Speed-Reading DNA Inches Closer

Continued from page 1

Oxford Nanopore can identify bases, but not yet in sequence. The system that it has demonstrated involves passing chopped-up DNA, not whole strands, through the nanopore. The company is now working on a setup for feeding long strands of DNA through the pore one base at a time. To do this, the researchers must attach an enzyme called an exonuclease to the nanopore. They hope that bases will be chopped off one at a time by the enzyme and will pass through the pore to the other side.

"There's some question [about] what will happen when you put long strands of DNA in front of the nanopore," says Schloss. "Will it form a hopeless knot?"

This is just one of several unknowns confronting researchers. To make the technology truly scalable and commercially viable, the pores will need to be grouped in large arrays, and the company will need to develop a less complicated way of reading the electrical signals from the pores. Oxford Nanopore says that it is currently working on both of these problems.

A potential pitfall of Oxford's nanopore-exonuclease approach, says Schloss, is that the DNA will be destroyed as it has been read, making it impossible to resequence a strand to check for errors.

However, there are other approaches to nanopore sequencing that are less destructive. David Deamer, emeritus professor of chemistry at the University of California, Santa Cruz, who first came up with the concept of nanopore sequencing in the 1990s and is a scientific advisor for Oxford Nanopore, points out that this is not the first demonstration of a nanopore system that can identify all DNA bases. Last year, researchers led by Reza Ghadiri at the Scripps Institute, in La Jolla, CA, sequenced a 10-base-long strand of DNA using another nanopore technique. The flow of DNA through the Scripps system, which is based on Deamer's original concept, is controlled by an enzyme that acts as a ratchet, moving the molecule forward one base at a time. But this system is much too slow, advancing at a rate of one base every 10 minutes, and the Scripps researchers are working on speeding it up.

Oxford Nanopore hasn't put all its eggs in one basket. It has licensed technology for several nanopore-sequencing methods, including Deamer's and another that uses an artificial nanopore: a silicon wafer punched with nanoscale holes and lined with carbon nanotubes the conductance of which changes as the DNA passes.

"One of these approaches will have a breakthrough and will be able to sequence at a rate faster and cheaper than what we do now," predicts Deamer. Neither the researchers at Oxford Nanopore nor those at competing labs are willing to speculate about just when this will happen, or what such a system would cost per genome. But Schloss says it's possible that one of the groups will meet the National Human Genome Research Institute's original target year of 2014 for successful nanopore sequencing.