A curious tweet this morning from personal genomics company deCODEme, barely a few weeks after the declaration of formal bankruptcy of parent company deCODE Genetics:
@decodegenetics: Migrate to deCODE this winter! Upload your genetic data for free. http://www.decodeme.com/data-upload
Here's a description of the service from the URL in the tweet:
deCODEme wants to give even more people the chance to enjoy the best in personal genomics. Our bioinformatics team has just launched a simple system to enable existing customers of 23andMe™ to migrate their data into deCODEme and to join our growing community. If you already have a 23andMe genetic scan, just click on the button below to begin the upload process and start to view your genome using deCODEme's many advanced features.
This service is available to existing 23andMe customers and for a limited time only.
Enjoy and spread the word!
Basically, the company seems to be providing its interpretation service free to customers who have already had their genome scanned by competitor 23andMe.
That's a bold and initially rather puzzling move. Those of you who've been following the personal genomics industry will know that the value of genome scans is not in the actual generation of the data (this is a straightforward procedure), but in the breadth and quality of the interpretation service.
Converting a series of half a million or so genetic data points into predictions of ancestry and disease risk is a non-trivial exercise, and requires the creation and constant, painstaking maintenance of a database of genetic associations. Parsing the literature to extract the required information can be a frustrating exercise, made even more difficult by the sheer rate that new associations are being generated.
Why, then, would deCODEme choose to give away its hard-won interpretation to customers of its most successful competitor?
Well, deCODEme is in a hard place right now. Since the launch of its personal genomics service two years ago, the company has worked hard to differentiate itself from its competitors using the scientific credentials of parent company deCODE, which has been spectacularly successful in converting its exclusive access to the genetic, medical and genealogical records of the Icelandic population into a string of high-impact publications (but, sadly for its investors, it has simultaneously been disastrously unsuccessful at converting those publications into cash flow).
Following deCODE's bankruptcy, deCODEme and other components of the parent company's scientific apparatus are in the process of being sold off to US-based investors. There's now, more than ever, massive pressure for commercial success from deCODEme's personal genomics offering; and that means finding some way of going head-to-head with the well-funded and Google-backed 23andMe.
Sales records are impossible to come by, but everything I've seen suggests that deCODEme's strategy of appealing to its scientific credentials has completely failed to drive personal genomics customers away from 23andMe's slick PR machine and appealing interface towards the more sober deCODEme product line.
So, why the free offer? I'm guessing deCODEme is gambling (quite reasonably) that offering free uploads will attract a non-trivial number of 23andMe customers over to deCODEme's interface. That then provides the Icelanders with an opportunity to give people a fair trial of their own interface, and hopefully to impress them with the quality and accessibility of the data provided.
What then? Well, bear in mind that the entire chip-based personal genomics industry is really just a transient place-holder for the real deal: interpretation of complete genome sequences. All of the personal genomics companies currently out there are simply positioning themselves for a share of the potentially enormous sequencing interpretation market that will emerge within the next couple of years as the cost of DNA sequencing plummets.
It's therefore crucial for deCODEme to place itself in the minds of likely early adopters of sequencing products as a serious and reliable player in the interpretation field. Right now it's failing to do that, due to the extraordinary market dominance of 23andMe - but pulling over customers with this free offer will help.
Anyway, I'll be interested to see what level of functionality is offered to 23andMe uploaders. I've already uploaded my own 23andMe data to the deCODEme website, and I'm told my results should be ready in around 24 hours; I'll hopefully have a chance to post my thoughts on the interface early next week.
Life Science
Medicine & Health
Comments
Do you have a blog post on your personal experience with your SNPchip data from 23andme? Looked but couldn't find one. Curious to know what you think of it. Thanks.
Posted by: squid | December 17, 2009 10:06 AM
I'm eagerly awaiting the next generation of DTC personal genomics.
Posted by: Russell | December 17, 2009 10:54 AM
Hi squid,
I haven't written about it yet. I'm saving my detailed analysis of the service for a separate project, to be described in the near future. Watch this space!
Posted by: Daniel MacArthur | December 17, 2009 11:46 AM
A similar message was posted on the GeneTrends blog
Posted by: Jordanna Joaquina | December 17, 2009 2:40 PM
Not too familiar with the 23andme site and would appreciate advice on how to download the necessary genetic text file to take advantage of the deCODEme offer?
Posted by: John Sullivan | December 17, 2009 7:53 PM
Hi John,
Click on Account > Browse raw data in the top right of the screen after logging into 23andMe. Then click on the "download raw data" link.
Posted by: Daniel MacArthur | December 17, 2009 11:29 PM
One of the main assets of deCODEme is their database. By accepting 23andME results, they increase the value of their database for next-to-nothing cost. It's great for them and for those 23andME customers who share their data with them. I suppose 23andME might not be too happy though.
Posted by: Richard R. Kenyon | December 18, 2009 4:03 AM
I don't think the raw data per se is particularly valuable, but I noticed that at least some reports ask you to fill out a brief (and optional) questionnaire about your medical and family history background before the results are revealed. This would circumvent some of the pitfalls you described in a previous post
http://scienceblogs.com/geneticfuture/2009/10/23andme_presents_novel_genetic.php
Posted by: Ann Turner | December 18, 2009 5:37 AM
Daniel,
I agree with the prior comments. This is a great way to poach data. Raw data is not that useful, but it gives the warning shot of, we are in this long term and we will steal your customers.......It could technically devastate any perceived value of these companies' "proprietary" database.....which if I remember correctly is what kicked Celera in the nuts....
Merry Christmas!
-Steve
www.thegenesherpa.blogspot.com
Posted by: Steven Murphy MD | December 18, 2009 8:44 PM
Maybe deCODEme is transforming their business model from test provider to data interpreter:
http://www.accessdna.com/blog/2009/12/is-decodeme-trying-to-migrate-their-business-model-along-with-23andme-customer-data/
Jordanna
www.AccessDNA.com
Posted by: Jordanna Joaquina | December 21, 2009 9:46 AM