This article was originally published Feb. 6.
Following the burst of next-generation sequencing technologies onto the scene, Sanger sequencing may have slipped out of the limelight, but it is still laboring away in a few laboratories.
Robert Lyons, the director of the University of Michigan DNA Sequencing Core, said that while he has seen a decline in demand for Sanger sequencing, it has not been large. "The business has gone down about 10 percent," he said. "We still do 300,000 samples or 300,000 reactions a year here."
Similarly, during an earnings call earlier this month, Greg Lucier, the chief executive of Life Technologies, said that Life Tech's capillary electrophoresis business will likely continue to be flat. "But on a fairly large franchise, we think that's pretty good, given that more and more customers in the diagnostics side see that as the gold standard for confirming next-generation results," he said. "So that continued resiliency of that business we see staying in place through 2013."
Despite its lack of glamour, Sanger remains a workhorse in research and clinical labs. There, it is often being used in small studies or, as Lucier noted, to confirm findings. While its use may still decline as next-gen machines become cheaper and better equipped to deal with small numbers of samples, there may be a niche carved out for Sanger sequencing for some time to come.
One of Sanger's strengths — though also a weakness — is that it can handle a small number of samples well. Michael Metzker, an associate professor at Baylor College of Medicine, noted that he turns to Sanger sequencing for small, pilot studies. "I think Sanger has a role in many small projects that don't require whole genomes or whole exomes or things like that," he said. For example, he and some colleagues have been exploring candidate genes associated with ketosis-prone diabetes using Sanger, which they will then follow up on using next-gen sequencing approaches.
Additionally, Michigan's Lyons said that, for some sample sets, Sanger is a more sensible approach. "If it is someone who is doing something that's small scale where the cost of a next-gen experiment would, in fact, be greater than the cost of Sanger — particularly if the difference is big — I'll frequently encourage them to do Sanger," he said, "even though they might have to put in a little bit of work to get those samples set up, in the long run they'll save enough money that it is worth their while."
Lyons added that projects coming into his core usually are small batches, prepared by postdocs, who want to confirm cloning junctions or mutations they introduced or want to do small-scale SNP typing.
Similarly, Karen Staehling, the director of the Stowers Institute for Medical Research's molecular biology facility, noted that her service also receives a number of projects to confirm mutations or constructs. "If they just want to look at a small part of a genome, it's so much cheaper just to amplify it and sequence it with Sanger rather than make a next-gen library," she said, noting that it does depend on the project.
"[If] you're sequencing one area in a plateful of clones, high-throughput sequencing would just be overkill," she added. Staehling estimated that her facility does about 80,000 Sanger reactions a year.
Sanger sequencing has a similar role in the clinic — for small batches and for confirming next-gen finds. Elaine Lyon, the medical director of ARUP Laboratories' genetics division, noted that the number of Sanger sequencing tests performed at ARUP is still increasing.
Indeed, she added that there are a number of clinical situations in which Sanger sequencing would be the tool of choice. "Sanger sequence is appropriate for a single-gene disorder when there is a clear diagnosis to confirm the diagnosis or identify the familial variant," she said.
As an example, she pointed to Factor VIII testing in hemophilia. While the diagnostic test for hemophilia is the activity level of the clotting factor, labs such as hers like to determine the particular genetic cause using Sanger sequencing. Once the familial variant is identified, then others in the family can be tested, again using a Sanger approach, for the disease or for their carrier status. "We don't need to sequence the whole gene; we simply sequence that particular region containing the known mutation," Lyon said.
That sort of role, testing a known mutation in family members, will likely continue for Sanger, she added, even as next-gen approaches make their way into the clinical lab. "Once a mutation is identified [using whole-exome or -genome sequencing], we don't need to do exome or genome sequencing for the rest of the family members," she said. "We simply Sanger sequence that one region where the mutation exists."
Additionally, she said, Sanger will have a role double-checking findings coming off of next-gen machines.
While Sanger is mostly used for single-gene tests, Lyon noted that it can also be used as a small gene panel test — her lab offers a Lynch syndrome test of four genes and a pancreatitis panel that includes three genes. Of course, she added, there is a limit to what Sanger can do. "Testing a single gene or a small gene panel is reasonable by Sanger, but beyond that the workload becomes too great. It then becomes more efficient to use a next-generation sequencing platform," she said.
While Sanger's share of the sequencing pie may be flat or even declining, it may be a while before next-gen sequencing fully eliminates it from the lab bench. To make it sensible to use more routinely in the clinical lab, Lyon said that the cost of next-gen sequencing would have to decline even further. While its cost has been dropping, Lyon said that she isn't sure whether the price will continue to fall or plateau out.
Similarly, Michigan's Lyons said that next-gen sequencers would also have to become better equipped to deal with small numbers of samples to replace Sanger. "What would displace Sanger is if another sequencer came along that was so cheap to use and was scalable, capable of operating on a single sample or relatively few samples, and could give the same quality of data, confidence of data, that are familiar to Sanger users," he said. "At that point, when that instrument is invented, it could supplant the classic ABI capillary electrophoresis instruments, but I don't know of such an instrument now."
There are some companies taking aim at the CE market with next-gen sequencing systems. GnuBio, for example, is developing a $50,000 microfluidic sequencer that will run panels of between 50 and 200 genes. The company has said that it plans to compete with Sanger sequencing for single-gene tests, which it expects to be able to run on its system for about 50 cents per sample.
For the moment, however, it seems that Sanger has a niche carved out for itself. "I'd say for the next few years, Sanger sequencing will be around," Stowers' Staehling added. "We have no plans to get rid of it. It serves a different area than next-gen does."