Following early access collaborations with academic research groups, Nabsys plans to start a beta-testing program for its single-molecule electronic DNA mapping platform by mid-year and to take orders for commercial systems in the second half of the year.
Two weeks ago at the Advances in Genome Biology and Technology conference in Marco Island, Fla., the company showed mapping data for viral, bacterial, and small eukaryotic genomes and demonstrated how its technology can help to correct de novo assemblies.
First applications for the commercial instrument, which will have eight independent detection modules and a not-yet-finalized list price of $149,000, will include "push-button" de novo assemblies of microbial genomes from a Nabsys map and short read sequence data, structural variation analysis, and quantitation of microbial strains in metagenomic samples.
Nabsys presented its technology and platform in detail about a year ago, but did not present whole-genome mapping data at the time.
Last year, the company was planning to provide both long-range maps of genomic variation and base-by-base sequence data. That has changed as sequencing technologies have continued to improve. "We are now focusing really on being complementary to sequencing," NabSys CEO and President Barrett Bready told In Sequence.
The firm will concentrate on mapping variations greater than a few hundred bases in size – including insertions, deletions, translocations, and inversions, some of which are difficult to infer from sequence data alone.
In brief, Nabsys hybridizes short sequence-specific probes to DNA fragments up to 250 kilobases in length and runs the DNA through a solid-state semiconductor nanodetector at a speed of more than a million bases per second. Electronic signals from the probes passing by the detector are recorded with a resolution of several hundred bases and are translated into positional maps. The company has 10 different probes available, but most applications work well with just one, Bready said. The probes recognize four to 22 bases of sequence, though the firm tends to use 5-mers to 7-mers. Several publications detailing the basic technology are currently under review.
While a year ago, Nabsys was mapping short synthetic DNA stretches, it now generates maps for entire microbial and small eukaryotic genomes and has a number of collaborators, including several major genome centers.
In one experiment, conducted with researchers at the Marine Biological Laboratory in Woods Hole and at the Massachusetts Institute of Technology, Nabsys researchers generated a map for a 75-kilobase bacteriophage genome that had previously been assembled from high-coverage Illumina sequence data and found that the original assembly was incorrect.
Nabsys researchers also ran a 250-kilobase previously sequenced BAC clone on its platform and found it to be incorrectly assembled. Some of the BACs went through the nanodetector as single intact molecules, Bready noted.
"What we have learned is, except for genomes that have been meticulously finished, very few things are actually correctly assembled," he said.
One of the first applications of Nabsys' technology will be hybrid de novo assemblies of microbial genomes from its maps and short read sequence data.
Nabsys' platform is also suitable for distinguishing closely related bacterial strains. The company ran five different E. coli strains on its platform, for example, and found "some really obvious differences" at the structural level, Bready said. The researchers were also able to map an unknown bacterial strain and see which other strain it was mostly closely related to. This could be used to quantify different microbial strains in environmental samples.
Longer term, the platform might have clinical applications, for example for mapping structural variations in human genomes with better resolution than array CGH, including inversions, which arrays cannot detect.
Future field applications, such as disease outbreak monitoring and safeguarding food and water supplies, are also possible, Bready said, since the technology is "stable in a variety of environments," requires no library preparation, and can run with a watch battery as its power supply.
Nabsys is currently looking for additional early access collaborators to test its technology on genomes up to about 100 megabases in size. "We're not encouraging use for human genomes right now," Bready said – mainly because data collection would take more than a day rather than just an hour – "but in spite of our best intentions, people are going to do that."
The first commercial instrument, for which Nabsys will start taking orders in the second half of this year following a beta-testing program mid-year, will have eight independent modules, each of which will take one chip with a single nanodetector, which is reusable. The instrument list price, which has not been finalized yet, is $149,000, and each experiment will cost about $300 in chips and consumables.
The company plans to increase the number of detectors per chip by about an order of magnitude each year or so, Bready said, and is already working on the next generation of chips. There might also be instrument versions with fewer modules in the future.
Other technologies – such as Pacific Biosciences' single-molecule real-time sequencer, BioNano Genomics' nanochannel array optical mapping platform, and, very soon, Oxford Nanopore's nanopore sequencer – also focus on providing long-range structural information and helping with de novo assemblies, but Bready said Nabsys' approach offers a number of advantages.
For example, Nabsys' platform analyzes longer DNA molecules – up to 250 kilobases – than any long-read sequencing platform to date. PacBio's current chemistry, for example, generates average read lengths of 8.5 kilobases, and a researcher at the AGBT meeting showed a 54-kilobase PacBio read. Early data from Oxford Nano's MinIon platform had a mean read length of 5 kilobases, with some reads more than twice that size.
Also, though it does not offer single-base resolution, Nabsys can analyze DNA at a much higher speed than other technologies, Bready said. And because its signal is electronic, it can obtain better resolution than optical mapping technologies, he added.