Skip to main content
Premium Trial:

Request an Annual Quote

Life Tech Plans New Ion Torrent Instrument to Deliver Whole Genomes in Hours for $1K

Premium

By Monica Heger

This story has been updated to include analyst opinions.

Life Technologies' Ion Torrent business is planning to launch a new sequencing instrument by mid-year that will enable a whole human genome to be sequenced in hours at a run cost of $1,000.

The company has also begun shipping the 318 chip for the Ion Torrent PGM. The new chip has a throughput of 1 gigabase per run, around 10-fold higher throughput than the current 316 chip.

The new machine, the Ion Proton, will be based on the same semiconductor sequencing technology as the PGM. The instrument will be around the same size as the original system, but will include upgraded electronics that have the ability to handle genome-scale data.

The company is planning to release two chips for the Proton this year: the Proton 1, which will be able to sequence a few exomes per run or one five-fold coverage genome; and the Proton 2, which will be able to sequence one whole genome at 20-fold coverage or eight exomes per run.

proton1.jpgRead lengths will be 200 base pairs, and the accuracy profile will be essentially the same as for the PGM, the company said. The machine will cost around $149,000, and will also require the purchase of a $75,000 server for analysis.

The company is taking orders as of today and plans to begin shipping the machine with the Proton 1 chip to early-access customers by mid-2012. Unrestricted commercial release of the machine is scheduled for the end of the third quarter, with the Proton 2 chip slated for release by the end of the year.

According to Andy Felton, senior director of product marketing at Ion Torrent, the main difference between the PGM and the Proton is that the Proton uses updated complementary metal oxide semiconductor technology, which enables a much higher sensor density on the chips.

proton2.jpgThe Proton 1 chip will have 165 million sensors and the Proton 2 chip will have 660 million sensors. By comparison, the initial 314 chip had 1 million sensors and the 318 chip has 11 million sensors.

The 314, 316, and 318 chips were all "of an era of a computing platform from 1995," while the Proton chips represent updated semiconductor technology, Felton told In Sequence. "The transistors have gotten better and faster," said Felton. "We can do the sequencing in the same time for the 165-million-well chips as the 1-million-well chips."

But in order to handle this increase in output, the electronics inside the machine also had to be updated, hence the need for a new system. "The electronics on the PGM were built to handle chips of 1 million to 10 million sensors," Felton said. "To go from that to a system that can handle nearly a billion sensors, you have to have electronics that are built to handle that kind of output."

Sample prep for the Proton will be the same as for the PGM, and will use both the Ion Express Library Kit and the Ion One Touch for amplification. As with the PGM, library construction on the Proton will take two hours.

Ion Torrent is also planning to release an exome capture kit in conjunction with the Proton. The kit will be compatible with the PGM 318 chip as well. The exome capture process will take a couple of days, similar to other exome capture kits.

Users of the Proton will also have to purchase a $75,000 standalone server. However, the machine does not have the same computational requirements for downstream analysis as systems such as Life Tech's SOLiD and Illumina's HiSeq, said Felton.

With those optical-based systems, there is a "relatively small computing platform that runs the instrument and does base calling," he explained. After a sequencing run, the user is left with a "mountain of data," so mapping and alignment becomes a "massive undertaking that requires a big server and IT infrastructure," he said.

Analysis on the Proton can also be done in the same day as the sequencing run, said Felton. For a multi-sample sequencing study, users can "handle the data piecemeal, rather than a deluge of data once every two weeks," Felton said.

The Ion Torrent team is also developing a cloud-based analysis system called the Ion Reporter. It will launch this system in the second quarter initially for the PGM and eventually to support exomes and genomes from the Proton.

The Ion Reporter will enable alignment, variant calling, and data storage in the cloud and will have a pay-per-use fee associated with it.

The company has already lined up several early-access users for the Proton, including groups at the Broad Institute, Baylor College of Medicine, and Yale University. Representatives from these organizations were unable to comment on the new system before press time.

Securing these early-access users is a particular boon to the company, according to analyst Isaac Ro from Goldman Sachs. The Broad's Chad Nusbaum, Baylor's Richard Gibbs, and Yale's Richard Lifton are all "leading minds in the field and whose opinions will carry weight among customers," he wrote in a research note.

As Ion launches the Proton, the machine's performance will be important for its adoption, particularly in the clinical space, Ro added. Illumina remains the "gold standard," and it is "not yet clear what types of accuracy (q-scores) can be achieved on the Ion Proton."

Amanda Murphy from William Blair added that the firm is "cautiously optimistic" about the Proton and that the new system will intensify competition in the benchtop sequencer market, which the firm estimates is currently about 6,000 instruments.

While "the battle has been lost" at the high-throughput sequencing level to Illumina, "the battle in the benchtop low-throughput segment is just beginning," she wrote.

Ion Torrent has also tapped a group at Carnegie Mellon University to develop a so-called "doctor in a box" for clinical interpretation of exome and whole-genome data. The software tool will be compatible with data from both the PGM and the Proton, and will use machine-learning algorithms to provide predictions on diagnosis, disease risk, and pharmacogenomics.

Felton said that the Proton will be aimed at "the average laboratory, [which to date] was not able to participate in genome-scale biology because of the price it costs to get into the game."

The news of the upcoming instrument boosted Life Tech's stock by 8.25 percent to close at $46.17 per share.


Have topics you'd like to see covered by In Sequence? Contact the editor at mheger [at] genomeweb [.] com.