water-filled pore. Uncontrolled permeation of vital molecules through the channel leads to osmotic swelling and the rupture of the cell membrane. In severe staphylococcal infections it is linked to the damage of red blood cells (NIH).

Kasianowicz was conducting investigations into alpha-haemolysin as part of a project he and colleagues had launched in the late 1980s to explore the physical properties of ion channels, proteins that form pores in membranes and enable ions - molecules with a net electrical charge - to flow across the cell membrane (Griffiths; Kasianowicz, Bezrukov). Alpha-haemolysin provided an ideal channel for Deamer to try out his idea. It formed a pore in lipid bilayer membranes just the right size for a single-stranded DNA molecule to pass through and was very stable and well behaved (Griffiths).

In 1991, Deamer took the opportunity to discuss the possibility of nanopore sequencing with Daniel Branton, a close colleague and friend from Harvard University when, at Deamer’s invitation, he gave a series of lectures at the UC Davis. During the visit the two scientists discussed ways to perform single molecule sequencing. Branton had been pondering the possibility of detecting individual bases in DNA by pulling a DNA strand through an interface (London Calling), but they decided that nanopore sequencing was a feasible alternative so began to collaborate on the project.

Being able to sequence DNA with a nanopore was highly attractive because the Human Genome Project had been launched the year before. Supported with public funds, this project aimed to sequence the entire human genome, made up of 3 billion nucleotide base pairs, within 15 years. This was a highly ambitious endeavour because until then the longest genome that had been sequenced was the Vaccinia virus, sixteen thousand times smaller than the human genome (Giani et al). In order to complete the human genome scientists needed a faster and less expensive means to carry out DNA sequencing (Branton, Deamer).

Once Branton returned to Harvard, he and Deamer began to look into applying for a joint patent between their two institutions. They decided to use the Harvard patent office to initiate the process because it had more resources to work on the application than the University of California. The Harvard office put them in contact with George Church, another Harvard faculty member who had disclosed a different idea to the office about nanopore sequencing. Church's idea was to use a protein isolated from a bacteriophage to drive double- stranded DNA through its pore. Branton in fact knew Church. Having Church on board turned out to be highly fortuitous because the Harvard Patent Office staff were initially hesitant about pursuing a patent on nanopore sequencing. Part of their reluctance stemmed from their belief that nanopore sequencing was a ‘wild idea that was never going to work’. Church managed to change their minds after pointing out that if three professors from different laboratories had independently come up with the same concept then it must have merit (London Calling). Filed in March 1995, the first patent for nanopore sequencing was granted in August 1998. It incorporated both Deamer and Church's techniques (US Patent 5,795,782).

With the patent application process initiated, in late 1992 Deamer flew to Gaithersburg to test out the haemolysin nanopore with Kasianowicz at NIST. Deamer took with him two forms of commercially available synthetic RNA which had been created with polynucleotide phosphorylase, the same enzyme he had previously used for his liposome experiment. One of the RNA's was polyadenylic acid, composed of adenylic acid monomers, and the other was polyuridylic acid, composed of uridylic acid monomers. After setting up the experiment and inserting a single haemolysin channel into a planar lipid bilayer, Kasianowicz added one of the RNA species and began to increase the voltage across the membrane 10 millivolts at a time. They could see the ionic current increasing with each step, but nothing much happened until the voltage reached 100 and then 120 millivolts. Suddenly they began to see many transient blockades of the ionic current, each lasting for a few milliseconds. In some cases 90 per cent of the current was blocked. The same thing happened when they repeated the experiment with the other RNA species (Deamer emails).

The current blockades were just what they expected if long strands of the single stranded nucleic acid were being drawn through the pore by the applied voltage. If this experiment had not worked, the idea of nanopore sequencing would probably have been abandoned (Deamer interviews; Deamer emails). Excited by what he had witnessed, Deamer invited Branton to witness the next experiment. Conducted in March 1993, this was again successful. So encouraging was the result that Deamer and Branton decided to seek funding from the National Science Foundation’s small grants programme. The advantage of such a grant was that did not require peer review. For them the grant would help them resolve whether the blockades signified individual bases coming into contact with the pore as the DNA flowed through and was not resulting from a folded up wad of DNA colliding with the pore (London Calling).

Awarded $50,000 by the NSF, Deamer and Branton had enough money to work for two years with Kasianowicz (Deamer interviews). In 1994 the team repeated and extended their earlier experiments. They found that longer RNA polymers caused longer blockades of the ionic current as they passed through the nanopore than shorter polymers. The result was encouraging because it provided a means to count and characterise each nucleic acid (Deamer, Akeson, Branton, 2016a).

By 1996 the collaborators had sufficient evidence to demonstrate that an electrical field was able to drive single-stranded DNA and RNA molecules through the alpha-haemolysin pore but the double-stranded DNA double helix was too large. They also showed that the nanopore could unravel a coiled up nucleic acid so that its nucleotides flowed through the pore in a single-file sequential order. Another crucial observation they made was that the chemical nature of the nucleotides - purines (A, G) or pyrimidines (C, T, U) - could be determined by how much they impeded the iconic current running through the pore (Branton, Deamer, Marziali).

Despite the team’s achievement, both Nature and Science, two highly prestigious scientific journals failed to recognise its relevance and declined to publish their findings. Eventually their article appeared in the Proceedings of the National Academy of Sciences (PNAS) in December 1996 (Kasianowicz et al 1996). Branton persuaded the journal to accept the article based on his membership of the National Academy of Sciences. According to Deamer a few months after the paper appeared, Science published a news report about a ‘remarkable new discovery’ that they had previously turned down as a manuscript (Deamer interviews).

The early experiments helped to establish that the nanopore could act as a sensor by measuring the ionic current blockades produced by RNA or DNA molecules. In 1997, Deamer obtained his first NIH funding for working on the project and persuaded Akeson to give up a full-time position to help him move it forward. The same year Branton also decided to spend a sabbatical with Deamer to join in on the work. The three scientists took the opportunity of being together to determine whether the alpha-haemolysin pore could distinguish between purine and pyrimidine bases (Deamer interviews).

Akeson kicked off the research by synthesising an RNA molecule composed of 30 adenine nucleotides and 70 cytosine nucleotides. When this molecule was driven through the alpha-haemolysin pore, the team was delighted to see that the blockades had two conduction levels, one related to the cytosine portion and the second to the adenine portion of the molecule. Even though the blockades did not resolve single bases, their work demonstrated that nanopore sequencing might be feasible because the nanopore could discriminate between chemically or structurally distinct segments in RNA or DNA molecules. When the team published their results in 1999 they took the opportunity to define the alpha-haemolysin channel as a ‘nanopore’. This marked the first time that the term was used and it soon caught on. Today a search of the word ‘nanopore’ on Google returns over 4 million results (Akeson et al; Deamer emails).

In March 2001 Akeson, Deamer and their graduate student Wenonah Vercoutere published a paper with an image from a computer monitor which showed a hairpin of DNA being captured by the alpha-haemolysin pore in real time (Vercoutere et al). For them the image marked a major turning point. The image showed the molecule initially wiggling around in the pore vestibule, producing a partial blockade, until the double-stranded stem portion unzipped and the molecule completed its passage through the pore, a process known as translocation. At this point it caused a distinctive spike that marked its exit. The most important observation was that the time spent before unzipping occurred was related to the number of base pairs in the stem. It took only a few milliseconds for three base pairs to unzip, but several minutes for eight base pairs, and the time was predictable from the amount of energy required to hold the base pairs together (Deamer interviews ; Deamer emails).

Another research path to nanopore sequencing

While progress was being made in Santa Cruz, scientists elsewhere were pursuing other independent research avenues that were equally important to the development of nanopore sequencing. One of these was Hagan Bayley, a British chemist and major figure in haemolysin research. An expert in engineering membrane pores, in the early 1990s Bayley had helped to unravel the crystal structure of the alpha haemolysin pore using x-ray crystallography. This he did with Eric Gouaux while based at the Worcester Foundation in Massachusetts (1990-1996). They found that alpha-haemolysin was shaped like a mushroom with a stem and a cap (Bayley interview). The stem had a tunnel-like pore around 2 nanometers in diameter, and the cap contained a larger volume called the vestibule.

Bayley saw the pore as a means of for ‘stochastic sensing’, a technique that makes it possible to measure the concentration and identity of a substance by measuring the change in an ionic current as the substance binds transiently through a single protein pore (Oxford University; Bayley 2015). He first began working on stochastic sensing in earnest in the early 1990s after receiving encouragement from Harold Bright at the US Office of Naval Research (ONR) who could see its potential as a tool to detect metal ions to track submarines as they move through water. In addition to providing funds for the work, Bright insisted Bayley take out a patent on the technique because he could see it had scope for many different applications beyond military ones. Bayley’s research into stochastic sensing subsequently also received funding from the US Department of Energy (Bayley interview).

One of the people Bayley worked with in the early years was John Kasianowicz. Between 1992 and 1994 they reported that they had managed to genetically engineer the alpha-haemolysin pore for use as a metal ion biosensor and for other applications in materials synthesis (Walker et al, 1992; Bayley et al 1992; Kasianowicz et al 1994). By early 2001 Bayley had taken up a position at Texas A&M University (TAMU) where he was continuing to develop rapid and sensitive biosensors for a range of purposes, from the detection of biological warfare agents and environmental contamination to pharmaceutical screening. Bayley hoped to combine the biosensors with microfluidics to create a lab-on-a-chip technology (Bayley, Cremer).

Based on his stochastic sensing work and the ideas put forward by Deamer, Branton and Church, Bayley also began to investigate whether an engineered nanopore could be used for DNA sequencing. In July 2001 he and his team reported the successful identification of individual DNA strands up to 30 nucleotides in length, each of which differed by just one nucleotide, using an engineered nanopore (Howorka, Cheley, Bayley).

Bayley, however, only really began to seriously consider the possibility of using scholastic sensing for DNA sequencing two years later. Two factors influenced his thinking. Firstly, the NIH’s launch of its '$1000 genome’ initiative to galvanise the development of budget-friendly systems for rapid genome sequencing. Secondly, Reza Ghadiri, who was an old friend of his. The two of them had first met in the 1990s when they each worked on different ONR projects under Bright’s auspices.

Bayley and Ghadiri first began discussing the possibility of nanopore sequencing during an ONR grantee meeting after they heard a presentation about a project which involved passing RNA and DNA strands through an alpha-haemolysin pore under an applied electric current. For Ghadiri the speaker's presentation was not that original. As he recalled, 'To me at least, based on our own research in pore-mediated ion and molecular transport, it wasn’t surprising that a polyanionic DNA or RNA could be forced by an applied electric gradient to pass through a sufficiently large transmembrane pore. Moreover, a difference in observed residual current between homopurine and homopyrimidine nucleic acids could readily be justified by their respective nucleobase size differences’. But he was immediately struck by the failure of the speaker to fully address two questions that were crucial if nanopore sequencing was ever going to be feasible (Ghadiri story).

Ghadiri and Bayley were already collaborating on the use of adaptor cyclic peptides for stochastic molecular sensing. Based on this work, Ghadiri suggested to Bayley that they design two experiments to address the fundamental issues not tackled by the speaker. The first aimed to determine whether the nanopore was capable of recognising 'the identity of single nucleotides (directly or via pattern recognition) at a particular location on a strand of DNA as it is traversing through the pore’. The second was to work out whether the speed of DNA translocation could be 'exquisitely controlled by one-base-at-a-time steps to produce sufficient time resolution for nucleobase interrogation and sequencing’. This was because 'unconstrained passage of DNA through the pore was simply too fast to allow any meaningful nucleobase identification’ (Ghadiri story).

It did not take long before Bayley and Ghadiri began experiments to see if it was possible to thread and hold a strand of DNA inside a pore and examine the extent to which single nucleotides were recognised. They continued to work on the project, with ONR funding, following Bayley’s move to Oxford University in 2003 where he became professor of chemical biology.

By 2005 Bayley and Ghadiri had successfully found a way to temporarily capture a single DNA strand in the alpha-haemolysin pore which made it possible to recognise a single adenine nucleotide at a specific location on the strand (Ashkenasy, Sanchez-Quesada, Ghadiri). But the method still needed refinement. As Bayley explained to a reporter, 'The DNA moves through much too quickly for us to distinguish between individual bases and it also moves through in a chaotic way that mitigates against recognition’ (Eisenstein). Their efforts to overcome this problem was helped by the fact that in 2005 they secured $4.2 million from NIH’s $1,000 genome programme (Oxford University; ONT News, 14 Sept 2010).

In 2006 Bayley and his team published a seminal paper which demonstrated for the first time that all four bases of DNA could be very easily distinguished by using an engineered alpha-haemolysin pore equipped with a molecular adapter. The same paper also discussed an approach called 'exonuclease sequencing' which involves cleaving off individual nucleotides one at a time from the ends of DNA and RNA polynucleotide chains, and capturing the bases after cleavage (Astler, Braha, Bayley).

A year later, in 2007, Bayley and Ghadiri disclosed to a National Human Genome Research Institute grantee meeting that DNA polymerase, an enzyme, could regulate the translocation of a DNA strand through the alpha-haemolysin pore one base at a time (Ghadiri story). This work was published by the Ghadiri group in 2008 (Cockroft et al). Together with their earlier experiments, it now looked to Bayley and Ghadiri that they might finally have a promising method for sequencing DNA. Advancing it any further, however, required far greater financial resources and personnel than could be provided in an academic setting.

Development

Fortunately, Bayley was able to turn to the support of Oxford Nanopore Technologies (ONT), a company that he had helped to found some years before. The company evolved out of another spin-out which Bayley set up while in Texas to commercialise stochastic sensing for various applications. He established the company, called STS Diagnostics GmbH, in Jena in Germany together with Christian Schmidt and Bernhard Seitz, two German colleagues. The company however, did not get very far because of obstacles with German bureaucracy to obtain funding (Bayley interview; Willcocks emails).

This situation proved more favourable once Bayley moved to Oxford because the financial environment there was more promising for pursuing academic spin-outs. His endeavour was also encouraged by Graham Richards, the head of the Chemistry Department who knew just where to get funding (Richards July 2009). Importantly Richards had strong contacts with David Norwood who had helped him negotiate a deal with Beeson Gregory Group, an investment bank, to build a new research laboratory for the department. In fact the laboratory became an important hub for Bayley's research once he arrived in Oxford. Back in 1999 Norwood had established IP Group Plc, a technology transfer enterprise to commercialise the intellectual property of British universities, and in 2003 it had created a £5 million seed capital fund to invest in spin out companies from Oxford University (Lofthouse; IP2IPO Group Plc, 2006). IP Group Plc typically invested in the seed stage of the spin outs and then ran private financing to help them develop.

Building out a company

In the summer of 2003 Bayley met Spike Willcocks from IP2IPO over sandwiches in the new Chemistry Research Laboratory. At that point STS Diagnostics GmbH was negotiating a licence for Bayley’s IP from TAMU so it was not entirely clear how IP2IPO could help because Oxford University did not own any of Bayley’s IP. Bayley, however,reapproached Willcocks in the summer of 2004 because the German company had been unable to raise the capital needed to sign the license deal with TAMU and that the university had another party willing to take the license instead. Fortunately by then Bayley had filed some new IP from his work in Oxford. Bayley and Willcocks worked out a plan for Oxford University to create a spin out company so that IP2IPO could become a shareholder, provide seed finance and at the same time acquire STS Diagnostics to secure the licence from TAMU (Willcocks emails).

In the late summer of 2004, just as Bayley and Willcocks began the arduous process of convincing Oxford University to create the spin-out company, Willcocks coincidentally met Gordon Sanghera who was looking for an opportunity to become a chief executive officer in a new spin. Sanghera was immediately attracted to the idea of joining the venture suggested by Bayley and Willcocks. He also had just the right expertise for driving the company forward (Willcocks emails). A bioelectronics engineer by training, Sanghera had led the development of a new generation blood glucose device for diabetic testing for Medisense, an Oxford University spinout, and then for Abbott Laboratories after it acquired MediSense. Sanghera immediately grasped the parallels between nanopore sensing and the glucose sensors he had helped to develop. As he put it 'When I joined MediSense, glucose measurements used to require ten minutes. Along came a digital device where you put in a drop of blood and you have a result in 30 seconds’ (Medeiros).

Oxford University was soon persuaded to form the spin out based on Gordon’s enthusiasm and experience and Bayley’s IP. The spin out was to be called Oxford NanoLabs. The next step required persuading Schmidt and Seitz that IP2IPO could acquire the company, which they argued would make it possible to secure Bayley’s IP at TAMU before it disappeared to another entity. Schmidt and Seitz agreed on the basis that they could spin out the ideas they had already begun working on for STS Diagnostics into a new company with the aim of forming a collaborative partnership with Oxford Nanolabs. The new company formed out of STS Diagnostics was named Jena Nanolabs to mimic the new UK company name (Willcocks emailss).

The original business plan was presented by Bayley, Sanghera and Willcocks to Norwood in early 2005. As was customary at IP2IPO at the time this was essentially a 15 slide powerpoint presentation. The plan laid out an initial focus on diagnostics based on emerging biomarkers to enable and drive platform development (ie device and disposable chip, similar to Sanghera’s previous products at Abbott Medisense), and then as the platform improved building towards longer term value of ion channel screening and the $1000 genome sequencing product. Norwood approved the plan and the company was officially registered on 13th May 2005 with an initial investment of £500,000 from IP2IPO (Willcocks emails).

The company took up residence in a laboratory in the Central Chemistry Building vacated by chemists who had moved across the street into the new Central Research built with the help of Beeson Gregory funding. Slated for demolition, Willcocks vividly remembers 'we always half expected a wrecking ball to come through the wall at any moment!’ The company’s initial team was Bayley, Sanghera and Willcocks who, as was expected for IP2IPO employees, worked for the company 1-2 days a week to help get its finance off the ground. In January 2006 Willcocks left IP2IPO to join the company as its vice president of business and corporate development (Willcocks emails).

Sourcing investment

Founded before the financial crash of 2008, which saw investors shift away from more speculative longer term projects, the company greatly benefited from Norwood’s contacts. Importantly, he helped to secure funding from private investors willing to risk putting money into ventures that might take up to fifteen years to provide a return. Bayley quickly learnt there were many different types of investors. As he recalls, those least 'willing to part with a penny’ were those whom they met in places with 'lovely wood panelled rooms’ and dressed in 'blazers wearing cravats’ who hired scientific advisors to grill him and his team. They had more success when they visited more down to earth places where 'some guy would come out with a tennis racket, wearing trainers and a sweater with holes’. It was only later that they found out 'that this person was worth 200 or 500 million'. Such investors were much less interested in the company’s business plan and the science than the impression they formed of Bayley and his team (Bayley interview). What also impressed the investors was Bayley’s $4.2 million grant from the NIH under the $1000 genome initiative which gave some sense of how important his work was seen within his field. His laboratory was one of the few outside the United States to receive such NIH funding (Willcocks emails).

By the end of 2005, the company had secured two rounds of seed funding from IP Group Plc and in June 2006 it raised £7.7 million from various sources. This funding allowed the company to rapidly expand its team and move into new incubator space in Begbroke Science Park in Oxford. In March 2008 the company had raised a further £10 million. This happened just before the collapse of Bears Stearns, an investment bank in New York, which saw the start of the global financial crisis and recession. Armed with the funding, the company was able to rapidly expand and move into new incubator space in Begbroke Science Park. By now the company had changed its name to Oxford Nanopore Technology (ONT, Funding Rounds; Willcocks emails).

Licensing patents

It also allowed the company to quickly begin to in-license every nanopore patent it could get its hands on. Getting as many patents as possible was a major priority for Sanghera because he wanted to avoid the difficulties he had witnessed at MediSense where a lack of investment in intellectual property meant the company could only secure a minority share of the diabetes care market (Brown interview; Anon 2018).

In 2007 Sanghera and Willcocks approached Deamer, Branton and Akeson to license all of their nanopore sequencing patents. The three scientists were immediately attracted to the proposition because previously they had had to beg for funds for their nanopore sequencing work. Just as crucial, Sanghera and Willcocks appeared to be earnest about investment in technology and realistic about how long it would take to develop. This was particularly important because they had previously had their hopes dashed by Agilent Technologies, a specialist in analytical instruments, which after providing initial funding had soon withdrawn their support because no progress was made for five years. According to Willcocks, one of the reasons the deal with Agilent fell apart was because 'they were buying alpha-haemolysin monomers from Sigma Alrich catalogues and 'expecting that to just sort of magically work' (Brown, Clarke, Willcocks interview). By 2008 ONT had signed a range of exclusive licensing agreements and entered several collaborations with leading nanopore researchers across the world whom it funded to do blue sky foundational work (Brown interview).

The turn towards DNA sequencing

When the company was first founded its initial vision included the possibility of developing diagnostics. This idea, however, was soon put on a backburner after Sanghera and Willcocks went on an odyssey around Europe and the United States to try and persuade diagnostic companies to partner with the company to develop new diagnostic tests based on its new nanopore platform. It proved what Willcocks calls 'a rather depressing exercise, as all of them said much the same thing, which was that, whilst it was fascinating science, without a demonstrated test, preferably a panel of tests, and preferably all FDA approved, it was a too early for them to get involved.' The experience led the company’s executive to make the radical decision in May 2007 to 'upend the original business plan, and instead focus on $1000 genome sequencing as the first application, to return to FDA approved diagnostics later' (Willcocks emails).

The company’s decision to focus on DNA sequencing was bolstered by the appointment of Clive Brown, a geneticist and computer biologist, and John Milton, a medicinal chemist, to ONT’s management team in June 2008. Both had previously worked for Solexa, a spin-out from Cambridge University set up to commercialise sequencing by synthesis technology before it merged in 2005 with Lynx Therapeutics, a Californian-based genomics analytics company (Brown interview).

For Brown, ONT offered a new opportunity to build an improved DNA sequencing platform which he had been attempting at Solexa which had been brought to a halt because of the company’s acquisition. Like other scientists at Solexa he felt crushed by its sell-off. As he put it, they felt like 'some silver-spoon investor' had 'pulled the rug out from under their feet' (Medeiros). Following Solexa, Brown joined the Wellcome Trust Sanger Institute, a non-profit British genomics and genetics research institute near Cambridge, where he helped to build a large Solexa-ILMN sequencing facility for one of the largest human genomics projects of the time (the 1000 Genomes Project) and Milton a small start up company (Brown, Clarke, Willcocks interview).

Challenging the DNA sequencing market

Entering the DNA sequencing market was a risky venture for ONT because a small handful of Californian companies dominated the sector. Funded by venture capital, all of these companies sold the same sort of system based on sequencing by synthesis. Conceived in the late 1990s by Shankar Balasubramanian and David Klenerman at the University of Cambridge, the Solexa sequencing by synthesis method is similar to that of Sanger but uses reversible terminators and the reaction took place in the solid phase, with amplified DNA fragments called 'Clusters' allowing it to be massively parallel rather than being limited by the number of capillaries as was the case with previous instruments. The Clusters technology was acquired from Manteia SA, a swiss company in 2004. Sequencing by synthesis involves four basic steps: library preparation, cluster generation, sequencing, and data analysis. The new method allowed for millions of DNA fragments to be processed in parallel (Illumina). Solexa reversed into Lynx Therapeutics in 2005 and the combined company, also called Solexa INC, was acquired by Illumina in 2007. Very few radical changes were made to DNA sequencers over the course of the next two decades. By 2008 most sequencing machines remained large, heavy, expensive and required costly reagents (Brown interview; Brown emails).

ONT’s team was driven by their conviction that nanopore sequencing held the key to completely change the DNA sequencing landscape and disrupt the way it was done. Importantly, it offered the means to make DNA sequencing much cheaper and faster and cut out the need for expensive equipment and reagents (Brown interview).

Most advanced DNA sequencing required high capital outlay for the equipment, ranging from hundreds of thousands of dollars to several millions of dollars, as well as the regular cost of reagents. The sequencers also remained the size of a refrigerator or desktop copier so they were not easily portable (Clarke et al). In addition the output of the sequencers could not be read easily by untrained staff. They also tended to only provide short reads which, while good for a number of applications, required a lot of piecing together to construct larger genomes (Pennisi; Baker).

Because of the expense involved, most DNA sequencing tended to be the preserve of well-resourced laboratories based in higher income western countries. Any work carried out in the field required blood and other samples to be shipped to a centralised laboratory with results then having to be sent back which could take many weeks. This made the process useless to any health workers trying to combat an infectious disease outbreak in remote areas (Spence Chubb ).

Coming from humble working-class backgrounds both Sanghera and Brown were very conscious that nanopore sequencing offered a means to make DNA sequencing more democratic. From the start their dream was to develop a cheap and portable technology that could go straight to individual laboratory researchers and 'get thousands of them sequencing without depending on a handful of gatekeeper labs who sit on top of a kind of pyramid of power' (Brown interview).

Unlike the DNA sequencers already on the market, which were largely constructed from pre-made components bought off the shelf, ONT needed to build its nanopore sequencing platform completely from scratch. ONT’s engineers had already begun to develop some nanopore instruments. The first one, presented to investors in April 2006, was 'fairly primitive: a battery-powered, calculator-sized box into which a plastic chip containing one nanopore (designed to detect and measure cyclodextrin molecules) was inserted. It had a button to apply voltage. The only way to turn it off was to dismantle it'. Willcocks recalls that Bayley thought they were 'bonkers' when they proposed running a live demonstration on it to a series of investors they planned to meet in London. Much to Bayley’s surprise, Willcocks managed to get the device working half way during Sanghera’s pitch to the first set of investors and show it measuring a molecule. Willcocks had to keep the device on as they travelled to meet the next lot of investors. He later admitted 'We actually amazed ourselves that we could keep this device going outside the lab' (Medeiros). Despite the ramshackle nature of the device it was enough to convince the first investors of a significant opportunity (Willcocks emails).

Developing the nanopore sequencing platform

Building a robust nanopore sequencing platform involved solving a long list of problems which the company’s scientists did both independently and in parallel with academic researchers. Throughout the process there was strong feedback between the scientists at ONT and in academia. In both cases it involved making certain breakthroughs in fundamental science. This was just as much the case for the scientists at ONT as for the academics. Getting the technology to work as a commercial product also required ONT’s scientists to crack 'problems nobody had ever cracked before' (Brown, Clarke, Willcocks interview). Brown also points out there is an 'enormous gulf' between what it takes to publish an academic paper and what it takes to make a commercial product that can work in other people’s hands. Academics never appreciate this - indeed one of them once comically said that he’d done the 95 percent and just the last 5 percent was left’ for the company to do. In reality Brown says 'it is the other way round' (Brown emails).

One of the key challenges the company faced at the start was developing a membrane to support the nanopores that would be stable enough to ship through the post. This was a major undertaking. Until now the state of the art was to use membranes made with lipids, fat-like substances which are important building blocks of living cells. In order to make the membrane the academics needed to 'hand-fabricate a little hole in a material, usually a sort of teflon type material, and then paint a lipid over the hole' ready for a pore to be inserted to start the sequencing process (Brown, Clarke, Willcocks interview). Such membranes, however, were extremely sensitive to pH and temperature (Loman 2012).

Just how fragile such membranes were is captured by Brown who comments that the 'system would typically last seconds to minutes' before it would collapse. This meant that academics had to repeatedly reconstruct the membrane 'hundreds of times a day in order to get a tiny amount of data'. James Clarke, a chemist who began working at ONT early on, recalls the very demoralising and painful process involved in making such membranes when he was a postdoctoral researcher in Bayley’s laboratory. He remembers that 'it could take almost a whole day of work to make the membrane and then only getting half an hour of data' (Brown, Clarke, Willcocks interview).

Clearly the membranes made by academics were never going to work for a commercial product. ONT’s team therefore went on a chemistry quest to find an alternative synthetic membrane material. Their aim was to develop one that could be manufactured in a factory that would be less fragile and not pop so easily. Eventually they 'came up with a design using surface tension and other physical properties to form extremely robust, long-lived membranes on a chip array'. The chemistry behind the membrane Clarke explains is 'exactly the same as the soap bubble, but it is liquid rather than air liquid. It's two molecules thick. So it's absolutely tiny. It has something that's 100 microns and two molecules thick. It is like people holding hands across the Grand Canyon. That's the dimensions for our membrane to span and the thickness to kind of distance.' The membrane is extremely strong. In fact, as Brown points out, the membranes are 'so good that you can, at room temperature, send them through the post, and even put them on a rocket and put them on the International Space Station, even when they have pores in them. And they'll sit there for months or longer. I have had one in my glove compartment of my car for six months. And you see people pulling them out of fridges and elsewhere, desk drawers, and they're still alive. Remember that the median lifetime of an academic liquid membrane was about three minutes or 20 minutes at most. These things last for months under very harsh conditions' (Brown, Clarke, Willcocks interview).

A second hurdle the company faced was finding the right nanopore. One of the pores that the academic community explored was Mycobacterium smegmatis porin A (MspA), an outer membrane protein derived from mycobacteria, a non-pathogenic species of bacteria. It had first been shown to be useful for nanopore sequencing in 2008 by Jens Gunlach, a researcher at the University of Washington (Butler et al). Unlike the alpha-haemolysin pore, which is mushroom-shaped, the MspA is shaped like a funnel. This meant that instead of looking at lots of the bases collected in the stem of the mushroom pore it was possible to see a smaller number of them going through the funnel (Deamer interviews).

Initially, ONT explored the use of the alpha-haemolysin pore, but they soon switched to seeking new pores and engineering variants of them to improve the read out. One of these included CsgG, an outer membrane lipoprotein found in Escherichia coli but also present in many other bacterial species. ONT looked at tens of nanopore types and thousands of mutants. The CsgG pore, currently used in ONTs platforms, helps bacteria to form biofilms to bind to host cells and colonise inert surfaces. Its structure had been unravelled in 2014 by a team led by Han Remaut, a molecular biologist at Flanders Institute for Biotechnology in Belgium (Goyal et al"). The pore had the advantage that it has a very narrow and well-defined passage for a DNA strand. Encouragingly it outperformed previous pores tried by ONT (Servick; Spence Chubb). The CsgG pore was subsequently engineered to have two reading heads which helped improve the signal and accuracy of the sequence read-out (Stoddart et al; Van der Verren et al).

The company also had to design an application-specific integrated circuit (ASIC). This is an array of circuits that can sense what the nanopores are doing at very high speed and very low noise. The electronics required was thought to be at the very limit of what is possible, detecting very small numbers of electrons per unit time. The circuits had to be as good as the gold standard 'Axopatch' style electrophysiology rigs used by the academics, save only measuring one pore at a time and in a tiny chip with many circuits in parallel. Without this no nanopore products would be possible (Brown, Clarke, Willcocks interview).

A third challenge was to find the right motor protein that would make it possible to slow down the speed at which a DNA strand went through the pore and ratchet it one base at a time so that it was possible to identify individual nucleotides. It also needed to be able to prevent the DNA from contracting or partly contracting into a ball by stretching it in the electric field (Bayley emails). A lot of work also had to go into creating a flow cell, which was to become an essential component of the company’s platform. Involving an enormous amount of engineering, the flow cell was designed to have multiple microwells, each containing a single nanopore, and an integrated specific application circuit connected to the sensor array with a heat mat to regulate the temperature. Software also had to be developed from scratch to provide control and monitor what was happening in each pore. Once it had overcome all the technical challenges, ONT then had to demonstrate the validity of the new technology over previous forms of DNA sequencing before it could secure consumers (Bayley emails; Brown interview).

From the start the company worked simultaneously on two different sequencing approaches. One dubbed 'BASE', an acronym for Bayley Sequencing, used the exonuclease sequencing method first described in Bayley’s 2006 paper. The second one was directed towards advancing Deamer and Branton’s original concept for sequencing a whole strand of DNA directly as it went through the pore (Brown, Clarke, Willcocks interview).

Initially, the exonuclease approach appeared promising. In 2009 both Bayley’s laboratory and ONT’s scientists published a paper announcing that they had devised a system that could identify the four standard DNA bases (A, C, G, T). Furthermore it could also detect bases that had been methylated, that is ones tagged with a chemical marker that helps to switch genes on and off. This opened up the possibility of their technique proving useful for epigenetics, a rapidly growing field of research that seeks to understand how modifications to DNA affects diseases like cancer (Clarke et al; Davies 2010).

In the end, however, the BASE technique proved a dead-end. Despite making major advances in protein engineering and in experimental design, ONT’s team struggled to get the biology and electronics to work together. After three years of throwing 'every trick and ingenuity' and lots of money at the project, they got nowhere and decided to abandon the approach around mid-2011. Not surprisingly everyone was demoralised by the experience. For Brown it was 'soul destroying' and 'the worst years of his life'. Willcocks described it as 'staring down the barrel of a shotgun' (Brown interview; Medeiros). To make matters worse the company was coming under pressure from Illumina, an American company, with which it had signed a deal in January 2009 for its exonuclease sequencing platform. A market leader in genetic sequencing, Illumina had agreed to invest $18 million in ONT in exchange for exclusive marketing rights to its exonuclease sequencing platform (Medeiros; Watson).

However, not all was lost because ONT had continued to work on the strand sequencing approach for the past three years in the background and had also continued to fund academic work in the area. By late 2009 ONT’s team had begun to pour its efforts into strand sequencing in earnest and regularly sent feedback on their progress to its academic collaborators. According to Clarke, the process helped to kick off a bit of a race between themselves and the academics who also intensified their efforts to advance the technique. Each group of scientists influenced each other in driving the project forward (Brown, Clarke, Willcocks interview).

Early on, ONT considered but rejected a method for the stepwise enzymatic strand translocation first demonstrated by Ghadiri’s team (Cockroft et al). Instead ONT took as its starting point Akeson’s team’s work on the Klenow fragment of Escherichia coli, a DNA polymerase enzyme, and an alpha-haemolysin pore engineered by Bayley’s laboratory. By June 2010 ONT had established that the fragment appeared to be a promising motor and the pore could discriminate between different nucleotides. Bringing together the pore and the enzyme, Clarke recalls, was 'an incredibly exciting moment at the company as it felt like there was finally a sequencing chemistry that could be built into a competitive sequencer' (Clarke emails). Soon after Clarke reported the results to the academic scientists in California, whom he visited in September 2010, the motor was upgraded to an enzyme isolated from the phi29 bacteriophage. Akeson’s group demonstrated that the Phi29 enzyme could slow down the number of nucleotides passing through a pore to around 50 per second. This work was led by Dr Kate Lieberman in Akeson's team (Lieberman et al).

ONT also investigated helicase, an enzyme that helps to unwind DNA strands. By using the helicase to unzip a double-stranded DNA molecule they hoped that a single strand would be able to step one base at a time from one side of the membrane, known as the 'cis' side, to another, called the 'trans' side. Each base was expected to modulate the ionic current while it occupied the pore, sending an electrical signal that made it possible to establish the base sequence. In fact helicase had been flagged up as potentially useful in one of the patents licensed from Deamer’s team. Helicase had the advantage that, unlike the polymerase which required all four nucleoside triphosphates to be present to function, it only needed ATP to work (Deamer interviews; Deamer emails; Brown interview). Despite its promise, it remained for ONT to establish if the helicase would truly work. To their delight it did (Medeiros).

Another piece of the puzzle that also needed to be solved before nanopore sequencing could succeed was the base caller, which is key to decoding the signal. Triggered by nucleotides blocking the pore, the signal provides the means to decipher the sequence of a DNA strand. The bulk of the base caller work was undertaken within ONT. Clarke recounts that such was work would have been difficult to do by the academics. This was because construction of the base caller required the help of an algorithm trained on the back of a large data set. ONT spent a year acquiring a training set to build the base caller. Having a stable membrane was a major asset to the company in this process. Critically, it made it possible to collect lots of data very quickly without having to construct a membrane from scratch each day as was required in academic laboratories (Brown, Clarke, Willcocks interview).

Developing a portable device

As ONT got to work on collecting data for the base caller, Brown also began to rethink the design of the company’s nanopore sequencing platform so that it could be as small and simple as possible. His goal was to come up with a device whose output was very easy to read and could be used by scientists and clinicians out in remote areas. A major inspiration for his idea came from the company’s breakthrough with the membrane. Importantly, the membrane made it possible to dispense with the need for pumps, lasers, cameras, reagent bottles and waste bottles required by conventional DNA sequencers. This made it feasible to make a much cheaper and smaller instrument than anyone had conceived of before (Brown, Clarke, Willcocks interview).

By 2011 Brown had come up with a workable prototype for a portable device that could fit into the palm of a hand. Consisting of a single circuit board and a metal case, the device, subsequently called a MinION, had only one moving part, a fan (Brown interview; Brown, Clarke, Willcocks interview). Much smaller than the typical DNA sequencer and the original instrument proposed in the company’s deal with Illumina, Brown struggled to persuade Jay Flatley, the CEO of Illumina who sat on ONT’s Board, that it would ever succeed. As Brown recalls Flatley was 'incredibly dismissive and chauvinistic' about his proposal. Despite this, the ONT Board backed Brown’s idea (Medeiros). All that remained now was for Brown to prove it worked. This was a major undertaking because it necessitated the development of several separate components, each with their own time-consuming technical hurdles. Nor was it guaranteed that all the components would integrate. And then there was the challenge of manufacturing the final device (Brown interview).

In March 2011 Brown and his team managed to sequence the first piece of DNA, a synthetic strand with the MinION. A year later they successfully decoded the 48,000-base genome of the Phi X 174 phage, a virus that infects bacteria, with the device. They achieved this 'first linking the ends of the two strands of its DNA, then threading the entire genome, first one strand and then the other, through a pore in one pass' (Brown Clarke; Pennisi). This a major breakthrough because it was the first time that an unknown piece of DNA had been successfully sequenced with a nanopore platform. Before this most nanopore sequencing had been carried out on known DNA fragments correlating expected signals back to each other and with no ability to decode an unknown molecule, this was not DNA sequencing. ONT was the first group to truly sequence DNA using nanopores and that remains the case to this day (Brown, Clarke, Willcocks interview).

Nanopore sequencing goes public

By February 2012 both the ONT and academic scientific teams had made considerable progress on nanopore sequencing to each report their results to a meeting held by Advances in Genome Biology and Technology (AGBT) held in Florida. Akeson and Gundlach took the opportunity to describe a nanopore sequencing chemistry they had put together with the help of talented students at UC Santa Cruz and the University of Washington, which incorporated both Gundlach’s MspA pore and Akeson’s phi29 polymerase. Nature Biotechnology subsequently published the academics' findings in April 2012 with its front cover showing an image of nanopore sequencing (Cherf et al; Manrao et al). Alongside the academics Brown presented the sequence data for an entire genome of PhiX, a bacteriopphage, which had been base called using nanopore sequencing at ONT (Brown AGBT).

At the same meeting Brown presented a desktop sequencer, called GridION, that ONT had developed which contains about 8,000 nanopores. The GridION he claimed had the potential to sequence a whole human genome in approximately 15 minutes (Hayden). But it was his announcement of the MinION which grabbed the most attention. Predicting that it would soon be able to 'decipher almost a billion DNA bases in 6 hours and sell for $900'. Brown’s news immediately caused a stir, generating what Sanghera recalls was an immediate 'audible gasp' from the audience (Pennisi). For Brown the reaction was so 'off the scale' that he had to hide in his hotel room after his presentation (Medeiros).

Just how much impact Brown’s news had can also be gauged by the reaction of Joshua Quick, an engineer who learnt of it from a blog posted by Professor Nick Loman, a professor of microbial genomics and bioinformatics at the University of Birmingham. After reading Loman’s blog, Quick decided to abandon his job at Illumina because he saw nanopore sequencing as the future. He had spent the past four years trying to make Illumina’s DNA sequencers, designed to do sequencing by synthesis, more compact and less expensive. Particularly struck by the small size of the MinION and the fact that it did not need any additional regents once it was running, Quick saw no reason to invest any more of his energy in DNA sequencing by synthesis technology that he believed would soon become redundant (Quick interview).

A few months after galvanising great excitement about the MinION, ONT disappointingly discovered a major design flaw in its microchip. Taking nearly two years to fix, the problem severely damaged the company’s reputation. As Brown recounted, the company was accused of 'peddling “vapourware” and cold fusion'. In November 2013 Illumina ended its agreement with ONT and divested its shares, worth $56.4 million, in the company (Medeiros).

Nanopore sequencers released to researchers

Despite the knockback, in April 2014, ONT decided to launch a MinION Access Programme (MAP) for researchers to begin testing the device for themselves and find out what applications it could be used for. Releasing it to researchers twenty-five years after Deamer first sketched out his idea for nanopore sequencing, the new device marked a major milestone for nanopore sequencing. Weighing less than 100 grams, the pocket-sized device offered by ONT was totally unlike any other sequencing technology previously commercialised. Critically it was portable, much cheaper and simpler to use than anything else on the market (Kono, Arakawa).

By June 2014 over 3,000 researchers had applied for access to the device, out of which approximately 500 researchers were accepted into the MAP. Each researcher was expected to pay $1000 as a refundable deposit for the device which came with some flow cell reagents and sample preparation kits. All researchers were free to use the MinION package and publish their results without any conditions imposed on them by the company (Medeiros; Loman 2014). The strategy reflected the fact that one of the goals of the access programme was to gain feedback from MinION users so that ONT could improve the technology. Quite a few of the innovations made to the device in recent years has resulted from feedback from this community (Quick interview).

New directions for nanopore sequencing

ONT’s nanopore sequencing tools are now being used in over 100 countries by researchers in human, plant, animal, microbiological or environmental genetics. It is also being deployed by industry for food safety (Ciscon). Researchers have also experimented with it on the International Space Station where in the future it could help evaluate biological responses to spaceflight and in the detection of extraterrestrial life on other planets (Castro-Wallace).

The outbreak of COVID-19 has also given ONT an opening into diagnostics, although this is still seen as a side line from the company's main focus which is genomic sequencing. ONT first began investigating the possibility of using its technology for diagnostic purposes after being approached by the UK government to see if it could help with the testing for the disease because of the shortage of reagents for conducting PCR tests. In May 2020 the ONT team, led by Dr Dan Turner, set to work on the problem, looking to combine nanopore sequencing with loop-mediated isothermal amplification (Lamp). Introduced in 2000, Lamp has the advantage that it does not rely on the same supply chain as PCR. It is also much simpler and less expensive than conventional PCR and produces considerably larger amounts of genetic material of interest. In addition, Lamp provides a much higher throughput and more time-efficient tool for sequencing patients samples than is possible with real-time quantitative PCR machines which have been the standard machines used for the pandemic.

The final test kit developed by ONT is called LamPORE. It can detect RNA from the SARS-CoV-2 virus. LamPORE is designated to specifically detect three target regions on the virus that are highly conserved from strain to strain together and also human B-actin, a protein involved in many important cellular processes. The actin acts as a control in LamPORE to check for false positives (Turner).

An early study of LamPORE carried out on over 500 samples from Oxford Public Health England, Porton Down, and Sheffield showed the test to have 99.1 percent sensitivity and 99.6 percent specificity. A further study of the LamPORE carried out by NHS laboratories across the UK which tested out the system on more than 23,000 patients swab and saliva samples reported in December 2020 that the LamPORE system was highly accurate for the detection of SARS-CoV-2 in infected people with and without symptoms (Beggs). If all goes well with COVID-19, the ONT team hopes LamPORE could eventually be used also to screen for other viral infections such as influenza and the Respiratory Syncytial Virus (Turner).

One of the advantages with LamPORE is that it is highly scalable. It can also be deployed in both high-throughput traditional laboratory settings and in more remote local environments. In January 2021 ONT helped to set up mobile LamPORE units in four areas of the UK to provide additional testing capacity for COVID-19 in remote areas (Osbourne). If used in conjunction with ONT’s desktop device (GridION) LamPore can process 15,000 samples a day. If used with the hand-held device (MinION) it can process 2,000 samples a day (Rapid Microbiology; James et al).

LamPORE is only one of the ways in which ONT is continuing to refine its technology. Another system that it has developed is the PromethION which deploys the same technology as the MinION but makes it possible to carry out long-read, direct DNA and RNA sequencing at a much larger scale. Released in 2018, the instrument is reported to have nearly 300 times the capacity of the MinION (Bayley emails). A desktop device, the PromethION was devised to help with the rapid sequencing of human genomes, which is seen as an important step towards personalised medicine. Being able to rapidly sequence the human genome can help in the characterisation of genetic disease, particularly in the context of cancer and provide information about a person’s likely response to treatment. Having the ability to rapidly sequence large numbers of human genomes also opens up the possibility of uncovering disease patterns and relationships that would otherwise not be evident. First launched through an access programme in May 2018, the device can now sequence multiple human genomes in rapid succession. A group of researchers at the UCSC reported, in 2019, that they had managed to complete the sequence of eleven human genomes in nine days with just one PromethION (Shafin et al).

Application

One of the applicants accepted into the MAP was Nick Loman who went on to play a significant role in demonstrating the utility of the MinION. Within a month of getting the device, Loman successfully produced a DNA sequence for a bacterial strain of Pseudomonas aeruginosa found in hospital water. A Gram-negative, rod-shaped bacterium, P. aeruginosa is commonly associated with hospital acquired infections which are difficult to eradicate due to antibiotics resistance. Listed by the World Health Organisation as one of the most dangerous superbugs, P. aeruginosa is linked to pneumonia and septicaemia. Loman sequenced the bacterium as part of a project he and his team were conducting to track its spread in water outlets and showers within Queen Elizabeth Hospital in Birmingham. Many of the hospital’s burns patients were getting secondary infections with the bacteria through hydrotherapy (Quick interview). Shared on Twitter, on 11th June 2014, Loman’s sequence of the bacterium was the first publicly released data from the MinION (Medeiros).

A year later, Loman and his team designed a laboratory protocol for using the MinION to carry out rapid genome sequencing in resource-limited settings to monitor epidemic outbreaks. The method was first tried out in the case of the Ebola virus, which was responsible for an outbreak of a lethal haemorrhagic fever which first struck forested Guinea in late 2013 and then spread to neighbouring countries. After conducting pilot experiments with a historical sample of the Ebola virus from Zaire, stored in the Defence Science and Technology Laboratory in Porton Down, Josh Quick, Loman’s doctoral student flew out to Guinea, in April 2015, to assess the capacity of the MinION to help in an ongoing outbreak of the Ebola virus in West Africa. Over 11,000 people had died from the virus since the first case had been diagnosed and it was proving very difficult to bring under control despite a strong coordinated international response. One of the major stumbling blocks was that genomic sequencing was reliant on access to well-equipped laboratories which slowed down access to real-time data on how the virus was evolving and its transmission paths. This greatly hindered the speed of efforts to curb the spread of the epidemic at ground level (Quick et al 2016).

Packed into Quick’s standard airline travel luggage, weighing less than 50kg, were three MinION devices, four laptops, a thermocycler, a heat block, pipettes and sufficient reagents and consumables to perform sequencing on the ground. Armed with this equipment Quick had at his fingertips a fully functioning genomic surveillance laboratory that he rolled out within two days of arriving at Donka Hospital in Conakry. The equipment was later moved to a dedicated sequencing laboratory in Coyah Prefecture. With the support of diagnostics laboratories on the ground, the team managed to sequence 142 Ebola virus samples collected between March and October 2015. The actual sequencing took as little as 15-60 minutes. Together with amplification, sequence library preparation and sequencing the whole genome the whole workflow after receiving a patient sample took less than 24 hours. The genomic data yielded critical information about the transmission paths of the virus. Within ten days the team had established that the persisting cases of Ebola in Guinea belonged to two major lineages and that the virus regularly travelled across the border between Guinea and Sierra Leone. Based on this insight epidemiologists were able to improve the diagnostic tools for detecting the viral strain responsible for the epidemic, and to swiftly allocate resources to households and villages identified in the transmission chain (Quick et al 2016; Loman n.d.; O'Carroll; Ladner et al).

Following their success with the Ebola virus, in 2016, Loman’s group along with collaborators from the UK and Brazil set up a mobile laboratory equipped with the new genomic surveillance system to travel 2,000 km across five federal states in the Northeast region of Brazil to help track the spread of the Zika virus carried by the mosquito. First detected in Brazil in March 2015, the rapid spread of the virus had prompted the World Health Organisation to declare it a 'Public Health Emergency of International Concern' in February 2016. What was particularly worrying was the fact that many pregnant women infected with the virus were giving birth to babies with severe microcephaly, a condition where the baby has a much smaller head and brain than expected.

The mobile laboratory formed part of the Zika in Brazil Real-time Analysis (ZiBRA) project, which brought together specialists in molecular biology, bioinformatics, entomology and genomic epidemiology drawn from FIOCRUZ Bahia, Instituto Evandro Chagas and the universities of São Paulo, Birmingham and Oxford. The goal of the ZiBRA project was to sequence a thousand genomes of the virus from both historical samples and patients with a range of clinical presentations collected from a wide geographical region in Brazil. Such data was seen as important to understanding when the virus first arrived in Brazil and to prevent the outbreak from becoming an epidemic. It would also provide clues for the development of a vaccine (Faria 2017; Quick et al 2017; Goes de Jesus et al ).

One of the major challenges with the Zika outbreak was that the virus was difficult to detect because it caused clinical symptoms that were very similar to those associated with the dengue, and chikungunya viruses that were prevalent in Brazil. In addition to this, many patients infected with the Zika virus only developed mild symptoms so they went unreported. Another problem was that relatively few genomes had been sequenced for the virus. In part this reflected the problem that it was difficult to obtain the full Zika virus from clinical samples to sequence the genome. This meant that the picture of the genetic diversity of the Zika virus remained fragmentary (Faria 2016).

Officially started in June 2016, the ZiBRA project managed to test RNA from the Zika virus isolated from 1349 patient samples collected between 2015 and 2016 and from more than 650 mosquitos, the vector that carries the virus. The results revealed that the Zika virus had been first introduced into northeast Brazil around February 2014, a year before its first detection in a patient, and had then travelled on to other regions in South and Central America and also the Caribbean (Faria et al; Quick et al 2017; Goes de Jesus et al).

Following the Ebola and Zika virus outbreaks, in 2016, Loman joined forces with Professor Ian Goodfellow, a virologist at Cambridge University and Professor Andrew Rambaut, an evolutionary biologist at the University of Edinburgh, to establish an international network of academics and public health researchers to deploy the same rapid genomic surveillance system with the MinION for other emerging diseases in remote and resource-limited locations. Known as the ARTIC Network, and funded by the Wellcome Trust, the aim was to develop a real-time molecular epidemiology tool to improve knowledge about epidemics as they unfold. Made up of scientists from the universities of Edinburgh, Birmingham, Cambridge, Oxford, KU Leuven, UCLA and the Fred Hutchinson Cancer Centre, the ARTIC Network soon became engaged in helping with an ongoing outbreak of Ebola in the Democratic Republic of Congo and the surveillances of polio in Pakistan and measles in Rwanda (ARTIC Network).

In January 2020, the ARTIC Network also quickly became involved in the unfolding of the COVID-19 pandemic, releasing a protocol for using ONT’s nanopore sequencing tools (GridION and MinION)to sequence the SARS-CoV-2 virus. Providing a set of materials, including a set of primers, laboratory protocols, bioinformatics tutorials and datasets, the ARTIC Network was key to the rapid sequencing of the SARS-CoV-2 virus, firstly in China and then globally (Cision). As part of this operation, ONT sent out 200 MinIONs to China to a broad network of public health laboratories in China to enable the rapid sequencing of the virus at the local level. Nanopore sequencing was pivotal to assembling the sequence of the nation’s genome of the virus (ONT News 10 Feb 2020).

Both the GridION and the MinION are also being used alongside Illumina sequencers, together with the Network’s protocol, in the UK by the COVID-19 Genomics UK Consortium (COG-UK). The Consortium sprang into action in early March 2020 with funding from the UK government and the Wellcome Sanger Institute to provide large-scale and rapid whole-genome virus sequencing to Public Health Agencies, the NHS centres and the government. Led by Sharon Peacock, Professor of Public Health and Microbiology at the University of Cambridge, COG-UK was established in the wake of a number of urgent phone calls and a subsequent round-table meeting between pathogen genome experts and specialists in the application of real-time sequencing (Peacock).

The consortium provides information on a proportion of SARS0oV-2 viruses causing COVID-19 in people across the UK. Genome data are generated in numerous academic laboratories, the four Public Health Agencies of the UK, and the Wellcome Sanger Institute. Combined with clinical and epidemiological datasets, the viral genome data collected by COG-UK has been vital for both tracking the evolution of new variants and spread of the virus and vaccine development. It has also helped in the identification of transmission networks involving both patients and staff in hospitals in the community. The Consortium was instrumental in the identification of the fast spreading B.1.1.7 variant around the country from November 2020.

Another country that has been sequencing the SARS-CoV-2 virus has been Denmark. Here the MinION has featured very heavily in the sequencing work. Led by Mads Albersen at the University of Aalborg, Denmark has been sequencing the genome of the virus from every positive sample collected from asymptomatic and symptomatic people tested positive either in hospitals or testing centres. Albertsen managed to use money he already had secured as a grant from the Grundfos Foundation for his previous research to kick-start the sequencing in March 2020 (Albertsen interview).

Using the ARTIC protocol and the MinION to do the sequencing, Albertsen's team successfully sequenced the first virus genomes within just a week of starting the project. Having scaled up the sequencing since the early days, Albertsen's laboratory is now able to sequence around 5,000 samples per week. In January 2021 they were sequencing approximately 10,000 samples a week to monitor the situation with the new B.1.1.7 variant. All of this work is carried out by a team of just 10 people operating 25 MinIONs in parallel. One of the advantages Albertsen sees with the MinION is that it is highly flexible and cost effective. Importantly, it makes it possible to run small batches of about 200 samples at a time. By contrast, in the case of the Illumina sequencer, they would have to wait for a batch of 1000 samples to be collected before they could start the machine. So far the vast majority of Denmark's sequencing of the SARS-CoV-2 virus has been achieved with the MinION. The process is also very quick, with the sequencing carried out in the afternoon and the results provided the next morning (Albertsen interview).

All the UK and Danish data is shared globally through the GISAID (Global Initiative on Sharing Avian Influenza Data) consortium. To date approximately 150,000 of the SARS-Cov-2 virus genomes in the GISAID database, which includes samples from around 77 countries, have been done on a nanopore device (around a quarter of the world's total) (McDougall email ).

Issues

Despite the growing popularity of nanopore sequencing, the technology is still in its infancy so it is not fool-proof. One of the issues noted early on with nanopore sequencing was that it had a lower read accuracy when compared with short-read technologies. The accuracy is largely dependent on the type of molecules being sequenced and library preparation methods. A number of improvements, however, have been made in recent years, such as the adoption of new protein pores, and improvements in library preparation, sequencing speed and software for deciphering the iconic current signal, which have made it more accurate (Bio-IT; Rang et al; Kono, Arakawa).

Overall nanopore sequencing is relatively inexpensive compared to other forms of DNA sequencing. Nonetheless, it does require disposable flow cells which have to be replaced on a regular basis. These flow cells add to the cost of doing nanopore sequencing, which can make the cost of sequencing a million bases similar to competing instruments (Deamer Branton).

Being a relatively new technology, nanopore sequencing also still has to gain acceptance more generally. Part of the problem is it challenges the prevailing culture of how sequencing is done. Many researchers have so far been hesitant about using the technology because they prefer to continue with PCR with which they have more familiarity. The ability to do nanopore sequencing away from centralised large laboratories also goes against the current mindset. Just how powerful such dogma is can be seen from the resistance ONT encountered with the UK government when at the start of the COVID-19 pandemic it offered to provide decentralised tests. The government instead opted to build large laboratories equipped with PCR machines because it was the tried and tested model. For ONT to develop a quick and easy portable test to diagnose COVID-19 or another disease at points of care could also be a challenge because of the strong reliance on yes/no results by users which lateral flow assay tests offer and which are currently easier to read and cheaper (Brown, Clarke, Willcocks interview).

Several companies are now exploring the possibility of developing nanopore sequencers, but so far none have succeeded in creating a product to challenge ONT’s lead in the area. This reflects the fact that developing such technology is very complex and expensive. One of the key ingredients of ONT’s success was its close partnership with academics whose basic scientific research laid the important groundwork for the development of nanopore sequencing and the dedication of its scientists to making sure that strand sequencing worked properly. Another important factor was finding the right investors who were willing to risk putting their money into a venture that could take many years to bear fruit.

ONT also has the advantage of a strong patent portfolio, including fundamental nanopore sensing and solid- state nanopores to protect its products and strong in-house innovation. Just how strong its position is can be seen from the fact that it managed to successfully fend off a lawsuit issued by Illumina in 2016. Illumina alleged that ONT had infringed one of its patents for using the MspA pore in its technology. The original patent for this pore had been granted to Gundlach and the University of Washington and to the University of Alabama, Birmingham, but in 2013 Illumina gained an exclusive license to the patent. Illumina’s lawsuit ultimately failed because by then ONT was using a different pore it had engineered for its technology (Deamer interviews).

Illumina was not the only company that filed lawsuits against ONT. Another was unsuccessfully issued in 2017 by Pacific Biosciences, a major provider of long-read DNA sequencers. Such lawsuits reflect just how competitive the landscape has become for companies developing the long-read technology (Genomeweb; Han). As Bayley reflects, lawsuits have become the 'cost of doing business', which is not only costly but can be intellectually draining on companies. ONT is strongly protected by its large patent portfolio which now numbers around 1400. These cover 100 different patent families and 12 of the key patents come from the work carried out in Oxford (Bayley interview).

Acknowlegements

This piece was written in March 2021 by Lara Marks with technical input from Daniel Power. Writing the piece was enriched by the generous insights provided by Mads Alberten, Hagan Bayley, Clive Brown, James Clarke, David Deamer, M Reza Ghadiri, Zoe McDougall, Joshua Quick, Sharon Peacock and Spike Willcocks all of whom have their own story to tell in the evolution of nanopore sequencing.

References

Albertsen, Mads, interview with Lara Marks, 25 Feb 2021.Back

ARTIC Network (n.d.) 'ARTIC Network provides protocol for rapid, accurate sequencing of novel coronavirus (nCoV-2019) '.Back

Akeson, M, Branton, D, Kasianowicz, JJ, Brandin, E, Deamer, DW (1999) 'Microsecond Time-Scale Discrimination Among Polycytidylic Acid, Polyadenylic Acid, and Polyuridylic Acid as Homopolymers or as Segments Within Single RNA Molecules', Biophysical Journal, 77, 3227-33. Back

Anon (23 Feb 2016) 'Illumina Sues Oxford Nanopore for Patent Infringement', GenomeWeb.

Anon (24 May 2018) 'Secrets of UK unicorn founders – Gordon Sanghera, Oxford Nanopore Technologies', Director.Back

Anon (5 May 2020) 'Fast, cheap, high-quality human genome assembly has arrived', Genetic Engineering and Biotechnology News.Back

Ashkenasy, N, Sanchez-Quesada, J, Ghadiri, MR (Feb 2005) 'Recognizing a single base in an individual DNA strand: a step toward DNA sequencing in nanopores', Angewandte Chemie International Edition, 44-9, 1401-04.Back

Astler, Y, Braha, O, Bayley, H (2006) 'Toward Single Molecule DNA Sequencing: Direct identification of ribonucleoside and deoxyribonucleoside 5', Journal American Chemistry Society, 128, 1705-10.Back

Baker, SC (1 Feb 2017) 'Next-generation sequencing challenges', Genome News, 37/3.Back

Bayley, H (29 May 2009) 'Nanopores for thought', Chemistry World.Back

Bayley, Hagan, interview with Lara Marks, 21 Aug 2020.Back

Bayley, Hagan, emails to Lara Marks, 25 Jan 2021 and 10 Feb 2021 .Back

Bayley, H (21 April 2015) 'Nanopore sequencing: from imagination to reality', Clinical Chemistry, 61/1, 25-31.Back

Bayley, H, Cremer, PS (13 Sept 2001) 'Stochastic sensors inspired by biology', Nature, 413, 226-30.Back

Bayley, H, Krishnasastry, MV, Walker, BJ, Kasianowicz, JJ (1992) 'Assembly of alpha-hemolysin: a proteinaceous pore with potential applications in materials synthesis', MRS Symposium Proceedings, 292, 243-52.Back

Bayley, H, Martin, CR (2000) 'Resistive-pulse sensings - from microbes to molecules', Chemistry Review, 100, 2575-94.Back

Beckman Coulter, The Coulter principle (1954-55).Back

Beggs, A (21 Dec 2020) 'Large study in UK NHS labs shows gold standard accuracy of Oxford Nanopore COVID-19 test LamPORE for both symptomatic and asymptomatic patients'.Back

Bio-IT World (29 Jan 2018) 'Oxford Nanopore MinION used to produce the most complete human genome', Bio-IT World.Back

Bowden R, et al (23 April 2019) 'Sequencing of human genomes with nanopore technology',Nature Communications, 10/ 1869.Back

Branton, D, Deamer, D, Marziali, et al (2008) 'The potential and challenges of nanopore sequencing', Nature Biotechnology, 26/10, 1146-53.Back

Branton, D, Deamer, D (2019) Nanopore Sequencing: An Introduction.Back

Brown CG, AGBT (2012) 'Single molecule strand sequencing using protein nanopores and scalable electronic devices', AGBT 2012 Presentation (Oxford Nanopore Technologies).Back

Brown, Clive, interview by Lara Marks, 20 July 2020.Back

Brown, Clive, emails to Lara Marks, 9 Jan and 21 Feb 2021.Back

Brown, C, Clarke, J (9 Aug 2016) 'Nanopore development at Oxford Nanopore', Nature Biotechnology, 34, 810-11.Back

Brown, Clive, Clarke, James and Willcocks, Spike, interview with Lara Marks, 11 Feb 2021.Back

Butler, T, Pavlenok, M, Derrington, IM, Niederweis, M, Gundlach (30 Dec 2008) JH,'Single-molecule DNA detection with an engineered MspA protein nanopore', Proceedings of the National Academy of Sciences, U.S., 105/ 52, 20647-52.Back

Castro-Wallace, SL (21 Dec 2017) 'Nanopore DNA sequencing and genome assembly on the International Space Station', Scientific Reports, 18022.Back

Cherf, GM, KR Lieberman, KR, Rashid, H, Lam, CE, Karpus, K, Akeson, M (April 2012) 'Automated forward and reverse ratcheting of DNA in a nanopore at five angstrom precision', Nature Biotechnology, 30/4, 344-48.Back

Chu, J, Gonzlez-Lopez, M, Cockroft, SL, Amorin, M, Ghadiri, MR (2010) 'Real-time monitoring of DNA polymerase function and stepwise single-nucleotide DNA strand translocation through a protein nanopore', Angewandte Chemie International Edition, 49, 10106 –09.Back

Church, G, Deamer, D, Branton, D, Balderelli, J, Kasianowicz, J (1998) 'Characterization of individual polymer molecules based on monomer-interface interactions', US Patent 5,795,782.Back

Cision News (23 March 2020) 'UK creates COVID-19 genome sequencing alliance for large scale analysis of the virus; Oxford Nanopore sequencing technology supports rapid sequencing capabilities in the network'.Back

Clarke, James, emails to Lara Marks, 21 Jan 2021.Back

Clarke, J, Wu, H-C, Layasinghe, L, Patel, A, Reid, S, Bayley, H (22 Feb 2009) 'Continuous base identification for single-molecule nanopore DNA sequencing', Nature Biotechnology, 4, 265-70.Back

Clive Bown, James Clark and Spike Willcocks, interview with Lara Marks 11 Feb 2021.Back

Cockroft, SL, Chu, J, Amorin, M, Ghaderi, MR (13 Sept 2007) 'A single-molecule nanopore device detects dna polymerase activity with single-nucleotide resolution', Journal American Chemistry Society, 130, 818-20.Back

Cookson, C (1 Sept 2015) 'Genes genie: Oxford Nanopore's Gordon Sanghera', Financial Times.Back

Damer, B (2 May 2019) 'David Deamer: Five decades of research on the question of how life can begin', Life, 36.Back

Davies, K (2010) The $1000 Genome: the Revolution in DNA Sequencing and the New Era of Personalized Medicine.Back

Deamer, David, interviews with Lara Marks, 2 and 8 July 2020.Back

Deamer, David, emails to Lara Marks, 4 July 2020, 13 Jan 2021.Back

Deamer D (1992) 'The role of water in proton-hydroxide conductance across model and biological membranes', in BP Gaber, KRK Easwaran, eds, Biomembrane Structure and Function - the State of the Art.Back

Deamer, D (1998) 'Daniel Branton and freeze-fracture analysis of membranes', Trends in Cell Biology, 8, 460-62.Back

Deamer, D, Branton, D (3 Nov 1967) 'Fracture planes in an ice-bilayer model membrane system', Science, 158/3801, 655-57.Back

Deamer, D, Akeson, M (2000) 'Nanopores and nucleic acids: prospects for ultrarapid sequencing', Tibtech, 18, 147-51.Back

Deamer, D, Akeson, M, Branton (6 May 2016) 'Three decades of nanopore sequencing', Nature Biotechnology, 34/5, 518–24.Back

Deamer, D, Akeson, M, Branton, D (6 May 2016a) 'Author response to John Kasianowicz and Sergey Bezrukov', Nature Biotechnology, 34/482.Back

Deamer, D, Branton, D (3 Nov 1967) 'Fracture planes in an ice-bilayer model membrane system', Science, 158/3801, 655-57.Back

Edwards, A et al (2 May 2017) 'Deep Sequencing: Intra-terrestrial metagenomics illustrates the potential of off-grid Nanopore DNA sequencing', bioRxiv.Back

Edwards, A et al (19 Jan 2019) 'In-field metagenome and 16S rRNA gene amplicon nanopore sequencing robustly characterize glacier microbiota', bioRxiv.Back

Eisenstein, M (March 2006) 'A “pore” man's sequencing solution?', Nature Methods, 3/ 157.Back

Faria, NR, Sabino ES, Nubnes MRT et al (25 May 2017) 'Mobile real-time surveillance of Zika virus in Brazil', Genome Medicine, 97.Back

Faria, NR (25 May 2017) 'A sequencing road trip at the heartland of Zika virus in the Americas', Nature Research Microbiology.Back

Genomeweb staff reporter (11 Sept 2019) 'Oxford Nanopore responds in court to Pacific Biosciences allegations of willful infringement'.Back

Garalde, DR, Snell, EA, Jachimowicz, D, et al (March 2018) 'Highly parallel direct RNA sequencing on an array of nanopores', Nature Methods, 15/3, 201.Back

Ghadiri, MR, Granja JR, Milligan RA, et al (25 Dec 1993) 'Self-assembling organic nanotubes based on a cyclic peptide architecture', Nature, 366, 324-27.Back

Ghadiri, MR, Granka JR, Buehler, LK (26 May 1994) 'Artificial transmembrane ion channels from self-assembling peptide nanotubes', Nature, 369, 301-04.Back

Ghadiri, M Reza'Story', unpublished document sent to Lara Marks, 21 Jan 2021.Back

Giani, AM, Gallo, GR, Gianfranceschi, L, Formenti, G (29 Aug 2019) 'Long walk to genomics: History and current approaches to genome sequencing and assembly', Computational and Structural Biotechnology Journal, 18, 9-19.Back

Goes de Jesus J, et al (7 Aug 2019) 'Acute vector-borne viral infection: Zika and MinION Surveillance', Advances in Molecular Epidemiology of Infectious Diseases.Back

Goyal, P, Krasteva, PV, Van Gerven, N, Gubelini, OF et al, (14 Sept 2014) 'Structural and mechanistic insights into the bacterial amyloid secretion channel CsgG', Nature 516, 250-3.Back

Griffiths, J (1 Jan 2008) 'The realm of the nanopore', Analytical Chemistry, 23-27.Back

Han, AP (18 March 2020) 'Jury invalidates pacific biosciences patents in lawsuit against Oxford Nanopore', GenomeWeb.Back

Hayden, EC (17 Feb 2012) 'Nanopore genome sequencer makes its debut', Nature.Back

Howorka, S, Cheley, S, Bayley, H (July 2001) 'Sequence-specific detection of individual DNA strands using engineered nanopores', Nature, 19, 636-39.Back

Illumina (n.d.) 'History of sequencing by synthesis'.Back

IP2IPO Group Plc (31 March 2006) 'Proposed change of name'.Back

Jain, M et al (29 Jan 2018) 'Nanopore sequencing and assembly of a human genome with ultra-long reads', Nature Biotechnology, 36, 338–45.Back

James, P, Stoddart, D, Harrington, ED, et al (30 Oct 2020) 'LamPORE: rapid, accurate and highly scalable molecular screening for SARS-CoV-2 infection, based on nanopore sequencing', medRxiv.Back

Johnson, KS et al (April 2017) 'Real-time DNA Sequencing in the Antarctic dry valleys using the Oxford Nanopore Sequencer', Journal Biomolecular Techechnology, 28/1, 2–7.Back

Kasianowicz, J.J., Walker, B.J., Krishnasastry, M.V., and Bayley, H. Genetically engineered pores as metal ion biosensors. MRS Symposium Proceedings 330, 217-223 (1994)Back

Kasianowicz, JJ, Brandin, E, Branton, D, Deamer D (Nov 1996) 'Characterization of individual polynucleotide molecules using a membrane channel', Proceedings of National Academy of Science U.S., 93, 13770-73.Back

Kasianowicz, J, Bezrukov, SN (May 2016) 'Letter to the editor: On three decades of nanopore sequencing', Nature Biotechnology, 34/5, 481-82.Back

Kono, N, Arakawa, K (29 April 2019) 'Nanopore Sequencing: Review of potential applications in functional genomics', Development, Growth & Differentiation, 61/5, 316-26.Back

Kotsev, V (5 Aug 2020) 'Oxford Nanopore's portable Covid-19 test launched in UK'.Back

Ladner JT et al (2019) 'Precision epidemiology for infectious disease control', Nature Medicine, 25, 206-11.Back

Lieberman, KR, Cherf, GM, Doody, MJ, Olasgasti, OF, Kolodj, Y, Akeson, M (1 Dec 2010) 'Processive Replication of Single DNA Molecules in a Nanopore Catalyzed by phi29 DNA Polymerase', Journal of American Chemical Society, 132/60, 19761-72.Back

Lofthouse, R (13 June 2018) 'Dave Norwood', Alumni Oxford.Back

Loman, N (n.d.) 'How do you stop outbreaks becoming epidemics in real time?', University of Birmingham.Back

Loman, N (17 Feb 2012) 'Oxford Nanopore megaton announcement: "Why do you need a machine?" - exclusive interview for this blog!'.Back

Loman, N (21 Oct 2014) 'Should I join the MinION Access Programme?'Back

London Calling, 2018, video 'The history of nanopore sequencing'.Back

Maglia, G, Restrepo, MR, Mikhailova, Bayley, H (16 Dec 2008) 'Enhanced translocation of single DNA molecules through ?-hemolysin nanopores by manipulation of internal charge', Proceedings of National Academy of Science U.S., 105/50, 19720-25.Back

Manrao, EA, Berrington, IM, Laslo, KW, Hopper, MW et al (2012) 'Reading DNA at single-nucleotide resolution with a mutant MspA nanopore and phi29 DNA polymerase', Nature Biotechnology, 30/3, 349-53.Back

Marx, V (29 Oct 2015) 'Nanopores: a sequencer in your backpack', Nature Methods, 12, 1015–18.Back

Zoe McDougall email to Lara Marks, 3 March 2021.Back

Medeiros, J (2 April 2015) 'Your genes can now be sequenced using your USB port', Wired.Back

Neher, E, Sakmann, B (1 April 1976) 'Single-channel currents recorded from membrane of denervated frog muscle fibres', Nature, 260, 799-802.Back

NIH (n.d.) 'Alpha-Hemolysin: Self-assembling transmembrane pore'.Back

O'Carroll, L (3 Feb 2016) 'From Ebola to Zika, tiny mobile lab gives real-time DNA data on outbreaks', The Guardian.Back

Olasagasti, OF, Lieberman KR, Benner, S, Cherf, GM, Dahl, JM, Deamer DW, Akeson, M (26 Sept 2010) 'Replication of individual DNA molecules under electronic control using a protein nanopore', Nature Nanotechnology, 5, 798–806.Back

Osbourne, S (30 Jan 2021) 'New Covid-19 saliva test effective at finding symptomless cases', The Independent.Back

Oxford University REF case study, 'UOA08-05: Oxford Nanopore Technologies: a successful company built on innovative DNA sequencing'.Back

Oxford Nanopore Technology, Funding Rounds.Back

Oxford Nanopore Technology, History.Back

Oxford Nanopore Technology, News (14 Sept 2010) 'Oxford Nanopore congratulates Professor Reza Ghadiri and collaborators on $5.1 million NHGRI'.Back

Oxford Nanopore Technology, News (10 Feb 2020) 'ARTIC Network provides protocol for rapid, accurate sequencing of novel coronavirus (nCoV-2019): first genomes released'.Back

Peacock, S (n.d.) 'A short history of the COVID-19 Genomics UK (COG-UK) Consortium'.Back

Pennisi, E (4 May 2012) 'Search for pore-fection', Science, 336, 6081, 524.Back

Quick, J et al (2015) 'Rapid draft sequencing and real-time nanopore sequencing in a hospital outbreak of Salmonella', Genome Biology, 114.Back

Quick, J et al (3 Feb 2016) 'Real-time, portable genome sequencing for Ebola surveillance', Nature, 530/7589, 228–32.Back

Quick, J, Grubaugh, ND, Pullan St et al (29 Jan 2017) 'Multiplex PCR method for MinION and Illumina sequencing of Zika and other virus genomes directly from clinical samples', bioRxiv.Back

Quick, Joshua, interview with Lara Marks, 21 Jan 2021.Back

Rang, FJ, Kloostrerman, WP, de Ridder J, et al (13 July 2019) 'From squiggle to basepair: computational approaches for improving nanopore sequencing read accuracy', Genome Biology, 19/90.Back

Rapid Microbiology Staff Reporter (3 Aug 2020) 'RT-LAMP With Nanosequencing - LamPORE COVID-19 Testing to be Rolled Out by UK Government'.Back

Richards, G (26 June 2009) 'Spin-outs: business built on chemistry', Chemistry World.Back

Servick, K (8 March 2016) 'Amid patent lawsuit, genetic sequencing upstart unveils new technology', Science.Back

Shafin, K, Pesout, T, Lorig-Roach, R, Haukness, M, et al (4 May 2020) 'Nanopore sequencing and the Shasta toolkit enable efficient de novo assembly of eleven human genomes', Nature Biotechnology, 38, 1044-53.Back

Song, L, Hobaugh, ME, Shustak, C, Cheley, S, Bayley, H, Goucaux (13 Dec 1996) 'Structure of Staphylococcal ?-Hemolysin, a heptameric transmembrane pore', Science, 274/5294, 1859-65.Back

Spence Chubb Law Office (8 March 2016) Letter to Lisa R Barton.Back

Stoddart D, Maglia G, Mikhailova E, Heron A, Bayley H (2010) 'Multiple base-recognition sites in a biological nanopore– two heads are better than one', Angewandte Chemie International Edition, 49, 556-59.Back

Turner, D (20 June 2020) Talk to London Calling.Back

Van der Verren SE, Van Gerven N, Jonckheere W, Hambley R et al (2020) 'A dual-constriction biological nanopore resolves homonucleotide sequences with high fidelity', Nature Biotechnology, 38, 1415-20.Back

Venkatesan, BM, Bashir, R (18 Sept 2011) 'Nanopore sensors for nucleic acid analysis', Nature Nanotechnology, 6.Back

Vercoutere, W, Winters-Hilt, S, Olsen H, Deamer, DW, Haussler, D, Akeson M (2001) 'Rapid discrimination among individual DNA molecules at single nucleotide resolution using a nanopore instrument', Nature Biotechnology, 19, 248-50.Back

Walker, B, Krishnasastry, MV, Zorn, L, Kasianowicz, J, Bayley, H (1992) 'Functional expression of the a-hemolysin of Staphylococcus aureus in intact Escherichia coli and in cell lysates', Journal of Biological Chemistry, 267, 10902-09.Back

Walker, BJ, Walsh, N, Bayley, H (1994) 'Genetically-engineered protease-activated triggers in a pore-forming protein' MRS Symposium Proceedings, 330, 209-15.Back

Watts, G (12 June 2010) 'Obituary: Alec Douglas Bangham', The Lancet, 375, 3070.Back

Watson, M (24 Feb 2016) 'Shots fired – nanopore wars part II'.Back

Willcocks, Spike emails to Lara Marks 19 and 25 Feb 2021).Back

Nanopore sequencing: timeline of key events

Science links: Science home | Cancer immunotherapy | CRISPR-Cas9 | DNA | DNA extraction | DNA polymerase | DNA Sequencing | Epigenetics | Faecal microbiota transplant | Gene therapy | Immune checkpoint inhibitors | Infectious diseases | Messenger RNA (mRNA) | Monoclonal antibodies | Organ-on-a-chip | p53 Gene | PCR | Phage display | Phage therapy | Plasmid | Recombinant DNA | Restriction enzymes | Stem cells | The human microbiome | Transgenic animals |