Technology for mapping genes is improving faster than Moore’s Law. New options range from advanced semiconductors and new materials to nanopores.
Slightly more than a decade ago an international consortium reached a major milestone by sequencing the human genome. Using laboratory systems called DNA sequencers, the Human Genome Project (HGP) determined the order of nearly 3 billion base pairs that make up the human genome. This, in turn, was supposed to pave the way to prevent, treat and cure diseases.
Then, in early 2014, Illumina hit another milestone, when the biotech company claimed that it sequenced a human-sized genome at a cost of $1,000. This compares to a cost of about $10 million a decade ago.
DNA is a molecule that carries the genetic instructions in living things. A complete set of DNA is called the genome. DNA sequencing is a technique that determines the order of DNA bases, which, in turn, is important for the study of genetics, gene mutations and therapies.
In any case, Illumina’s claims are somewhat debatable, as the actual cost is about $4,000 or more to sequence the human genome. In fact, there are still a multitude of challenges in the DNA sequencing field. One of them is to bring affordable DNA sequencing for genetic analysis to the masses. Still, Illumina’s DNA sequencing technology, dubbed sequencing by synthesis (SBS), has dominated the market and scaled much further than expected.
But seeking to dethrone Illumina in the fledging arena, several companies have been developing next-generation DNA sequencers based on various and competing technologies. The technology contenders include DNA nanoballs, nanopore, single molecule and semiconductor-based devices. And in the lab, researchers also are working on DNA sequencers, based on carbon nanotubes, graphene, microfluidics, among others.
The goal is to bring the cost-per-genome closer to the $1,000 target in the near term, but the industry is now talking about a $100 genome in the future. So, in the next-generation DNA sequencing race, the eventual technology winners and losers will be determined by cost, functionality, manufacturability and scalability.
“Some of the (DNA sequencing) technologies are working today. Some of them are not working — yet. In this case, people haven’t figured out how to design the device, much less build it. You can build one to do an experiment. But figuring out how you would cost-effectively make arrays with enough high-quality features that would work in a device is very challenging,” said Jeffery Schloss, director of the Division of Genome Sciences at the National Human Genome Research Institute (NHGRI). The NHGRI — one of 27 institutes and centers that make up the National Institutes of Health (NIH) in the United States — is involved in several projects, including the $1,000 genome sequencing technology development program.
Hopes and challenges
DNA sequencing is used in agriculture, forensics and medicine. The DNA sequencing market, which is a $20 billion business, is growing by 25% to 30% per year, according to analysts. Illumina holds 80% of the DNA sequencing market, followed by Thermo Fisher Scientific, according to Ross Muken, an analyst at Evercore ISI. Complete Genomics, Nabsys, Oxford Nanopore, Pacific Biosystems and Roche also compete in the market.
In the original HGP, DNA (deoxyribonucleic acid) sequencing was conducted using Sanger-based chemistries and capillary instruments. The HGP and other research revealed that the human genome is organized into 23 pairs of chromosomes. Chromosomes are divided into 30,000 smaller regions called genes. Each gene consists of a string of nucleotide bases, dubbed A, C, G and T. In total, human DNA has 3 billion bases. The exact order of these bases is known as the DNA sequence, according to Illumina.
Sager-based chemistries are still used for some applications, but this technology is too expensive and has been displaced by newer techniques. In general, DNA sequencing has enabled new breakthroughs, at least in the lab. “The goal (of DNA sequencing) is to unlock complex diseases that have genetic characteristics,” Muken said. “Oncology would be a good example. Hopefully over time, the idea is to develop therapies and means of treating patients such that we can be more preventative and have better clinical outcomes. We are getting close to some clinical discoveries. We are starting to get some significant breakthroughs and adoption on the pharma side.”
One of the big challenges is to reach the true $1,000 cost-per-genome target, which could occur within the next two to three years, NHGRI’s Schloss said. The $1,000 target is significant for good reason. At that cost, insurance companies may start reimbursing patients for genetic analysis using DNA sequencing.
There are other challenges. In DNA sequencing, a system generates millions of short reads of the individual DNA bases. A computer must assemble and map the data to a reference genome. But there are gaps in the reference genome. And repetitive short reads are sometimes excluded from sequencing, resulting in incomplete data.
In mainstream applications, there are also some roadblocks. For example, many physicians are offering, or could one day use, clinical genome or exome sequencing (CGES) for patients. CGES is an abridged, and a less expensive, version of whole-genome sequencing. “(CGES) is sequencing the exomes, or the protein coding regions, of the genome, plus DNA nearby the protein coding regions. That’s about 5% of the genome. You can do that now on the order of $500 to $600 per person,” Schloss said. “For the patients for whom it works, it’s quite informative. But it only works in about 25% of patients. We don’t know exactly why it doesn’t work for the other patients. One obvious guess is that the problem is not within that 5% of the genome. It resides somewhere else in the genome.”
On the other hand, CGES has identified causative or contributory gene variants in a host of diseases, including Charcot-Marie-Tooth disease, mental retardation, metabolic disorders, epilepsy, cardiomyopathy, cancer and amyotrophic lateral sclerosis (ALS), according to the NHGRI.
Wanted: New breakthroughs
Today’s next-generation DNA sequencers range from bench-top units for targeted DNA work, to high-end systems for large-scale, whole-genome sequencing. With a high-end system, Illumina said it cracked the $1,000 barrier. Based on an optical methodology, Illumina’s SBS technology uses fluorescently labeled nucleotides to sequence hundreds of millions of clusters on a flow cell surface in parallel. “There are huge data quality issues. Right now, you get a better data quality per dollar with SBS,” NHGRI’s Schloss said. “(Illumina) has done an astounding job of continuing to scale the technology. At some point, it seems as though it should stop scaling. But it hasn’t stopped yet.”
Still, Illumina’s solution is expensive, according to one competitor. “You have to buy $10 million worth of equipment and run tens of thousands of samples over a number of years to get to $1,000,” said Andy Felton, head of product management for genetic analysis at Thermo Fisher Scientific.
Meanwhile, with a related technology, Pacific Biosciences also reached a milestone. Generally, the average read length for sequencing is 100 to 200 bases. Pacific Biosciences introduced a new chemistry that enables average read lengths from 10,000 to 15,000 bases, with the longest reads exceeding 40,000 bases.
Pacific Biosciences’ DNA sequencers are based on single molecule, real-time (SMRT) technology. DNA sequencing is performed on tiny SMRT cells. Each cell is patterned with 150,000 zero-mode waveguides. “They are the furthest along in producing these long reads,” Schloss said, “(but) it costs about $50,000 to sequence a human genome. This is way more than $4,000. Still, people are using that information to build better reference genomes.”
Others are developing DNA sequencers based on semiconductor technology. DNA sequencing chips are single-use, CMOS-based devices built on trailing-edge geometries. In this area, companies borrow many of the same techniques used in the IC industry. OEMs utilize a CMOS-based fab flow, enabling them to scale their devices.
Semiconductor-based DNA sequencers have some challenges. In the lab, for example, Illumina is developing a DNA chip. “The biggest challenge is the interface,” said Mostafa Ronaghi, senior vice president and chief technology officer at Illumina. “You are basically marrying dry science with a wet science. The other challenge is the model data.”
Others have a different viewpoint. “The goal is to do it faster, cheaper and better,” Thermo Fisher Scientific’s Felton said. “The promise of our technology, which is based on semiconductors, is to do a $1,000 (genome) in a single run. We are not that far away.”
In 2011, Life Technologies, part of Thermo Fisher Scientific, rolled out its first DNA sequencing chip, dubbed the Ion Torrent. The top portion of the CMOS-based device has millions of tiny microwells, each of which resembles a vertical trench. In the flow, the microwells are patterned using lithography. Then, a dielectric film stack of silicon dioxide and silicon nitride layers are deposited on a substrate. An etcher is used to create the wells.
In each microwell, there is a sphere-like structure, which is embedded with DNA. Just below that, the device consists of a layer of transistor-based pHFET sensors. Then, the bottom of the device consists of a silicon substrate.
Basically, a nucleotide flows into the microwell. “We initiate a chemical reaction in each well that releases hydrogen ions. And that changes the pH of the solution in that well. Then, we are picking up that signal,” Felton said. “It’s really the world’s smallest pH meter with millions and millions of wells on the surface of one of these semiconductor devices.”
The first Ion Torrent chip had up to 11 million wells. Each well had a feature size of 3-micron. In comparison, the current chip has 165 million wells, which have feature sizes of 1.5 microns. It also consists of analog-to-digital converters running in parallel with 24 high-speed links at 2.5-Gbs.
Now, the company is working on a new chip, dubbed the P2. The 0.5-micron device, which incorporates 660 million wells, has a power consumption from 1 to 2 Watts. “It’s been a challenging project to build a chip with 660 million wells and get the data rates in the chip at reasonable temperatures,” Felton said. “There is also biology going on inside the chip. So, we can’t run the chip at very high voltages and generate a lot of heat. It will just kill the biochemistry.”
There are other technology options as well. Oxford Nanopore is developing DNA sequencers based on nanopore technology. “These are 2nm diameter pores or holes,” NHGRI’s Schloss said. “For this, there is no polymerases molecule. It’s actually just an electronic readout of the DNA as it goes through the pore.”
The next goal for the industry is the $100 genome. It remains to be seen if the industry can reach its lofty goals and revolutionize healthcare. But the trends are going in the right direction. “If you look at the increasing output of sequencers, the technology has been outpacing Moore’s Law for the last 20 years,” Thermo Fisher Scientific’s Felton said. “We can argue if it’s going to stay at this pace in the future.”