What’s New in Human Genome Research?
Jeffrey A. Schloss, Program Director for Technology Development Co-ordination at the National Human Genome Research Institute at the National Institute of Health, joins Pharma IQ to discuss developments in his field, the recent achievements of the Institute and the way ahead for genome research.
Pharma IQ: You claim that technological leaps will reduce the cost of DNA sequencing by four to five orders of magnitude to enable sequencing of an individual human genome for $1,000 or less. What technologies are you developing to make this happen and how long until this is a reality?
J Schloss: When we proposed this research programme in 2002, the cost to sequence a genome at reasonably high quality was more than $10m. Reducing the cost by four orders of magnitude would take the cost to the $1,000 range. Of course, lower would be better, for using sequence information for understanding human biology and disease, and potentially for personalised medicine, as well as in many other areas of science and industry. To achieve this goal, our programme supported research on a wide variety of technologies. In the early days we supported miniaturisation and integration of Sanger sequencing chemistry and electrophoretic fragment sizing – the same basic technology that was so effective in achieving the Human Genome Project. This was a rather conservative segment of the portfolio in case more ambitious technologies, such as sequencing-by-synthesis, failed to pay off. The two other approaches that received the most support were sequencing-by-synthesis (and its cousin, sequencing by ligation), and nanopore/nanogap sequencing. We have also supported research on methods to detect DNA polymerase activity in real time, and quite a number of other methods.
Pharma IQ: Your current programmes have successfully managed to commercialise several new sequencing platforms that have been put into use in laboratories worldwide. Are there any new developments in the pipeline ready to hit the market?
J Schloss: Several of the technologies that received support from our programme have been commercialised and others are close. I want to be very clear that NIH doesn’t commercialise technologies, but supports research to enable or accelerate the commercialisation that appropriately occurs in the private sector. Those commercialisation activities require a much larger investment than NIH makes in any particular technology.
NIH support has played different roles in different technologies. For example, the technology underlying important aspects of Helicos’s platform had its origins in a new investigator grant by NHGRI to Dr.Stephen Quake when he was an assistant professor at Caltech. Quake received additional sequencing technology funding, and Helico’s received grant support for its system development as well as continuing efforts on novel assays. NHGRI supported a Centres of Excellence project in Dr. George Church’s lab that developed many aspects of sequencing-by-synthesis as commercialised by several companies. For example, Church’s lab originally developed and implemented the concept of diluting DNA fragments and then amplifying them locally, to enable fluorescence detection from ensembles of identical molecules, and doing that with thousands (now millions) of fragments arrayed across a surface.
One company that picked up aspects of that work was Agencourt, which received a grant from NHGRI to develop a version of sequencing-by-ligation that was subsequently commercialised by Life Technologies. And Church worked with Danaher to commercialise an ‘open-source’ platform. NHGRI supported early development, in a grant to Ronald W Davis at Stanford University, of electronic detection of nucleotide addition to a growing chain; this method was subsequently developed by Ion Torrent which received an NIH grant to accelerate technology scaling (and is being commercialised by Life Technologies). And 454 received NIH support that accelerated development and subsequent commercialisation of a higher-throughput version of its original commercial system. I want to point out that NHGRI has provided grant support for research in a number of other academic and private sector labs, such as Li-Cor, that, while not resulting in commercialisation of independent systems, has nevertheless found its way into commercial systems by licensing of inventions, as well as by contributing to knowledge in the field through peer-reviewed publications.
The next systems coming onto the scene detect incorporation of fluorescent nucleotides into single DNA strands in real time by single DNA polymerase molecules. NHGRI supported an SBIR grant to a small spin-out from Cornell University, and subsequently provided grant support to Pacific Biosciences whose system is now in early access in almost a dozen labs worldwide. NHGRI also provided an STTR grant and subsequently a research grant to Visigen, and this technology is being commercialised by Life Technologies with anticipated release later this year. These systems offer potential advantages over today’s systems – particularly longer read lengths and other types of genomic information.
NHGRI has made a substantial investment in nanopore-related technologies. The main variants here are protein nanopores, synthetic nanopores, and nanoelectrode gaps in nanochannels. The underlying concept of nanopore sequencing – single DNA molecules threading through a sensor whose diameter is that of a DNA molecule, with rapid electronic read-out of nucleotide sequence – was introduced in the mid-1990s, at which time support from NHGRI began. In conceiving the current technology development programme, NHGRI recognised that this approach, while very attractive if successful, was risky and would need support for several years of basic research to understand the physics and develop fabrication and detection methods. Progress has accelerated recently with demonstrations of ability to distinguish the electronic signals from individual nucleotides and of fine control of DNA translocation past the sensor, and of means for deploying multiplexed systems. Even longer sequence reads and discrimination of epigenetic marks, as well as higher speed and lower cost, are among the potential advantages, if nanopore sequencing can be achieved and commercialised.
Among other sequencing technologies receiving NHGRI support are additional approaches for monitoring polymerase action in real time, and methods based on mass spectrometry, electron microscopy and force spectroscopy.
Pharma IQ: Looking at the bioinformatics challenges of next generation sequencing; with shorter reads requiring more reads and a much higher depth of coverage for acceptable quality sequences, how are you targeting technological scalability so advances in technology don’t surmount analytical capabilities?
J Schloss: In the ‘old days’ of DNA sequencing – five to ten years ago – we felt comfortable that the pace of development of computing capacity would stay ahead of the needs of the sequencing community. However, in the past very few years the cost of producing sequence data has fallen at a rate that exceeds the Moore’s Law curve. So the community of technology developers and users have collectively moved the bottleneck from the production of sequence data to its analysis. While this opens immense opportunities to study biology and human disease, and potentially to improve patient care, there’s no question that it introduces critical challenges. We anticipate that some of the emerging, or next-to-emerge, sequencing technologies may reduce the proximal challenge of putting together a genome from reads information, if the read length is significantly longer and data quality is at least as good as that from the short-read platforms. This will help, but the greater challenge – interpreting the information from large numbers of genomes once each genome is ‘assembled’ – remains, and must be solved by means other than improving the sequencing technology, per se. That’s outside the scope of the sequencing technology development programme, but very much on the radar of NHGRI as we move into our next strategic plan implementation of taking human genomic information toward the clinic.
Pharma IQ: Jeff, at the 2nd Annual Next-Generation DNA Sequencing conference in March you will be delivering a presentation on the Revolutionary DNA Sequencing Technologies Emerging and on the Horizon. What impact does the work of the NHGRI have on the industry as a whole, and what can people learn from the best practice methods you have employed?
J Schloss: The NHGRI, and federal agencies in general, I think, have a small but crucial role to play in stimulating the development of new technologies that are critical for the advancement of our missions. Among these roles are (1) identifying areas in which effort is needed and making a clear case for why the need exists, (2) articulating goals (technical, cost, quality, timing), (3) providing financial support with a longer view than the private sector may be able or willing to offer, (4) requiring that those who receive support share information with each other to ‘lift all the boats,’ and (5) ‘keeping it real’ by judiciously connecting the technology development effort with users. Your question also gives me the opportunity to attribute progress in sequencing technology development to other agencies and entities in addition to the NIH. For example, the NSF, DOE and DOD (DARPA in particular) in the US, and the Wellcome Trust and EC all have lent support to this effort; I’m sure there are others.
Pharma IQ: Finally, to round off, at the event in March, what do you expect will be the most valuable discussion points, and what do you hope to gain from being there?
J Schloss: The various groups that are adopting sequencing technologies have a huge challenge because the technology is improving so fast, and there are so many different technologies available each with subtle advantages and disadvantages, that the field is sparking extensive creativity. I look forward to hearing what biomedical problems your participants are trying to solve, how they are evaluating available platforms and creating their own tools to adapt those platforms to achieve their goals, from sample to base-calling to analysis. I am particularly interested in how they balance current productivity with future flexibility.