Ion Torrent ups the ante against MiSeq -- Omics Omics Blog

[ECO]

Keith (krobison) wrote a great article about the battle for the benchtop...check it out here:

http://omicsomics.blogspot.com/2011/...ized-lead.html

[ECO]

From this (PDF warning) press release:

Life Technologies to Launch Ion 318 Chip in Second Half of 2011 with 1Gb of Data Output PGM to Scale 100-Fold in Sequence Output in Less Than a Year CARLSBAD, Calif. – Feb. 23, 2011 – Life Technologies Corporation (NASDAQ: LIFE) today announced that the Ion 318 semiconductor sequencing chip will be available for early access in September of this year, complete with RNA-Seq kits and analysis software, providing up to 1Gb of data output – 100 times more than original Ion 314 chip introduced two months ago with the Ion Personal Genome Machine (PGM™) sequencer.

“Ion Torrent has increased the number of accessible sensors on the Ion 318 chip to 11 million, from 1.2 million on the Ion 314 chip and 6.1 million on the Ion 316 chip,” said Gregg Fergus, President of Ion Torrent. “The Ion 318 is ideal for applications like transcriptome sequencing, miRNA sequencing and ChIP-Seq. The Ion 318 requires no changes or upgrades to the Ion PGM sequencer – with Ion Torrent "The Chip is the Machine,‟ so sequencing is easier and more economical than ever before.”

Scalability
Internally, Ion Torrent has already achieved read lengths in excess of 300bp and improvements in chemistry and loading are expected to increase chip utilization by fivefold. By combining additional reads with longer read lengths, the company will achieve exceptional scalability on its Ion PGM sequencing platform and expects to reach read lengths of up to 400bp in 2012.

Simplicity
As with the Ion 314 and Ion 316 chips, the Ion 318 chip will use simple PostLight™ sequencing chemistry, which requires no modified nucleotides, fluorescence, chemiluminesence or enzymatic cascades. As a result, Life Technologies will continue to deliver low-cost reagents and low-cost detection. The Ion 318 will also leverage four decades of exponential improvement in semiconductor technology, also known as Moore‟s Law. This enables the Ion 318 to be priced the same as the Ion 316, even while delivering a 50-fold reduction in sequencing cost – less than a year after the Ion 314 launch.

Speed
Because of its simple and flexible workflow, the Ion PGM enables parallel processing of 8 samples per operator, which allows researchers to go from sample to result in less than one day, making it the fastest long read sequencers on the market. Ion 318 chips will continue to offer incredible sequencing speeds, while delivering up to 1Gb of highly accurate sequence.

[genlyai]

I tend to be more skeptical than KR on this one.

First, this is all vaporware until they a) have 318 chips in the field, which is half a year away; b) improve their sample prep time as they claim they will; and c) we get some idea of the real-world performance of their current system.

Second, I am confused by their claim above that the 318 and 316 will be priced the same, as they previously indicated that there would be tiered pricing (I think $500, $900, and $1200 for the three chips).

None of this is to say that they won't deliver (and of course I hope they do), just that this announcement comes across as marketing-driven more than anything else.

--

The other thing I'd point out is that the "ideal" PGM scenario from Keith's post -- shallow sequencing from a number of samples on a semi-continuous basis -- is not really in line with the strengths of these devices. For that application, as long as you have a decent lead time, wouldn't you just multiplex onto a larger machine for much less $ and labor?

To me, these machines need to be fire and forget. Have a sample and a couple hours to spare? Know all about it tomorrow. This is very empowering, but is a different set of applications from the big boys.

[krobison]

Thanks for the feedback (and I'll respond in the comments to some more tonight)!! I think it is fair to say I wrote yesterday in line with IT's confidence they can bring these improvements to market.

On the pricing ($500 for the 318, rather than a premium over the current price of $500 for the 314), this does seem to be a shift from the prior statements-- perhaps driven by the pressure from MiSeq. Of course, as is pointed out, until they are in the field it is all hypothetical.

WRT sequencing scenarios, it's really going to depend on how time-sensitive your information is. I'd agree that for many projects, time isn't so critical and so you'd rather wait & pool. On the other hand, for a number of applications (biosurveillance, use in the clinic, rapid method prototyping & optimization, strain optimization) time will be critical and samples will come in at unpredictable rates. And, I see a number of companies which have been quite successful running high-volume on-demand Sanger sequencing. What I imagine (and would love to have access to), is a similar service leveraging either PGM or MiSeq. For a number of recent projects, submitting a pool of sequences for one of these platforms would have been just as easy as the collection of wells for Sanger.

Still, a system more suited for mass crunching (such as PacBio or OxNano) with inherent multiple sample handling would clearly be preferable.

[The_Dark_Base]

When "IT's confidence they can bring these improvements to market" depends on their receiving $350M, everything they say without data to back it up is suspect. This is the biggest conflict of interest I have seen in the genomics field.

The media should be reporting on published data, not treating their crafted press releases as gospel truth. Whether they get the $350M or not, I bet the reports come back to reality when real versus press release data is shown. What ever happened to peer review?


"Over the past few years, Life Technologies (LIFE - Analyst Report) has been expanding its product portfolio through acquisitions; the latest being privately held Ion Torrent, a DNA sequencing company, for $375 million in cash and stock. In addition, Life is also liable to pay $350 million if certain milestones are met through 2012."

i can't think of any tech company that releases a product first, then advertises current specs. I see no reason this would change for a sequencing company. Of course, you have to judge a company by how often they really meet these projections. That takes effort.