23 And Me...And Google...And Your Genome
February 3, 2008 12:51 PM   Subscribe

Want your genome on a hard drive but don't have the money? 23andMe can give you almost that: a scan of your SNPs, presented online and complete with analyses derived from up-to-date medical research (and a few educated guesses). Eight months ago, blogs were rife with speculation of who 23 could be and what the connection with Google could mean.

But only more recently did 23 launch, and were present at the World Economic Forum last month, obtaining the raw resources (i.e., spit) to develop the genome scan of hundreds of the world's most powerful people, something the Center for Genetics and Society refers to as the beginning of a "fashionable fetish" for the elite; Wired refers to the general trend as retail genomics. More seriously, the CGS also questions the ethics and goals of the scheme, positing that only big business will ultimately benefit (largely from patents and pharmaceuticals), while consumers may have something to lose—their genetic privacy.

And to end on a note ripe for humor, 23 plans to release social networking features later this year. To learn more about the actual 23andMe procedure and web interface, the Wired article (linked above) and the NYT are on the case.
posted by artifarce (33 comments total) 8 users marked this as a favorite

 
I fear that in 2020 I'll be having to decide whether to forego a mortgage or bite the bullet and submit my DNA for authentication..... which of course gets genome-typed by Experian and affects my FICO score and health insurance rates.
posted by crapmatic at 1:13 PM on February 3, 2008


@Crap: That would be scary. Not just for health insurance. Imagine if debt mongers (i.e. credit cards) have a good idea of when you'd die, and thus set an interest rate ensuring they'd get the most money possible from you? We need to get some laws passed to stop discrimination here. But then again, it gets muddy with DNA, because as we get to know it better, we might find the genes we best associate with talent/work ethic/merit/genius/etc. Banning genetic discrimination in those regards would turn our capitalist system on its head. But then allowing it would likely lead to a situation like in Gattaca, where gene tests determine if someone is too mediocre for certain jobs. Clearly, we need someone smarter than me working on the laws of the future. Thanks to the 96% incumbency rate, those people will probably be today's senators. It's gonna be quite a ride.

This all assumes genetics trump environment in most cases. Here's hoping, for the human race, environment keeps things murky enough that everyone gets a chance to prove themselves, regardless of their past.
posted by mccarty.tim at 1:21 PM on February 3, 2008


crapmatic...I fear it will be much, much sooner that 2020. Health insurers are already looking closely at genotyping applicants.
posted by Thorzdad at 1:21 PM on February 3, 2008


Crap, I meant to say "genes," not "past." If there's an edit function in MeFi, sorry I didn't use it. I'm kinda a noob.

So, would they look over the genes again as the knowledge of the genome expands? $1000 seems like a lot for me to spend, but if more of the genetics associated with autism are found, and could be found from the sample I sent, that would be great for me in deciding if I should have kids. Autism seems to run in my family (I'm not autistic, but my grandmother, great grandmother, and sister are, and I think some of my relatives show a few traits). My mom thinks environmental factors, like pollution and vaccines, are to blame, so a clean environment may keep a child of my own from becoming autistic, if that is the case. However, if the autism in my family is innate, adoption would probably be a better idea.
posted by mccarty.tim at 1:25 PM on February 3, 2008


So, would they look over the genes again as the knowledge of the genome expands?

They intimate something like that: "We'll keep you updated on the latest genetic discoveries, and you can keep checking your account to see what those advances may mean for you."
posted by grouse at 1:29 PM on February 3, 2008


You know, I wasn't exactly blown away when I first saw this site. They only seemed to test for a few traits and diseases. It didn't seem to be worth the money they were charging.

But then, the first flight at Kitty Hawk probably wasn't too impressive-looking either.
posted by jason's_planet at 1:39 PM on February 3, 2008


After these revelations, there's a somewhat sick part of me that really wants to see a genetic-compatibility-based dating service.
posted by roll truck roll at 1:41 PM on February 3, 2008


Finally a way to genotype my earwax type. Now I won't have to taste used Q-tips anymore! Thanks 23andme!
posted by grouse at 1:45 PM on February 3, 2008


Gross gross icky Gattica insurance yuck I'm afraid for the future.
posted by localhuman at 2:01 PM on February 3, 2008


Decode Genetics (decodeme) has been doing the personalised SNP business for longer than 23andme, and offers roughly 10x the SNPs of 23andme for the same price. The best thing is that the company safely removed from the privacy-hostile US environment and is based in Iceland, which has one of the stricter data privacy regimes in the world (even when you're dead, which protects your descendent's genetic privacy as well). Lodging your genetic data with any US-based company is basically tantamount to public genetic exhibitionism. Do it, by all means, but don't act surprised when your genome ends up exposed on random Google searches.
posted by meehawl at 2:06 PM on February 3, 2008 [6 favorites]


No privacy, and they'll sell your data to big pharma and big insurance. Stay away.
posted by orthogonality at 2:09 PM on February 3, 2008


Eight months ago, blogs were rife with speculation of who 23 could be and what the connection with Google could mean.

Without reading the links or comments, I'm calling 23 chromosome pairs.
posted by spiderwire at 2:16 PM on February 3, 2008 [1 favorite]


Oh... was the question who the company is? Um, it's a present for Sergey Brin's girlfriend, what more was there to know?
posted by spiderwire at 2:20 PM on February 3, 2008


orthogonality: No privacy, and they'll sell your data to big pharma and big insurance. Stay away.

Ortho, you're so uncreative. Did Gattaca teach you people nothing?

1. Think of the healthiest person you know who looks somewhat like you. Next time you run into them, ruffle their hair playfully when you say hello.
2. Take one of the strands of hair that gets caught in your fingers and bring it to 23andme to get sequenced. Give them allllll your personal information like you have no idea what they're going to do with it.
3. Get cheap-ass insurance, pharmaceuticals, and health care for the rest of your life.
posted by spiderwire at 2:27 PM on February 3, 2008 [3 favorites]


FYI, a lot of information and useful utilities about deCODEme, 23andme, SegWright, Knome, and other full-genome-scan companies are collected on this webpage.

Personally, I use the terrific FamilyTreeDNA for all my genework, since genealogy is my main focus rather than disease-tracking, but I hear rumors that they too are going to start offering similar targeted gene tests in the near future. Lots of money to be made in this arena...
posted by Asparagirl at 2:53 PM on February 3, 2008 [1 favorite]


There already is a genetic "compatibility" dating site based on the MHC stuff that determines how you smell/ your immune system.

It was discussed here before but search isn't finding it for me for some reason... based in boston.
posted by Maias at 2:57 PM on February 3, 2008


also, yeah, i didn't see speculation.. when it was first announced, it was announced as brin's girlfriend's project.
posted by Maias at 2:58 PM on February 3, 2008


While this topic is certainly ripe for some analysis of the ethics and repercussions that mass genome typing might lead to, I was also excited by the detail in the Wired and NYT articles. To me, the the increased rate of which we're learning about the human genome and the accessibility to the layman is akin to seeing computer technology not only get better, but cheaper, and exponentially so. The unique identity that is one's own genetic information and a fractional sense of what it means is the draw of 23andme. Maybe I'll have to keep looking for something that's more private--thank you for the link to decodeme, meehawl.
posted by artifarce at 2:59 PM on February 3, 2008


And forgetting preview, thank you, asparagirl, for more resources.
posted by artifarce at 3:00 PM on February 3, 2008


This may have already been posted, but here is the wired article on 23andMe, which has a vid interview with the founders.
posted by Lutoslawski at 3:22 PM on February 3, 2008


One word: creepy.
posted by -t at 3:24 PM on February 3, 2008


On multiple levels.
posted by -t at 3:24 PM on February 3, 2008


Here's another interesting and skeptical article on 23andMe.

Another Genealogy based DNA service, Sorenson Genomics, always made a point to stay away from recording the DNA for used for disease tracking. It was founded by Utah's medical inventor/billionare James Sorenson, who died just a couple of weeks ago. (H.e invented disposable surgical masks, among many other things)
posted by eye of newt at 4:14 PM on February 3, 2008


Its disheartening to think that you could put out so much money for one of these scans and already be using outdated technology. All the cool kids are doing disease association studies on copy number variation these days.

From the second link:
Most research has focused on small alterations, called single nucleotide polymorphisms (SNPs). It may be, said Scherer, that some diseases are caused by copy number variations rather than SNPs. In fact, recent research has already linked such variations to kidney disease, Parkinson's disease, Alzheimer's disease, and AIDS susceptibility.


Personally, if I'm going to pay my hard earned cash for someone to analyze my genome, I want to get something a bit more lasting out of the deal.
posted by Hutch at 4:57 PM on February 3, 2008


All the cool kids are doing disease association studies on copy number variation these days.

Of course, the coolest kids would know that you can infer the presence of a copy number polymorphism using SNPs. They'd also know that CNV studies are not a replacement for SNP studies.
posted by grouse at 5:25 PM on February 3, 2008


A prediction from a friend of mine who's knowledgeable about such things: 23 (and by association, Google) are really aiming at compiling the biggest human genomic dataset anywhere, and then providing that raw data to places like Celera Genomics and the HGP. While I'm not nuts about monetizing people's genetic code, it is a fascinating notion.
posted by Emperor SnooKloze at 6:26 PM on February 3, 2008


Of course, the coolest kids would know that you can infer the presence of a copy number polymorphism using SNPs.

The efficacy of the COPPER algorithm has yet to be widely assessed (although the results from the NEJM autism study are pretty solid). Hence the CNV probes on the latest iteration of the Affymetrix arrays. One potential drawback of exclusively relying on SNP data to infer CNVs might be an uneven distribution (when examining small windows) of arrayed SNPs across the genome. As the algorithm (as I understand it) relies on finding correlated changes in SNP intensities, you might expect to overlook CNVs in areas with lower amounts of coverage.
posted by Hutch at 7:18 PM on February 3, 2008


A thousand dollars?

Can't I just click on some ads or watch some videos or something?
posted by Ynoxas at 7:56 PM on February 3, 2008


As someone who studies genomics, I'll tell you that unless your family has a history of some rare genetic disorder, there's not much point right now. Give the biology a few years to catch up to the technology. This stuff will only get cheaper and more powerful.
posted by chrisamiller at 8:31 PM on February 3, 2008


A thousand dollars?

Can't I just click on some ads or watch some videos or something?


A quick Google search reveals that the average cost per click for Google Ads is about $0.50, so you should only have to click on about two thousand. Four thousand and you get a new laptop!

... Holy shit, I just realized how totally broken the Internet economy is.
posted by spiderwire at 8:36 PM on February 3, 2008


@Ynoxas: Why not just sell the right for ad companies to analyze your genome for marketing research? I'm scared to say that once the technology hits the sub-$100 range, that could very well be the case, and I might just fall for it because of how darn fascinating it is.
posted by mccarty.tim at 12:34 AM on February 4, 2008


Startup money from Google? I can see it now...

Welcome to GEne (beta)!

GEne is different. Here's what you need to know...

1. Don't sequence your genes - search!

2. You don't like junk DNA. Neither do we.

...and so on.
posted by caution live frogs at 11:21 AM on February 4, 2008


A quick Google search reveals that the average cost per click for Google Ads is about $0.50, so you should only have to click on about two thousand. Four thousand and you get a new laptop!

... Holy shit, I just realized how totally broken the Internet economy is.
posted by spiderwire at 10:36 PM on February 3 [+] [!]


I could make a decent living clicking ads at $0.50 a pop. I've seen probably 200 ads today, and that's from about 30 minutes reading CNN through the day and maybe an hour on various sites (non-porn even) tonight.

Let's see... 1000 ads would be 125 per hour for 8 hours, so a little over 2 per minute.

Yeah, I think I could do that pretty easy.

Take a job like overnight security guard or motel front desk, you could earn 2 incomes.

Ok I'm sold. Where do I sign up?
posted by Ynoxas at 9:26 PM on February 4, 2008


« Older Ayahuasca: A Strange Brew....  |  Photo albums of German soldier... Newer »


This thread has been archived and is closed to new comments