A commenter on yesterday's piece on Intelligent Biosystems scolded me on not mentioning GnuBio and said they had released data. This had totally slipped my my notice, and indeed it seemed to have slipped past the GenomeWeb and BioIT worlds as well, judging from some Google searches. There is certainly a press release out there and covered by several outlets, but amazingly stealthy for public relations. Strange! But many thanks to my anonymous correspondent for flagging me on this!
But, it does make a set of stunning claims. When GnuBio launched last year and announced they would have alpha instruments in collaborator's hands by the end of 2010, I was skeptical.
Apparently, they really have gotten machines out (perhaps not quite on that schedule, but quite close). While they haven't released data generally or published it, they claim on a PCR amplicon (chosen to be 58% G+C and bearing homopolymers up to 6 bases) to have acheived 99.993% raw accuracy (phred 41.5!), 100% sequence coverage, 100% useable reads, 1700X coverage of the amplicon and read lengths of 126. More stunningly, they claim "ead lengths of 1Kb can routinely be achieved today". The system is proposed to launch at $50K. One other neat angle, which I covered a bit in the earlier post, is the fact that your sequencing cost is proportional to how much sequencing you want to do rather than a fixed size determined by a flowcell. Fully dial-able sequencing capacity!
If these really hold up more generally, then this would blow away any existing platform. Cost of an Ion Torrent, read length of a PacBio, accuracy of a SOLiD and throughput of a HiSeq, all in one box. Apparently, faster and cheaper to run than any of them. At least that's the promise.
Can they deliver? As always, good question. Demonstrating this much progress this quickly is impressive, but nothing is available publicly. Both of the academic scientists (Justin Lamb of the Broad & Michael Phillips Montreal Heart Institute) of describing the results are on the Scientific Advisory board for GnuBio. I'm not one to immediately discount their opinions due to their closeness, but they can't be seen as fully independent. Other than general descriptions of microfluidics, their website is devoid of descriptions of the chemistries used, overall operational scheme or anything else useful in technical evaluations. Perhaps this system won't have an Achilles heel, but it will be hard to find it in a single PCR fragment, and one which represents a little difficulty but not true sequencing hell.
I fervently hope they are on the right track and can start pushing out some peer reviewed publications so it is possible to really kick the tires., Ideally those will be a bit higher profile and garner more notice than this week's press release.