Monday, September 17, 2012

BGI Gobbles Complete Genomics

An email this morning alerted me that BGI Shenzhen is acquiring Complete Genomics.  I hadn't been following Complete's business very carefully and had missed (or forgotten about) their quarterly report in August warning that they were dangerously low on cash and had engaged a firm to look at strategic alternatives.  If you are holding Complete Shares (with the wonderful ticker symbol GNOM), you'll get $3.15 cash for each for them.  If you were unfortunate enough to buy them at the top, that's over $11 a share in capital loss to put on next year's tax returns.  I'm not one to watch stock prices closely.  Buy-and-hold an index fund is my primary investment strategy; A Random Walk on Wall Street is the best route to sound sleeping and good returns.  I've never held any GNOM stock.

The big question is what is BGI Shenzhen getting out of all this and how does it make sense?  They're plunking down $117.6 million to acquire Complete Genomics.  In the press release, CEO Jun Wang states "Complete has developed a proprietary whole human genome sequencing technology that, together with other sequencing platforms used by BGI, will fit well with our research and business requirements and position Complete to become an even more successful global innovator".  BGI has been omnivorous in terms of sequencing platforms, so in that sense it fits.  Complete has claimed very low cost-per-base, but has been unable to dominate the human genome sequencing world as they would have liked; Illumina has hacked prices to similar levels.  I haven't talked to them in a while, being out of human genome business for a year now, but a serious challenge for Complete had been the long lead time for their process, though they had demonstrated sequencing from very tiny input amounts (Complete's early service required tens to hundreds of micrograms of input DNA). 

BGI also gets a U.S. sequencing facility to go beyond their office in Kendall Square; it will be interesting to see if they keep the factory open.  On the one hand, much of their business advantage comes from centralization and (presumably) lower labor costs in Shenzhen than the U.S..  However, for many services speed is becoming an issue, with the few days for a sample to transit the Pacific a serious hindrance.  Furthermore, BGI also gets CG's long-term deals with various customers, which weren't sufficient to maintain a publicly traded company but could prove valuable for a different business structure.

BGI also gets Complete's Long Fragment Read technology.  For me, this is exciting because it opens the possibility of BGI making it available for other genomes.  From the outset, my business brain saw the operating advantages to CG's 'any DNA sequence you want, as long as it's a complete human genome' business model, but my scientific brain was always frustrated by it.  Now, there's a chance that this exciting approach to getting long-range sequence information in general, and specifically haplotypes, may be applied broadly.

A final thought: Complete decided not to drag out the process of struggling.  On the public market for less than 2 years, they saw that their business model just wasn't working and headed for the exits.  This can be contrasted with Helicos, which has had a perils-of-Pauline experience for multiple years now.  As someone who once rode a biotech until it sank beneath the waves, I must express a certain admiration for those who continue to believe in the company and keep it going.  But at some point, perhaps it is more merciful to admit defeat and jump into the arms of any reasonable suitor.  Of course, that requires a suitor that one views as reasonable, and so the open question for Helicos is whether there have been no reasonable suitors or Helicos has just been too stubborn to entertain reasonable offers.


Gerry Higgins said...

There is certainly enough evidence now that single molecule sequencing will take that 'great leap forward' past even CG's long fragment read technology, so I wonder why BGI was interested in the acquisition of GNOM?

Anonymous said...

If you are talking about LifeTechnologies licensed platform 'Ion Torrent', or Pac Bio's 'RS' platform, then neither is ready to go commercial or even reserach scale at this time. Pac Bio has been working on thier's for 10 years, and still can't get over the initial problems, meanwhile 44 machines are on order, and few ship every year. Ion Torrent has the same issues, recording single molecule events with electronics sounds great in principle, but fails in execution. Biology is far more complicated than the movement of electrons we try to associate it with in a electronic detection environment

Anonymous said...

Ion Torrent isn't a "licensed platform"- Ion Torrent is wholly owned by Life Technologies. It is not a single molecule platform either, rather, it used monoclonal beads.
PacBio has already shipped 40+ instruments, they are no longer on order, and is on the path to reliability.
And I have no idea about your distinction between "commecial or even research scale"- this means nothing.

Anonymous said...

Well, in some respects, Ion Torrent (or at least semiconductor sequencing) is a licensed technology. The underlying technology is licensed non-exclusively from DNA Electronics.

Dean McNeal said...

This website is a terrific resource to come for important info! Will you be mind if I reblog several of your blog posts on my portal?

Keith Robison said...

Dan: my preference is that content stays here. I'm certainly happy if you comment on this elsewhere and link back