54 replies to this topic

[manofsan]

The age of personal genomics is now beginning, with the launch of a gene-scanning service by 23andme.com, a company started by the wife of Google founder Sergey Brin.

http://news.google.c...123702198&hl=en

For $1000, they will broadly scan your genome, and allow you to see the results via the web. A rival company in Iceland called DeCode is now also offering a similar service for $15 less.

It's no great bet to recognize this industry will take off dramatically in the years to come. Does anyone remember the movie Gattaca?

[apoptosos]

Also Navigenics

http://www.navigenics.com/

[niner]

The age of personal genomics is now beginning, with the launch of a gene-scanning service by 23andme.com, a company started by the wife of Google founder Sergey Brin.

For what it's worth, Decode must know a hell of a lot more about this stuff than the Google Wives Club. (Some other Google Wife is also involved...)

[Aegist]

Cool. Can't wait

[Johan]

Nice
I wonder if we'll see any insurance companies using this... could lead to insurance problems for individuals with high-risk genotypes.

[apoptosos]

For what it's worth, Decode must know a hell of a lot more about this stuff than the Google Wives Club.  (Some other Google Wife is also involved...)

The "Google wife" also happens to be the owner of DNA Direct, arguably the best direct-to-consumer genotyping provider.

[manofsan]

I think right now there's a bill going before the US Congress on genetic discrimination.

Anyway, Decode.com is part of a much larger bio-pharma group, from my understanding. They're offering a million data points, as compared to <600000 offered by 23andme.com

There's also knome.com , which is headed by Harvard's George Church. He's working on the nanopore technology, and knome.com plans to offer full genome sequencing.

I hope they all get flooded with customers, so that a clear signal is sent that personal genomics is the new bonanza industry. For $1000, I think I'm more than willing to pay that to get myself sequenced. Maybe I'll just want to find out about the privacy and legal issues first, including implications on health insurance, etc.

[Athanasios]

http://ichart.financ...oo.com/w?s=DCGN

[John Schloendorn]

Can't wait

Wait? You don't need to wait. What are you waiting for?

[Aegist]

Can't wait till i have $1000 spare?

*and by spare, I mean $1000 I would happily throw away....

[mitkat]

I'd be interested to hear some predictions (sound or otherwise) as to an eventual price drop in services like these, as 1000$ is 1000$ of student loan, or rent.

[niner]

For what it's worth, Decode must know a hell of a lot more about this stuff than the Google Wives Club.  (Some other Google Wife is also involved...)

The "Google wife" also happens to be the owner of DNA Direct, arguably the best direct-to-consumer genotyping provider.

Yeah, but no one has Decode's database or depth of experience.

[manofsan]

I for one would be willing to spend $1000 for this info. It seems to be a more than reasonable price. Although for the sake of the world at large, I hope that the price will one day drop to a few bucks.

But I'd still like to do a little research and talk to others who have already gone through the process, to find out more directly what they felt they got out of it.

A million SNP data points is more than enough to work with -- at least until full-genome sequencing comes out.

I too would be leaning towards Decode.com
I hope they're not all booked up

[manofsan]

Btw, here is the webcast from decode.com's latest announcement:

http://www.decode.com/Webcast/

And here is the video for how to collect your DNA sample using their supplied kit:

http://www.decodeme.com/buccalvideo

Seems pretty straightforward.

[caston]

Any idea if they do international orders?

[manofsan]

Of course they do. I'm in Canada, and I've just signed up.
They take payment by Paypal or credit card.

But they won't deal with a few places, like New York, California, Arizona, etc.
I presume there are some kind of legal hurdles there.

[unisonus]

So, what practical benefit does this procedure afford?

[manofsan]

Well, decode.com will give you 1 million data points on SNPs for $985, while 23andme.com will give you 550,000 + 30,000 custom ones for $999.

Sure, it's not total end-to-end sequencing, but it's pretty good, because it'll tell you what combination you have on the major genotype variants.

knome.com is by Harvard's George Church, and they're offering future end-to-end whole genome sequencing, but they don't seem to be up and running yet. Their website seems to be just a placeholder for now.

Look, as soon as the orders/money start pouring in, then more players are going to jump into the game, and the cost will come down. Hell, it's worth considering buying stock in such companies, if they make public offerings. After all, there are 6 billion people on the planet, and they all have genomes, and they're all very likely to be concerned about their health at some point. Can you think of a larger market than that?

On the other hand, until more players bring the service availability up, the few service providers could be quickly inundated with orders, and it could be harder to get sequenced than finding a Nintendo Wii for xmas.

Here, read some of this stuff:

http://www.nytimes.c...7/us/17dna.html

http://www.nytimes.c...s/17genome.html

[caston]

So, what practical benefit does this procedure afford?

It allows all those single mothers to pin point you as the father of their child and come forward seeking child support payments... ;-p

But seriously from the video it offers 2 benefits :
1) You got to trace back your genetic ancestory
2) You can find out if you are predispositioned to certain genetic diseases.

[niner]

But they won't deal with a few places, like New York, California, Arizona, etc.
I presume there are some kind of legal hurdles there.

I've heard that people in these backward regions drink the blood of children and wear clothes made of live tarantulas. I wouldn't deal with them either.

So manofsan, let us know what happens! This is pretty cool.

[caston]

Also

http://www.familytre...m/products.html

http://www.genelex.com/

[manofsan]

Here, read this:

https://www.23andme....win/sequencing/

[manofsan]

What I'm thinking is that once genome information becomes more commonly available, the next thing you'd want to know is how it relates to your metabolism.

That's where network mathematics comes in.

If you have a million genotype data points, then from that you'd want to construct your network of metabolic pathways -- the "control circuitry" that runs your metabolism.

If you could feed in your genomic data into some software that could construct your metabolome for you, then that would be even more informative and useful for you.
Because based on how your pathways behave and interact with various substrates and each other, then you could more accurately model yourself to extract useful information.

You could see which of your biochemical control loops, feedback loops, etc work well, and which don't work as well, thus warranting greater care/precautions on your part. Hey, isn't that how the big financial trading companies model the "metabolism" of the stock market to optimize their asset growth in the right direction? In a way, that's what you'd want to be doing to optimize your lifespan.

This again then brings to the systems biology stuff that would be a more rigorous and robust way of dissecting your own biology.

[apoptosos]

For what it's worth, Decode must know a hell of a lot more about this stuff than the Google Wives Club.  (Some other Google Wife is also involved...)

The "Google wife" also happens to be the owner of DNA Direct, arguably the best direct-to-consumer genotyping provider.

Yeah, but no one has Decode's database or depth of experience.

Decode's database is based on an Icelandic population so this may work against them.

Not trying to take anything away from Decode - they identified some very important polymorphisms associated with diabetes, heart disease, obesity and schizophrenia - but I just can't see any company being able to compete with the might of Google.

The company that ends up dominating this area will do it based on making the information understandable, relevant and fun to the average person. If its just gloomy disease risk predictions most will avoid it. 23andme are looking to fold the data into a Facebook style social networking paradigm where genenauts will be able to compare genomes and look for genetic compatibility. Sure, Decode is also featuring this on their site but Google has the advantage on the social networking stuff.

[caston]

I know this sounds stupid but even at 26 I'm still confused about my orienation. Is it possible that such sequencing could help me understand myself more?
Perhaps one day I will even know what genes I can insert to get a little less confusion and a little more action

The company that ends up dominating this area will do it based on making the information understandable, relevant and fun to the average person. If its just gloomy disease risk predictions most will avoid it. 23andme are looking to fold the data into a Facebook style social networking paradigm where genenauts will be able to compare genomes and look for genetic compatibility. Sure, Decode is also featuring this on their site but Google has the advantage on the social networking stuff.

Imagine including your genetic profile in an online dating site and searching for breeding matches based on other peoples genetic profiles. That is just mind boggling.

[manofsan]

I agree that Google is ahead on the deep brain trust score, for doing all kinds of fancy things with this info. But on the other hand, a Google rival like Microsoft, Yahoo, etc could easily pair up with Decode to offer various hi-tech web services.

I dunno if I want to be locked in with one particular website, though. It'd be nice to get a copy of your DNA for yourself, and then you can do with it as you like, without being trapped into only viewing it thru the one website.

For that purpose, it would be nice to have some common widely used format for the genomic data. (*.gen ?)
You could then feed your genome file into 3rd party applications designed to analyze whatever aspects you wanted.

Like I said, it would be nice to be able to convert your collection of 1 million genotypes into a metabolic picture/map, because a metabolic picture is more meaningful than some mere collection of statistically correlated genotypes.
So there could be some "standard metabolic model" onto which you could overlay your genotype data, and this would be used to tweak that standard model to become your personalized metabolic model.

This presumes that all of us have more or less the same metabolic model, with various tweakable differences.

Your personalized metabolic model would then be used to predict or provide you with relevant information about your biochemical processes and resulting phenotypes (physical traits) particular to you.

[maxwatt]

But they won't deal with a few places, like New York, California, Arizona, etc.
I presume there are some kind of legal hurdles there.

I've heard that people in these backward regions drink the blood of children and wear clothes made of live tarantulas. I wouldn't deal with them either.

So manofsan, let us know what happens! This is pretty cool.

Not so. The tarantulas are dead, and the children are willing donners.

[manofsan]

Regarding that whole social networking thing, where "genenauts" would compare with each other, if you re-watch that webast from decode.com, you'll see Dr Stefannson pointedly mention that it's useless to arbitrarily compare some snippet of your genome with someone else's, because there's not enough context.

You might look at some gene that appears to increase your risk of cancer, and compare against a friend who doesn't have that gene, thinking that you're worse off. But then he could end up being the one getting the cancer while you didn't, because he had other high-risk genes you didn't compare against, or he had bad behavioral habits that you didn't account for, or environmental factors, etc.

But from what I see, this is then all the more reason for more genotypes and phenotypes to be recorded for collective comparison, in order to statistically drown out all the random uncontrollable ("noise") factors, and allow the gene's influence ("signal") to shine through.

If you consider Wikipedia/Youtube.com/Web2.0/etc, these things mushroomed by accumulating valuable information from user-submitted content (aka. "crowdsourcing")
Since we all have genomes and phenotypes, these are things we might potentially all contribute to some giant datastore for comparative analysis.

None of us wants to see our privacy violated, nor our rights, dignity, etc.
So therefore there needs to be a way to anonymously submit your genotype and phenotype data to some central datastore, in a way that won't adversely impact you.
It has to be absolutely secure and untraceable back to you, just like a swiss bank account.

Perhaps some wary people could even be given an option to have their data submitted when they die -- sort of like an organ donor card. After all, just like your organs, you may not consider your genomic privacy to be worth keeping after you die. You may consider it to be a useful gift to help others or to help society as a whole.

[manofsan]

Notice this recent article relating to a Microsoft announcement last month:

http://blogs.zdnet.com/BTL/?p=7078

Microsoft HealthVault launched in October with an impressive lineup of partners. Technically, HealthVault isn’t a personal health record, but a way to straddle numerous health records and combine them in one place. Microsoft understands the trust issue and spends a good amount of time talking about privacy. Here’s the privacy statement. From an FAQ:

    Q: What is Microsoft’s approach to privacy for the HealthVault platform?

    A: People willing to try Microsoft’s HealthVault must trust that their data will not be lost or misused by us or anyone else. Microsoft designed and built HealthVault with a strong foundation of security and privacy while consulting with experts inside and outside the company to augment our significant expertise in these areas. HealthVault’s privacy principles show that we’re committed to putting people in control of their health information:

      1. The Microsoft HealthVault record you create is controlled by you.
      2. You decide what information goes into your HealthVault record.
      3. You decide who can see and use your information on a case-by-case basis.
      4. We do not use your health information for commercial purposes unless we ask and you clearly tell us we may.

    Q: How can you promise consumers that their data will be safe? What measures have been taken to protect data “in the cloud?”

    A: HealthVault was developed using Microsoft’s Security Development Lifecycle, which means security concerns have been factored into every stage of the development process and our work has been subjected to internal and external security testing. HealthVault grants access to partner programs only to the extent a user allows such access. A partner program cannot access a user’s HealthVault record without the permission of a record custodian.

    Microsoft’s systems operate with extra precautions. With HealthVault we isolated traffic onto a virtually separate network and located our servers in physically separate, locked cages. All data that moves among our systems is encrypted, including all traffic to and from HealthVault, its users and its partners. Access to HealthVault data by Microsoft employees is tightly controlled and extremely limited to a small group of personnel necessary to perform essential operations.

    All of our back up data is encrypted, and every stage of its transportation is logged. We also log every time records are created, changed or read, leaving a clear audit trail.

So while it's not explicitly genomic, it's obviously a response to Google imminent foray into healthcare.
MS is particularly focusing then on the privacy issues.
But in light of the recent news on 23andme.com and decode.com, will MS now be forced to make some acquisitions in the genomics realm?

Part of the game is in gathering the genotype info, and the other part of the game is in gathering the corresponding phenotype info, including the disease pathologies, med history, etc.

So Google appears to be starting out from the genotyping end, while MS seems to be starting from the opposite physiological end. Will they meet in the middle? Who will race fully across the full span first?

[manofsan]

Here's the automated email response I got from knome.com , the venture being led by Harvard's George Church:

Dear ...,



Thank you for your interest Knome’s whole-genome sequencing service.  Knome is the first personal genomics company to offer whole-genome sequencing and analysis for individuals. Along with the opportunity of being among the first people in history to be fully sequenced, we provide you with a personal analysis and private consultation to review your results.  An additional point of distinction is that each Knome client receives a digital copy of their sequenced genome and retains full ownership of it.



Another vital and complementary mission of Knome is to provide researchers with access to sequenced whole genomes in order to enable Genome Wide Association Studies (GWAS).  We believe that approaching interested individuals directly is the fastest way to create the critical data needed to make significant scientific breakthroughs. That is why we also offer each client the option of allowing researchers to have access to their anonymous sequenced genome.  While we hope that a significant percentage of our customers will do so, it is not required.



Because of the significant costs involved in whole-genome sequencing, our services are targeted towards high net-worth individuals who have the personal desire and financial means to sequence their own genome.  The price for our service starts at $349,000.



I hope this information has been helpful.  If you would like to pursue this opportunity further or learn more I would be happy to be of assistance.



Best Regards,



Ari Kiirikki

VP - Sales

Knome, Inc.

101 Main Street, 16th Floor

Cambridge, MA 02142

Direct: (617) 528-1672

www.knome.com

Hmm, $349,000 ?
Well, that's a little bit on the high side, when you can get the million genotypes from decode.com for $1000

Yeah, it may be full genome sequencing, and it does seem a bit cheaper than the million-dollar cost of sequencing Dr James Watson, but frankly I think it might be worth it to wait awhile until the cost of full sequencing comes down, and meantime make do with the genotype sampling.

But if there are people who are rich enough, they may see price as no object, so if they want to pay that, then that's great for them. They can subsidize the cost of bringing it down to where the masses can get it.

Oh well, if people can fork out a few hundred thousand bucks to go to outer space, this trip to inner space seems much more worthwhile for the money.