By Kevin Davies
September 22, 2009 | Whether Clive Brown, vice president of development and informatics for Oxford Nanopore Technologies (ONT), is indeed “the most honest guy in all of next-gen sequencing,” (as described by The Genome Center’s David Dooling), is perhaps debatable. But as someone who has already stamped his mark on the sequencing world, his views certainly count for something.
Five years ago, Brown was the director of computational biology and IT for Solexa, helping to spearhead the British company’s successful entry into the next-generation sequencing market, which spurred a $650-million acquisition by Illumina. After a spell at the Wellcome Trust Sanger Institute, Brown joined fellow Solexa alum, vice president research John Milton in moving to Oxford to commercialize nanopore sequencing. An intriguing subplot to the business of next-gen is whether Milton and Brown can catch lightning in a bottle again.
Oxford Nanopore is based around the pioneering nanopore research of Oxford University chemistry professor Hagan Bayley. CEO Gordon Sanghera remains close lipped about the firm’s platform specs, but an elegant paper in Nature Nanotechnology earlier this year showed that ONT’s nanopores can neatly discriminate between the four bases of DNA, based on the degree of current inhibition across the lipid bilayer (see, “Breathtaking Biology,” Bio•IT World, Mar 2009). ONT received further validation by inking an $18-million marketing deal with Illumina.
Under the watchful eye of Oxford Nanopore’s communications director, Zoe McDougall, Brown has to be more circumspect than is his true nature. “Things are on track—without telling you what the track is,” he says helpfully. What he will say is that many of the key risks in ONT’s technology have been addressed, and his team has built the entire informatics subsystem of the instrument.
Another key milestone -- achieved earlier this year, Brown says -- was to couple an exonuclease enzyme to the nanopore so that it can successively snip off bases from the end of a DNA strand, which will then tumble into the pore for the sequence to be read. ONT was recently granted a patent for its stable bilayer design. “This is a core element of our nanopore sensing system, not just for DNA sequencing,” says McDougall. Milton calls these bilayers “the workhorse of our nanopore chemistry. We use the bilayer chip to focus on single nanopores and we also operate multiple-channel versions for higher throughput experiments.”
Before ONT can produce an instrument, it has first to become essentially a small genome center to test the product for months in house. Brown hired his former Sanger Institute colleague Roger Pettett to build up that infrastructure, as well as the software that goes on the instrument. It is “revolutionary new stuff, but we’re reluctant to talk about that at the moment,” says Brown, though he would say, “It does break the conventional instrument software paradigm.”
Brown says the data throughput on ONT’s sequencer will be high, “many tens of Megabytes/second.” Not as high as some high-tech military applications, but “significantly higher than traditional lab equipment.”
“Even before we were running chemistry,” Brown says, “we made software that simulated data streams at launch spec rates. We designed interconnects and wiring, computer boards and live software that would process that data. We did it all in parallel. So when it came to plugging a chip in, it all more or less worked.” But Brown knows from experience that the system must be “very, very flexible to change.”
Another priority is move the data processing close to the point of data generation. ”We have already put a huge effort into not outputting raw data, but outputting optimized processed data instead.” Brown has considered running some of the algorithms on GPUs, but worries about the power consumption demands. “The other option is to use FPGAs. They’re good accelerators, very low power requirements, but a bugger to program and so not very agile.” Brown says FPGAs might be used at the end of product development, but not before. “So far we haven’t had any problems in terms of compute speed when dealing with our data, either at the instrument level or centralized datasets.”
Brown says the data processing simulations have been instructive. “It’s quite early, and we’re not scared,” he jokes. Meanwhile, Brown is quietly checking out potential software partners, which he hopes will deal with the quality scored DNA sequence output. In addition to genome centers and large-scale laboratories, ONT is also targeting the bench-top. “In order to have a bench-top sequencer,” says Brown, “we have to provide pretty easy to use bioinformatics solutions. Otherwise, it’s just not going to happen.”
“One of the problems with all these existing sequencers is, even if you automate the sample prep and make the sequencer easy to use, you still end up with a file with a billion short reads in it. This is still beyond the capability of most non-bioinformatically trained postdocs to do anything sensible with.” ONT aims to generate even more sequence with longer individual reads.
Brown’s goal is to provide, for want of a better term, a “turnkey” bioinformatics solution sitting alongside the sequencer. Brown has met with several potential partners, including one unnamed company that demonstrated that its “software can deal with a whole human genome-type workflow in a day or 6 hours on a typical workstation.” Brown says that looks quite promising.
He also plans to find a partner to liaise with user IT groups and “help us to smooth the early adoption of lots of our systems. I’m more worried about the bench-top side than the high-end side.” Once ONT is fully launched, Illumina will have a large say in that part of the workflow.
Besides targeting the genome centers and the bench-top sequencer sitting next to a lab researcher, Brown thinks that service organizations such as Complete Genomics might prove another fertile market—in other words, “very large sequencing centers that are not traditional genome centers,” focusing on medical sequencing applications. “I think Complete Genomics is a perfect customer for us. In fact I think our machine’s better suited for what they want to do than theirs is!”
As Brown talks, it sounds as if ONT stands for “on track.” Surely there are problems somewhere? “I don’t want to oversell things, and remember we are still very stealthy as a company,” he responds. As at Solexa, “things just aren’t linear in a company like this. You have days when things work beautifully, and long dry periods when things aren’t working. Half of it is just keeping your nerve.”
Certainly Brown has assembled a strong team to build the IT/informatics infrastructure. Nava Whiteford, another Sanger Institute recruit, is adapting existing algorithms and developing a novel file format called Fast5 for scored sequences. Physicist Stuart Reid is driving data quality measurements and some of the basic science feeding into the platform. Lukasz Szajkowski joined from Illumina to manage the writing of the instrument software, which Brown calls “one of the most risky areas, but it’s all on track thanks to him.” Molecular modeler Mick Knaggs has implemented much of his software on GPU-enabled systems.
I don’t suppose the Sanger Institute is too happy about some of their top people being poached. “Yeah, we did have a chat about my recruitment methods,” says Brown honestly.
(Read the full interview here.)
Clive Brown was not alone in feeling the paper by Stanford’s Stephen Quake in Nature Biotechnology and colleagues was “a little bizarre.” While the first report of a single-molecule human genome was “perfectly worthy of a good Nature paper,” with a “respectable throughput” of about 2 Gigabases/day, Brown was mystified by the “ridiculous back-door marketing” in the paper.
“Their own cost seems to be exactly what Illumina is citing for their service sequencing now [$48,000],” although the paper referenced outdated 2008 figures of $250,000 and up. Says Brown: “They’re using the number of names on the Solexa paper [Nature 2008] as evidence of how many people are required to run an instrument! Well, that Solexa paper was the culmination of 8 years work encompassing the entire development of the platform, so it had everybody’s name on it. The CEO’s name was on there, and he didn’t run any instruments.” He adds that the original Helicos paper (Science 2008) had more than 20 authors for the sequencing of a tiny virus.
“They’re setting themselves up for a tagline: ‘Look, we only need three people to run a Helicos machine. And Illumina needs all these people and it’s much more expensive.’ If they’d just stuck to the high ground, i.e. they’ve got a working system that does single-molecule genomes, they’d be a lot better off….”
“I think Helicos deserves some kudos. They’ve stuck with it, they’ve had a rough time as a company, and they’ve made it work about as good as it can work with single-molecule fluorescence, with the cameras they have. People have taken the technology outside and they’ve used it successfully. And that’s not trivial…. They should stick to the high ground—you can quote me on that.”
This article also appeared in the September-October 2009 issue of Bio-IT World Magazine.
Subscriptions are free for qualifying individuals. Apply today.