The National Human Genome Research Institute last month awarded more than $20 million to 11 teams under its Advanced Sequencing Technology program, an initiative launched in 2004 with the goal of drastically reducing the cost of DNA sequencing.
The current round of grants is likely the last to support so-called “near-term” technologies, which aim to lower the cost of sequencing a human genome to around $100,000. Next year’s awards, outlined in a request for applications issued in early August (see In Sequence 08/12/08), will be focused on “revolutionary” technologies that will reduce that cost to $1,000 by 2014.
This year, three out of the 11 grants went to near-term sequencing projects, which fetched a total of $7.15 million, or about a third of the total funding. “I expect that that is the last time we are going to do that under these RFAs,” Jeff Schloss, NHGRI’s program director for technology development, told In Sequence last month.
“We see the landscape, and we think that basically the $100,000 genome goal is close to being met,” he said. “That does not mean that all the problems are solved. It does not mean the technology is perfect. We are certainly willing to receive applications through regular investigator-initiated grants, and we will fund ones that look really good. But in terms of a concerted push, the effort now goes toward the $1,000 genome — another 100-fold drop in cost.”
Schloss noted that when NHGRI kicked off the Advanced Sequencing Technology program in 2004, “we had stated a goal of about five years for achieving the $100,000 genome, which would put it at 2009, so we are pretty pleased that it seems likely that will be achieved.”
Indeed, several groups claim to have already broken that barrier. In February, Illumina said that it had spent $100,000 to sequence a HapMap sample, an African Yoruban man from Ibadan, Nigeria (see In Sequence 02/26/08). A month later, Applied Biosystems said that it had sequenced the same sample for $60,000 (see In Sequence 03/18/08).
To date, however, neither company has either published its genomes in a peer-reviewed journal or provided complete details on the quality of the assemblies, the variation they detected, or other parameters that would indicate whether these genomes meet the quality standard that NHGRI has set for the Advanced Sequencing Technology program — the mouse genome assembly that was published in 2002.
“I know there are people who are claiming $60,000,” Schloss said, “and I think the question is always, ‘What is the product that you get for $60,000?’”
He said that he has not had the opportunity to examine the data from the Illumina or ABI human genomes to determine whether they meet NHGRI’s quality criteria, but noted that there are other factors at stake in determining whether the $100,000 genome is actually in hand.
“The other question is, ‘How widely available is access to the $100,000 genome?’” he said. “While it is fantastic if a few large sequencing centers can produce a genome sequence of the specified quality or better for $100,000, I think our true goal is that that should be widely available.”
Another issue, he said, “is that the ability to assemble those sequences is a huge challenge because of the informatics requirements,” which means that true de novo assembly of a large mammalian genome may still be on the horizon. “I do believe that de novo assembly of genomes with the current platforms and their capabilities is still quite challenging,” he said.
Nevertheless, while there are still some hurdles ahead for $100,000 genome technologies, NHGRI has decided that these approaches will likely meet the program’s goals sometime next year, and that the emphasis for future funding will be on $1,000 genome methods.
Schloss noted that NHGRI is casting a wide net in its search for groundbreaking sequencing methods and stressed that despite the fact that the majority of this year’s $1,000 genome grants went to nanopore sequencing projects, the agency is not placing all of its bets on that approach.
While the agency has no intention to “de-emphasize” nanopore approaches in the future, “we just want to ensure that people understand that we don’t believe that nanopore is necessarily the only approach to achieving these goals,” Schloss said.
He said that in its recent solicitation for the program’s 2009 awards, NHGRI tried to “emphasize that we are looking for people to propose other approaches, because I think we need a more diverse portfolio of ideas and pursuit of those ideas.”
In order to foster these emerging approaches, the program offers smaller “feasibility” grants on the order of several hundred thousand dollars that give researchers the ability to “bootstrap and get other ideas into the mix and see if they can conceivably work,” he said.
Even so, Schloss said that nanopore-based methods do appear to be making progress — particularly in light of the relative uncertainty surrounding the approach when the program launched in 2004.
“We thought that, perhaps, after four or five years of research, that … we might have learned that there were real showstoppers to the nanopore approach, and I think we can state emphatically that we have not seen showstoppers.”
As an example, Schloss noted that several groups have done modeling “that shows one ought to be able to distinguish signals from the four bases if several conditions can be met. And that’s good, because that modeling could have said we can’t delineate a condition under which we can distinguish electrically among the bases of the four nucleotides.”
However, he added, modeling has also shown that “there are tremendous challenges to achieving those conditions of position and orientation of the nucleotides, relative to the sensors.”
In addition to solid-state systems, Schloss said that there have also been some successes in the area of protein pores, such as “work demonstrating the ability to distinguish … the one-by-one movement of the DNA molecule through a protein pore, and then, in parallel with that, the ability to distinguish deoxy nucleoside monophosphates from each other in the protein pore.”
Details of this year’s funded projects are outlined below. Further information is available on the NHGRI website here.
Eight awards went to groups focusing on the “$1,000 Genome:”
Dan Branton and Jene Golovchenko, who jointly head the Harvard Nanopore Group at Harvard University, will receive $6.5 million over four years, or about a third of the overall funding in this year’s round, to develop electronic sequencing in nanopores. Branton told In Sequence this spring that the group is working on a carbon nanotube detector for solid-state nanopore sequencing (see In Sequence 4/22/2008). The goal, according to the grant abstract, is to be able to sequence DNA at a rate of 10 kilobases per second. One hundred of such nanopores in parallel would be able to produce a draft sequence of a human genome in 20 hours at a cost of $1,000. The new grant follows a three-year, $5.2 million award that Golovchenko received in 2005 for the same technology.
“We see the landscape, and we think that basically, the $100,000 genome goal is close to being met.”
Last month, UK-based Oxford Nanopore Technologies secured intellectual property generated by the two Harvard researchers and others, and said it would fund research in the Harvard Nanopore Group (see In Sequence 8/12/2008).
Stephen Chou, a professor of electrical engineering at Princeton University, was awarded $920,000 over three years to work on a so-called “nanogap detector” for sequencing, which combines a single nanochannel with a detector that is formed by a pair of electrodes with a gap, which measures the electrical signal transverse to the DNA backbone as the molecule moves through the channel.
The grant abstract does not mention any commercialization plans. But Han Cao, who invented a nanochannel technology as a postdoc in Chou’s lab, founded BioNanomatrix in 2003, a startup that is working on a nanochannel-based sequencing technology (see In Sequence 4/8/2008).
Marija Drndic, an assistant professor of physics at the University of Pennsylvania, won $820,000 over three years to develop nanopore-nanoelectrode devices for DNA sequencing. She will collaborate with her UPenn colleague Ken Healy. Drndic told In Sequence last week that she will also use a pair of nanoelectrodes, positioned on either side of a nanopore, to measure the current across a nanometer gap.
The main advantages over other approaches would be higher resolution, because the nanoelectrodes can be thinner than the membrane that contains the nanopores, and the ability of the electrodes to constrain the DNA molecule.
“One of the most significant obstacles to achieving nanopore-based DNA sequencing is that DNA molecules do not travel smoothly through the nanopore,” Drndic explained. “But if we can constrain the molecule, then we can minimize this erratic motion.”
Though two other groups funded in the current round are also taking the nanoelectrode approach, she said, “we are confident that our transmission electron beam ablation lithography [TEBAL] technique for fabricating nanoelectrodes has the edge in terms of precision and ease of implementation.” She added that her group has been able to fabricate nanopore-nanoelectrode devices for about 18 months now.
It is too early to say when the technology might be commercialized, according to Drndic. “Even when this project is complete, further scientific research will most likely be required before the technology is ready for commercialization.”
Stuart Lindsay, a professor of physics at the Chemistry Biodesign Institute at Arizona State University, will receive $370,000 for one year to combine hydrogen-bond mediated molecular recognition with DNA translocation through a nanopore. The technology also involves a nanopore and a pair of electrodes, tethered to two types of recognition molecules, that span a nanoscale gap.
The main difference between his and other nanopore sequencing approaches is the base recognition chemistry, Lindsay told In Sequence last month. “We have a new way to make chemically specific electrical contacts to the bases,” he said. “We also have a new approach to making nanopores, but this is not quite ready for prime time.”
The biggest challenge will be “manufacturing reliable nanostructures, so cheap devices could be made,” he said.
Lindsay said that the hope is to have a prototype ready within three years. “If it lives up to its potential and is readily manufacturable, perhaps devices could appear as products not long thereafter,” he said.
This is Lindsay’s third grant under the program: In 2004 he received an award to work on a molecular reading head for single-molecule sequencing, and in 2007 he won a grant to explore a new approach to “sequencing by recognition” in nanopores. The idea funded in the current round stems from 2006, he said.
Di Gao, an assistant professor in the department of chemical and petroleum engineering at the University of Pittsburgh, will receive $370,000 over two years to develop a method that could replace electrophoresis-based Sanger sequencing by pulling DNA strands off a solid surface when stretched under an electric field. DNA strands of different lengths produced in a Sanger sequencing reaction could thus be separated. Fluorescence resonance energy transfer would then detect those DNA strands that have become detached.
Gao told In Sequence last month that the technology might be able “to increase both the read length and the read speed of the sequencing process” and “does not require complicated and expensive fabrication processes that are needed to fabricate nanopores.” The challenge will be to “integrate the electronic device with the optical signal detection equipment and to synchronize them to obtain a fast and reliable reading on the fluorescence signals.” He estimated that the technology could be commercially viable in “less than 10 years.”
Xiaohua Huang, an assistant professor of bioengineering at the University of California, San Diego, and his collaborator and UCSD colleague Pavel Pevzner, won a four-year, $2.5 million award to develop “natural sequencing by synthesis” technology, a cyclic approach that will sequence amplified DNA using DNA polymerase, natural nucleotides, and a small percentage of cleavable fluorescent nucleotides.
Huang told In Sequence last week that his new grant “differs significantly” from the two previous grants he received under the same program — for “sequencing by denaturation” in 2005 and for “single-molecule sequencing by ligation” in 2006.
He said he hopes to develop his new technology within the next four years “to the point that a whole human genome can be sequenced within a day with one single instrument using our chemistry.” He is “not working with any specific company yet.” Pevzner will contribute to the development of algorithms and software for de novo sequence assembly from paired-end short reads.
Predrag Krstiæ, a senior staff scientist in the physics division at Oak Ridge National Laboratory, will obtain $720,000 over two years for developing a nanoscale quadrupole Paul trap, in collaboration with Mark Reed of Yale University. The trap is an “alternative to nanopore sequencing with enhanced control capabilities both in translocation and detection,” according to the grant abstract. The DNA will be stabilized by combined static and radio-frequency quadrupole trapping electric fields, controlling its translocation. This resolves “one of the main obstacles in reproducible DNA sequencing in a nanogap or nanopore based on reading electrical characteristics of the bases,” the abstract stated.
The main advantage of the Paul trap is its ability to relax “critical dimension control, which simplifies the device fabrication” as well as the possibility to use arrays of Paul traps in parallel, according to the abstract.
Jiali Li, an associate professor of physics at the University of Arkansas and a former postdoc in the department of physics at Harvard, where Golovchenko works, won $830,000 over three years to focus on sequencing DNA that contains biotin-labeled nucleotides using solid-state nanopores. She will collaborate with her University of Arkansas colleague David McNabb.
In 2004, Li won another three-year grant from NHGRI for solid-state nanopore identification of proteins, though that grant was not awarded under the advanced sequencing technology program.
The biggest challenge of her new project will be “to develop a high-resolution nanopore sensing system that is capable of obtaining the DNA sequencing signal,” she told In Sequence last week, adding that the technology might not be commercially viable yet by the end of the funding period.
NHGRI also made three “$100,000 Genome” awards under the current funding round.
Mostafa Ronaghi, who recently joined Illumina from the Stanford Genome Technology Center, will receive $5.1 million over three years, or about a quarter of the total funding in this year’s round, to develop a 10-gigabase pyrosequencer. Illumina said in July that it will acquire Avantome, which Ronaghi founded earlier this year (see In Sequence 7/29/2008).
He collaborates under the grant with Helmy Eltoukhy, the former CEO of Avantome, who also joined Illumina, and Stevan Jovanovich, CEO of Microchip Biotechnologies, which has been developing an automated sample prep front-end for the sequencing system. Ronaghi won a grant under the same funding program in 2004 to develop a pyrosequencing array for genome sequencing.
According to the grant abstract, the researchers will develop a low-cost 10-gigabase pyrosequencer for de novo DNA sequencing “that would enable any lab to perform high-throughput genome analyses.” The platform will combine automated sample preparation with pyrosequencing, potentially enabling “mammalian genome sequencing in a single run.” The researchers have already developed a “sensitive CMOS image sensor” that is specialized for the pyrosequencing chemistry and is integrated with a fluidic platform. Sixteen chips, each enabling two million sequencing reactions, will be combined.
Steven Benner from the Foundation for Applied Molecular Evolution in Gainesville, Fla., won $1.1 million over three years to develop reagents and enzymes for genome sequencing. These reagents will provide new capabilities “that are not available from standard DNA or RNA, but that have technological value,” Benner told In Sequence last week. “These capabilities can be built into any sequencing architecture; existing, in-development, or not yet invented,” he said.
Among the reagents to be developed are some that will be used to move specific DNA from one place to another in a complex biological environment, such as body fluids; others that are used when many probes need to be introduced at the same time, for example in highly multiplexed PCR; and yet others that allow the synthesis of nucleic acids to be primed with the specificity of long sequences, but the discrimination power of short sequences, he said.
These reagents are unrelated to reagents that Benner developed under previous awards from NHGRI, such as DNA polymerases that can handle unnatural nucleic acids and reversible terminators for sequencing for synthesis, he said. Those reagents are “on their own commercial tracks,” he added. In 2004, he won a grant to develop polymerases for sequencing-by-synthesis and another grant for DNA sequencing using nanopores under the same program.
Benner said that his foundation is working with Firebird Biomolecular to commercialize several of the currently developed reagents. He has already “made contacts with several companies and organizations that offer, or are developing, platforms whose performance could be exploited or improved by adding one or more of these reagents, including Luminex.” He and his colleagues are using Luminex’s bead system to beta-test the reagents for small-scale sequencing-by-synthesis.
“At present, we are extremely enthusiastic about the platform being developed by Intelligent Bio-Systems,” he added, a company that has been commercializing technology developed by Jingyue Ju at Columbia University.
Ju won a two-year, $950,000 award under the current funding round, in collaboration with Columbia colleague Nicholas Turro, to work on sequencing with reversible dNTPs and cleavable fluorescent ddNTP terminators.
The new method is a hybrid between Sanger dideoxynucleotide sequencing and sequencing-by-synthesis. Ju and his colleagues published a proof-of-concept of the approach in the Proceeding of the National Academy of Sciences this summer (see In Sequence 7/1/2008). “We hope to optimize and further increase the read length of the approach during the lifetime of this grant,” he told In Sequence this week. Turro will contribute to the study of the photochemistry of the fluorescent nucleotides and the fluorescence detection, he said.
It is unclear it the technology will be commercialized by Intelligent Bio-Systems, which previously exclusively licensed an SBS chemistry developed in Ju’s lab (see In Sequence 10/23/2007). “IBS will decide on the best SBS technologies and strategies to commercialize,” Ju said.
Ju previously received a grant in 2004 under the same program for developing an integrated system for DNA sequencing-by-synthesis; another award in 2005 for modulating nucleotide size in DNA for detection by nanopores; and two awards in 2007, for developing 3’-O-modified nucleotide reversible terminators for pyrosequencing and for developing an integrated system for DNA SBS.