Jeffrey Perkel has been a scientific writer and editor since 2000. He holds a PhD in Cell and Molecular Biology from the University of Pennsylvania, and did postdoctoral work at the University of Pennsylvania and at Harvard Medical School.
It’s been nearly 10 years since Harvard University geneticist George Church and Jonathan Rothberg, of 454 Life Sciences, kicked off the next-gen DNA sequencing (NGS) revolution with a pair of papers in Science and Nature, and to say the field has exploded would be a gross understatement. Costs are down by orders of magnitude, and base outputs are through the roof.
To wit: The original publication documenting the approach of 454 Life Sciences (now part of Roche) sequenced a 580-kb bacterial genome in a single four-hour run [1]. In January 2014, Illumina announced its new top-of-the-line system, the HiSeq X™ Ten Sequencing System. Capable of sequencing complete human genomes at $1,000 a pop, the HiSeq X Ten will produce up to 600 billion bases per day.
Today, next-gen DNA sequencing has transformed the life sciences. Yet the NGS field itself is undergoing something of a transformation, too. One key player decided to mothball its first-of-its-kind platform, while another announced it was finally making its next-next-generation platform available to early-access customers. Meanwhile, the Archon Genomics XPRIZE, an ambitious sequencing challenge first announced in 2006 and set to be decided in the fall of 2013, was abruptly cancelled in August 2013, essentially because the field had nearly caught up to the competition’s goals.
Church, who led one of two teams that had registered to compete in the competition, says whether any team could actually meet the XPrize goals is hardly a foregone conclusion. But in any event, he says, he feels burned by the announcement. “If you offer a prize, you should let the teams finish.”
All in all, it’s been a busy year of sequencing advances. Here, just in time for this year’s annual Advances in Genome Biology and Technology (AGBT) meeting, we round up the latest developments.
454 Life Sciences
In October 2013, 454 Life Sciences announced it would phase out its pyrosequencing-based instruments, the GS Junior and GS FLX+, in mid-2016. According to Beth Button, head of strategic marketing and communications for sequencing at Roche, the 454 platform—the first NGS system commercially available—had essentially “run its course” in terms of scalability and cost. “It’s difficult to upgrade [the system] further than it has [been].”
Button says Roche is now looking for new partners in the sequencing space. In September 2013, the company announced a collaboration with competitor Pacific Biosciences to, as GenomeWebreported, “develop diagnostic products based on PacBio’s SMRT technology."
Illumina
Illumina announced in January the release of two new sequencing platforms based on its Solexa sequencing chemistry, the HiSeq X Ten and the NextSeq™ 500.
According to a company press release (Illumina declined to speak with Biocompare.com for this article), the $250,000 NextSeq 500 includes both sample preparation (cluster generation) and sequencing in a single desktop instrument. “Its push-button operation delivers a one-day turnaround for a number of popular sequencing applications, including one whole human genome and up to 16 exomes, up to 20 non-invasive prenatal testing samples, up to 20 transcriptomes, up to 48 gene expression samples and up to 96 targeted panels.” Runs can take as little as 12 hours for 75 sequencing cycles.
Sold in sets of 10 instruments, the HiSeqX Ten Sequencing System “[breaks] the ‘sound barrier’ of human genomics by enabling the $1,000 genome,” according to the company. The system, available in Q1 and costing $10 million, offers throughput of 1.8 trillion bases per instrument per three-day run, or about 600 Gb per day—10 times better than the HiSeq 2500.
Elaine Mardis, co-director of the Genome Institute at Washington University, says the HiSeqX Ten essentially targets the human-genome sequencing service industry, as few research facilities could hope to achieve the required throughput. “To get the $1,000 genome to play out on a day-to-day basis, you need to have the right number of samples and bandwidth,” she says. Keeping up with the data analysis could be even more difficult. “You could be quickly overwhelmed if you were to take on the analysis of everything coming through [the pipeline].”
In fact, Mardis says, her team ran the numbers and concluded they would probably need to spend at least $8 million on informatics infrastructure to keep up with the HiSeqX Ten, plus run 18,000 to 20,000 genomes per year to get the pricing down to $1,000 apiece. “I do a lot of tumor genomes,” she says, “but there aren’t enough tumor genomes to feed the beast.”
Life Technologies
Life Technologies (which was acquired by Thermo Fisher Scientific in early February) has two NGS platforms, SOLiD and Ion Torrent. But according to Andy Felton, vice president for product management in Thermo’s life sciences solutions group, the company’s efforts are devoted largely to the latter technology. “I think it’s fair to say that most R&D has shifted from SOLiD to Ion,” he says.
Current specs on the Ion PGM™ system say it can, using the latest Ion 318™ chip, produce up to 6 million 400-base reads, for 2 Gb max per four-hour run. The newer Ion Proton™ with the Ion PI™ chip produces up to 80 million 200-base reads in the same time frame, or 10 Gb to 14 Gb per run.
A new Ion PII chip for transcriptome sequencing is expected some time this year, Felton says; it will enable up to 300 million 100-base reads per chip, extending to 200 bases as the system’s chemistry improves.
Also in development from Life Technologies is a new polymerase, called Hi-Q. Providing a “significant boost in accuracy,” the Hi-Q polymerase offers 90% fewer insertion/deletion errors, Felton says. Currently available to early-access customers, Hi-Q should enter full commercialization by mid-year.
Pacific Biosciences
The current long-read champion among NGS technologies, Pacific Biosciences reads are now 8.5 kilobases thanks to a new sequencing chemistry called P5-C3, says the company’s founder and chief technology officer, Steve Turner. P5-C3 “provides a protective scaffold” on the nucleotide analogs that prevents physical contact between the enzyme and fluorophore—and thus the potential photodamage to the enzyme that can occur as a result.
“Photodamage has been reduced to an extent that we believe the limitation [in read length now] is the sample itself,” Turner says. That is, as reads get longer and longer, it gets increasingly difficult to produce sequencing templates that have no damage or abasic sites that can cause premature termination.
The current PacBio® RS II instrument and SMRT Cell configuration produce 50,000 reads, or about 400 million bases per run. And that number will only increase throughout the year, Turner says. “We are predicting that in 2014 we’re going to be adding another 50% of read length and reach 20,000 bases in 2015.”
Recently, researchers have demonstrated the feasibility of de novo eukaryotic genome assembly using only PacBio chemistry, thanks to a new assembly method called HGAP (hierarchical genome-assembly process) [2]. In January, the company announced a complete diploid assembly of the Drosophila melanogaster genome, producing contigs as long as 24.6 Mb and with an N50 length of 15.2 Mb. (N50 is a reflection of contig length.) This week, the company announced a haploid human-genome assembly to 54x coverage, producing “a 3.25 Gb assembly with a contig N50 of 4.38 Mb, and with the longest contig being 44 Mb.”
Oxford Nanopore
Oxford Nanopore Technologies, arguably the leading developer of so-called third-generation technologies based on nanopore sequencing, announced in October that it would be making its USB key-sized MinION™ sequencer available to early-access customers starting early in 2014.
Unlike most early-access programs of this type, Oxford Nanopore opted not to restrict access only to traditional sequencing centers, but rather to open the program up to everybody wide audience, as it noted in a recent email to customers.
The company is now reviewing applications and plans to begin sending out invitations “in the coming weeks.” [Editor's note: A spokesperson indicated after this article posted that the company would be sending out invitations on Friday, Feb. 14, to "a few hundred people."]
In the email, the company noted that not all MinION Access Program (MAP) applicants would be invited in the first wave, and those that are will likely receive fewer numbers of MinIONs than they requested. The instruments will be distributed in six-week cycles.
MinION was one of two sequencing systems the company described at AGBT in 2012; the other is a production instrument called GridION™. But according to a company spokesperson, the community latched on to MinION for its size, perceived flexibility and potential to “democratize” next-gen sequencing.
According to Oxford Nanopore’s web site, the company hopes to use the MinION Access Program to assess usability, suitability to different applications, logistics of shipping and reagent distribution, data analysis and so on. Those data will then inform another access program for GridION, though they have not yet announced when that will launch.
Application advances
As technology advances, so too does its application. One fast-growing application area, says Mardis, is in the clinic, where tumors increasingly are being subjected to multigene panels and exome sequencing to identify druggable targets. So, too, are children with undiagnosed genetic conditions, who increasingly are being spared the traditional “diagnostic odyssey” they used to endure in favor of whole-exome sequencing. “Do they always get an answer? No. But it’s remarkable that insurance companies are paying for this.”
Such testing has two benefits, Mardis adds. It can identify new treatments. But it also allows parents to assess the risk to future children from the same condition. Indeed, says Church, “carrier screening” is, in his opinion, the “killer app” for NGS.
“Anyone who is considering reproduction … should know whether they or their prospective sperm donor [have] a severe recessive allele,” he says. “There are now hundreds of [genes] that are … extremely predictive, and there are several services that allow you to do that accurately.”
Recently Church coauthored a study (led by Good Start Genetics, for which Church is also a scientific advisor) that is, he says, the first published application of NGS to carrier screening, an approach that “adequately drives quality up and cost down.” The team used molecular inversion (“padlock”) probes to pull out and sequence 15 clinically informative genes from 194 cell lines, 55 from individuals carrying a mutation in one of the 15 genes [3].
Another emerging application area is single-cell RNA sequencing (e.g., here) . “It’s really fascinating to show the stochastic nature of RNA expression from cells that should be identical,” Mardis says.
Church agrees, and says that he has a paper in press demonstrating a new sequencing approach that “might blow away single-cell sequencing.” Called fluorescent in situ sequencing, the technique combines fluorescence in situ hybridization and RNA-Seq, enabling researchers to, for instance, both count and determine the spatial location of individual transcripts throughout tissue slices.
Using that approach, Church’s team was able to document that transcripts are asymmetrically distributed with respect to wound healing in fibroblasts. (Further details are under embargo by Science. [ETA: See update, below])
A decade ago, such data would have been unimaginable. What’s even more unimaginable is what the next decade of NGS will bring.
References
[1] Margulies, M, et al., “Genome sequencing in microfabricated high-density picolitre reactors,” Nature, 437:376–80, 2005. [PubMed ID: 16056220]
[2] Chin, CS, et al., “Nonhybrid, finished microbial genome assemblies from long-read SMRT sequencing data,” Nat Meth, 10:563–9, 2013. [PubMed ID: 23644548]
[3] Umbarger, MA, et al., “Next-generation carrier screening,” Genet Med, published online June 13, 2013. [PubMed ID: 23765052]
Update (3/4/14): Church's paper on fluorescent in situ RNA sequencing (FISSEQ) was published online Feb. 27. Read it here.