Personal DNA Sequencing Machine One Step Closer

oxide7 writes "A new, low cost semiconductor-based gene sequencing machine has been developed and may unlock the door to advanced medicines and life itself. A team led by Jonathan Rothberg of Ion Torrent in Guilford, Conn is working on a system which uses semiconductors to decode DNA, dramatically reducing costs and taking them closer to being able to reach the goal of a $1000 human genome test. The current optical based system costs around $49000 and is already on the market and being used in over 40 countries."

Comments

Yes!

[JamesonLewis3rd]

It's about time.

Re:Yes!

[MightyYar]

and life itself.

Agreed! Finally, LIFE!

Re:Yes!

[Compaqt]

You know it's coming: it'll be an attachment for the iPhone. (not an iPhan)

Re:Yes!

[RandomFactor]

No, no. This is DEFINITELY an Android app...

Re:Yes!

[luceth]

Actually, the Ion Torrent PGM has an iPod dock. In fact, it comes with an iPod in the box. Yes, really. See, for example, this link. Sadly, they don't have an iPod app yet - they were more focused on getting the product to market. (Yes, my lab has one.)

Re:Yes!

[RDW]

The world's most expensive iPod dock and it doesn't even come with speakers :-)

How are you finding it? Looks useful for smaller -scale (subgenomic or bug) projects where the capacity of a bigger machine would be wasted. And the concept of a 'massively parallel pH meter' is certainly cool.

For human whole-genome stuff there are already relatively cheap options. Illumina now offers a genome service for $4000 per sample in quantity:

http://www.bio-itworld.com/news/05/09/2011/Illumina-announces-five-thousand-dollar-genome.html

So the '$1000 genome' can't be far off, and may well be available as a commercial service before Ion Torrent can offer a competing solution for their machine. They already have a bit of a reputation for hype:

http://pathogenomics.bham.ac.uk/blog/2010/12/ion-torrent-hype-cycle-status-disappointment/

Note that the original article is misleading about the competing technologies, implying that the (existing) Ion Torrent costs $49000 because it uses 'optical based' technology. In fact the Ion Torrent (unlike the Illumina system, which does use an optical system) has always been semiconductor-based:

http://en.wikipedia.org/wiki/Ion_semiconductor_sequencing

The key issue, of course, is the cost and sequencing capacity of the (disposable) chips it uses, not the one-time cost of the machine. Right now, the comsumable costs are pretty high per sequenced base when compared to Illumina's. The Ion Torrent chip technology needs to be scaled up substantially if they want to offer a viable genome solution.

Re:Yes!

[luceth]

Well, it was just installed a few weeks ago - hasn't seen a whole lot of use yet, so not too many experiences on which to base an opinion. A single run costs ~$400 (an order of magnitude less than an Illumina Hi-Seq run!), but only gets you 10-20 million bases of sequence. (A human genome has ~3 billion bases, and you need ~10-fold coverage to make sure you get each base. In the Nature paper, they sequenced Gordon Moore's genome - and it took about 1000 chips to do it.) There's some talk of using it to QA libraries before a Hi-Seq run - for $400, I think that makes a whole lot of sense. I'm not sure that talking about Illumina approaching the magical $1000 genome is quite on target, either. The other thing most people don't talk about is read length - an Illumina run will get you enough sequence, but it's only in 50-100 base chunks. Reassembling an entire 3 billion base genome (6 billion if you consider both haplotypes!) is either extremely difficult, or actually theoretically impossible, depending on who you ask. The $1000 genome X-prize is to sequence a single genome, for $1000, in one week, to the quality standards of the original Human Genome Project (1 error in 1000 bases, IIRC). To do that, you're going to need long-range sequence data. I think that the nanopore sequencing approaches, that read a single piece of DNA for hundreds of thousands of bases at a stretch, have a much better shot at getting there in the next few years. The other interesting thing about the Ion Torrent PGM is that the magic happens on the disposable chip - and that can be upgraded without the rest of the machine (sensor interface, fluidics) changing. The next generation of chips is spec'ed to produce 10x as much sequence, and it's already sampling at early access laboratories like Baylor and MIT.

Essentially a Proprietary Hydrogen Ion Sensor

[eldavojohn]

From the short video on their site (youtube alternate), it appears that this technology relies on a DNA template across thousands or millions of wells on a chip that emits hydrogen ions every time a base is incorporated into a DNA strand by a polymerase. I'm not a biologist but it looks like a pretty neat idea and I certainly hope it works as well as they say it does. I guess even if your sensor isn't that great at classifying between A, G, C or T then you can just build more wells on the chip and look at the statistics. I'm not sure how they ensure that one process is going on in each cell but I'm hoping this yields some cheap and fast accuracy. This would be a huge boon for research -- hell you could start up some hobby work very quickly and (relatively) cheaply since it's such a straight forward process.

Re:Essentially a Proprietary Hydrogen Ion Sensor

[ColdWetDog]

hell you could start up some hobby work very quickly and (relatively) cheaply since it's such a straight forward process.

They're hoping to get the cost down for a total organism sequence to around $1000. That's one hell of an expensive hobby even before you start asking yourself what you plan to do with that rather impressive amount of information. I suppose you could use it to blackmail your family and friends by turning up all sorts of paternity issues - whatever floats your boat.

Researchers, OTOH, should be very happy about the trend.

Re:Essentially a Proprietary Hydrogen Ion Sensor

[Anonymous Coward]

All,

Be advised that scientists now recignize 8 bases of DNA: http://www.physorg.com/news/2011-07-scientists-seventh-eighth-bases-dna.html

Re:Essentially a Proprietary Hydrogen Ion Sensor

[Anonymous Coward]

The problem at this moment is that no machine is able to meet all of the criteria set by the gold standard Sanger technology. Sanger gave long reads with high accuracy which gave good assemblies but was terribly expensive and slow. The new technologies tend to focus on high throughput (gigabases of raw sequence) at the expense of read length and accuracy. Imagine trying to assemble a 4 Gb genome from 200 Gb of 100 bp fragments (or less) and suddenly it doesn't look quite as cost effective. Some of the newer technology is supposed to give very long read lengths which helps the assembly problem, but the accuracy is horrendous. I don't know what the stats are on the new machines, but the first company that can come out with a technology that can do everything Sanger does at a hundreth the cost will be sitting pretty.

Re:Essentially a Proprietary Hydrogen Ion Sensor

[fuzzyfuzzyfungus]

You can already do most paternity and forensics stuff, to a quite usable degree of confidence, with much smaller snippets. Trying to do so on the very cheap might well get you results from a lab that can't be bothered with minor stuff like negative controls or not fucking up on a regular basis(luckily, this never, ever, ever happens at crime labs); but you can get it today, cheap. Here's an over-the-counter option for $150(no particular endorsement implied, of course, just an example of what you can find in totally mainstream shops with 30 seconds of searching...)

Whole-organism sequencing will likely remain a research tool for quite some time. The snippet-based stuff is already as or more accurate than the people doing it, and whole-organism for medical purposes will be largely snake oil(although there will certainly be people selling it) until we actually have the knowledge necessary to make meaningful inferences from those sequences.

Re:Essentially a Proprietary Hydrogen Ion Sensor

[Anonymous Coward]

Impressive amount of information?
My porn collection is bigger!

(To be honest, I have two collections. I use this program called "Firefox" to browse the second one. Whatever you type in there, it results in porn. ;)

Re:Essentially a Proprietary Hydrogen Ion Sensor

[fusellovirus]

Each well is just large enough to allow one microsphere in it. This microshpere has copies of the DNA to be sequenced attached to it. Each of the 4 bases are then added to the chip one at that time and the pH is measured. When a nucleotide is incorporated it changes the pH and the signal is recorded. The $50K pricetag is a little decieving, you also need the machinery to produce the DNA coated microspheres and a hefty server to process the millions of ~100 base reads form the machine and assemble them into a useful file. I'm not sure of the specs on our server but I believe it has 12 cores and around 20 gigs of ram, and takes 2-8 hours of processing for each run. There is also the consumables to think about. The DNA oligonucleotides that server as primers and the chemicals. All tolled it is more like $150K for the parts adn 500-1500 per run.

Re:Essentially a Proprietary Hydrogen Ion Sensor

[mlvlvr]

The technology deposits DNA fragments, many copies of a single fragment, in each of millions of wells on a chip. A, G, C, or T are flooded over the chip sequentially, and if one is incorporated by the polymerase, the reaction, not the chip, emits a proton and the sensor beneath each well picks up the change in voltage. The cycle is repeated hundreds of times, until the end of the fragment is reached or the quality of the reads falls off. Lets say you get a few million fragments of a few hundred basepairs, you get close to 1B bp per chip. One of Jonathan Rothberg's insights was that the large, well funded core labs with the largest most expensive instruments, could not give the individual investigator the turnaround time she needed. So, he sought to deliver the solution in a low cost instrument that allows the investigator to decouple herself from the core labs. I've heard him say that Ion has the largest installed base of any sequencer, and I've heard from many investigators that what appeals to them is this very independence.

Re:Essentially a Proprietary Hydrogen Ion Sensor

[Xemu]

They're hoping to get the cost down for a total organism sequence to around $1000. That's one hell of an expensive hobby even before you start asking yourself what you plan to do with that rather impressive amount of information.

www.meet-your-genetic-match.com

You heard it here first.

Re:Essentially a Proprietary Hydrogen Ion Sensor

[WorBlux]

and whole-organism for medical purposes will be largely snake oil(although there will certainly be people selling it) until we actually have the knowledge necessary to make meaningful inferences from those sequences.

It's called statistics. Research is an important part of medicine. Really cheap whole genome analysis will let you more easily and cheaply find the short sequence test you want to make as a diagnostic, and if lucky clue you into the metabolic pathway involved leading to a place to start with drug design or which drug to try for off-label treatment of rarer genetic diseases or deficiencies. Or you can test to see what genetic factors are associated with side effects or particular efficiency of a drug compound. Having a full sequence on file will eventually save money time. Instead of paying someone to take a cheek swab, postage to a lab, waiting a few days, paying the lab techs, and then finally getting a result, the doctor can just pay a subscription to a database which will grep your DNA for any relevant issue whenever the doctor make a prescription, or enters symptoms to try to make a diagnostic. Having an opt-in central database would also make research a lot easier, as you would just have to write a filter to pick out candidates and analyze their medical history against their genome. IF done properly, medical research and medical practice will an integrated feedback loop with a low delay. Add in other data gathered with inexpensive new tech, there's no reason most people wouldn't live to see 100. The biggest problem I think would be the patent issues that would be stepped upon by grepping for certain patented sequences.

Re:G, A, T, C

[TaoPhoenix]

Time to view all news through an Evil filter.

Nice try for these guys coming up with tech. Now the Insurance companies will require one so they can save $50,000 per "hit" on a "predisposed condition".

Hello, Gattaca.

Investment

[Anonymous Coward]

Might be time to invest in hard drive manufacturers.

Gaaa the numbers are all wrong

[Anonymous Coward]

it does not cost $49000 to sequence a human genome. it's cheaper than $1000. this system is more expensive than current models. The difference is that this has some hope of having more headroom to mature. But eventually it's going to come down to whichever method has cheaper pre-processing. in the long run the machines will be cheap.

Re:Gaaa the numbers are all wrong

[Anonymous Coward]

In addition, these guys are still talking about read sizes of hundreds of base pairs. This technology will help make reading data faster and cheaper, but reading the data is not the bottleneck today. The bottleneck is still post-processing, i.e. alignment, assembly, scaffolding, heretozygosity resolution and so on.

Re:Norway

[Anonymous Coward]

Hey, (offtopic) was there a story about the terrible situation in Norway?

Yes

Or is it not news for nerds?

Doesnt matter.

Re:G, A, T, C

[chemicaldave]

Already covered with the Genetic Information Nondiscrimination Act.

Re:G, A, T, C

[TaoPhoenix]

Valiant attempt sir.
Too bad we don't bother to follow laws anymore.
We're 7 days from economically blowing up our country, so a little "ethics" law won't stop anyone.

Coming to ...

[PPH]

... police cars near you.

Re:Coming to ...

[c0lo]

Is there an app for that?

Re:Norway

[tbird81]

Do you have a link?

I had a look through the recent slashdot stories and did a few searches, but couldn't find it.

Thanks.

Sounds great?

[Grizzley9]

This sounds great and all, but can someone please explain to me what this will mean for those that get the test done and have their genome on a flash drive? I can understand yeah you might have male pattern baldness or increased risk of cancer, etc but there are other ways of determining that currently including yearly blood tests and the like.

What will having your genome sequenced actually do for you, today, right now? Why should I pay $1k or even $50k for something like this?

Re:Sounds great?

[ColdWetDog]

What will having your genome sequenced actually do for you, today, right now? Why should I pay $1k or even $50k for something like this?

Virtually nothing. There have been several companies that have tried to cash in on the 'personal genomics' craze (23andMe comes to mind) that actually didn't do a whole sequence, just SNP (single nucleotide polymorphisms) that purported to help you determine your risk of various diseases. Except that they found precious few diseases that had clear links to SNPs. Whole genome sequencing will be even harder to figure out.

So other than bragging rights, it does you little good. For research purposes, getting fast, accurate (and see the AC's post above concerning the Sanger Method and accuracy) and cheap sequences will be very useful. For personal use, not so much.

Re:Sounds great?

[Anonymous Coward]

What will having your genome sequenced actually do for you, today, right now? Why should I pay $1k or even $50k for something like this?

Give you an edge against the house when deciding what bet to place against your health insurance company.

It may not be a winning game for an individual to pay $50K (that is, 1000 individuals paying $50K each would likely not win themselves $50M in having the healthy ones cheap-out on insurance premiums, and the ones doomed to cancer/alzheimer's/etc loading up on insurance in anticipation of getting big payouts) to get that edge, but at $1K per user ($1M over our hypothetical 1000-user population), it probably is.

"The sky was the color of a TV tuned to a dead channel. Feeling like a refugee from a Cyberpunk novel, AnonCow had just left the back-alley DNA sequencing shop after receiving a clean bill of health. One of his parents had Alzheimer's, but now he knew himself, and his children, to be in the clear. Basking in the warm glow of the sun, he smiled at the beautiful blue sky, and cut his family's health insurance plan - which, due to his genetic history, had heretofore cost him $6000/month - to the bare bones. They say that once upon a time, televisions tuned to dead channels looked different. "

Re:Sounds great?

[Rich0]

In the 1950s computer scientists speculated on the creation of computers that wouldn't cost millions of dollars and fill entire rooms. Imagine what such power in the hands of the individual consumer could bring!

However, this was completely silly thinking. As anybody who actually had access to a computer knew, being able to program one was far beyond the reach of individual consumers. No, computers are strictly the domain of research, and have little practical use for individuals.

Or, perhaps what is the domain of research today, will be practically employed tomorrow. We only first sequenced the human genome, what, barely more than ten years ago, and now we're talking about doing it for every person on the planet!

Re:Sounds great?

[fusellovirus]

Testing for genetic differences that lead to well characterized diseases are better done methods other than sequencing. Methods such as real time PCR and microarray analysis costs large amounts of money to set up and validate but are much cheaper once they are up and running. This is why places like 32andme are $400 and not $1000s. The power of next gen sequencing like ion torrent is that you can go fishing for things you don't know about. Deep sequencing is a good example. Many cancers result from spontaneous mutations or chromosomal rearrangements. Catching them early requires identifying sequence differences in a very rare cells. Because Ion torrent and many of the other next gen sequencers are capable of massively parallel sequencing they can sequence regions of DNA prone to rearrangement or oncogenes over and over, finding that one in a 1000 cells, or one in 100,000 cells that is the beginning of cancer. They can also be used to sequence novel genomes, such as newly emerging strains of pathogenic E. coli, and find what changes have led to the change.

Re:Sounds great?

[Urkki]

If we ignore that GPP talked about "today, right now", then home gene sequencing alone will be of very little use, just like home computers alone were of very little use except as a hobby for a decade or so. However, once we have molecular construction machinery available, then we can have customized drugs (including, but not limited to the medicine kind...) made at home. But that's quite a few decades away, possibly not happening in this century, and most likely will be illegal as long as corporations rule the world.

Re:Sounds great?

[julesh]

There are two things:

1. Full genome sequences are useful for research. Lowering the cost opens up new avenues of research (e.g. comparing genomes of thousands of individuals, rather than the much smaller samples used today). Imagine performing automated correlation tests on, say, ten thousand sufferers of a particular disease versus ten thousand control subjects. The results could be quite interesting.

2. Patented genetic tests normally work by patenting a method of isolating a particular gene and determining its presence. If this can be done by a quick grep, the chances of a patent actually being upheld become rather slim. This should drive down the costs of genetic testing.

Neither of these means you should rush out and get your genome sequenced. It's just the possibility that it can be done cheaply if necessary that's important.

Re:Sounds great?

[Rich0]

Well, you don't need to make your own drugs to benefit - we already have some knowledge about the correlation between genetics and response to commercially-available drugs. If this were expanded there would probably be a greater variety of drugs on the market, since it would be easier to get them approved (once you figure out who gets the side-effects vs the benefit you can just make the indication "... in people with SNPs x,y,z").

Home chemical synthesis is an entirely different technology, and might someday be practical. However, if a particular chemical has commercial value, I'm sure some lab somewhere will sell it.

As far as legality goes - you can make any chemical you want as long as it isn't patented. Drugs aren't patented unless commercial companies do the development work (ie they pay for the clinical trials). The easiest way to have patent-free drugs is for national governments to actually start funding drug development (not just basic research). That is to say, after discovering some groundbreaking new biological mechanism the NIH or whatever would actually start testing compounds for activity/etc, and then pay to have tens of thousands of people take the most promising compound and figure out if it works. They could then get marketing approval and never patent the drug, or patent it defensively but license it freely, or perhaps freely to domestic manufacturers or overseas companies whose governments reciprocate and spend a similar amount (per-capita) on R&D. Private companies would still be free to compete, but of course prices will be influenced by the competition. The government could even outsource some of the development to private industry/etc, while retaining patent rights.

I'm not a big fan of just getting rid of drug patents - it seems like killing the goose that laid the golden egg. In order to have new drugs without patents you need a lot more government R&D. However, simply having a lot more government R&D can probably solve the problem without even getting rid of the patents. In a world where we have many competing systems for discovering drugs we can evaluate the costs of all the models and figure out what actually works best.

More than 4 bases in DNA

[Thorfinn.au]

There are at least 8 bases in the DNA sequence and this will be only looking for 4 of them
http://www.sciencedaily.com/releases/2011/07/110721142408.htm

Re:More than 4 bases in DNA

[ColdWetDog]

Your article talks about methylated bases - the basic DNA base pairs that have been modified AFTER replication. And you are correct, non of the sequencing methods (AFAIK) can determine the extent of methylation or demethylation of a given base. This is likely to be rather important although the mechanism and the level of importance has yet to be determined.

Re:More than 4 bases in DNA

[Pseudonym]

If by "sequencing method" you mean the whole process from wet lab to assembly, then we can indeed detect methylation. Look up, for example, methylation-specific PCR.

Re:More than 4 bases in DNA

[Anonymous Coward]

There are at least 8 bases in the DNA sequence and this will be only looking for 4 of them

That's not a limitation of this particular technology. There are suggestions that the IonTorrent may be able to detect methylated bases (because there would be a change in pH that is slightly different from the change after incorporation of a standard base), and there's no particular reason why you couldn't flood the chips with more than 4 bases, if in some future experiment you considered that necessary.

Re:More than 4 bases in DNA

[julesh]

Yes, but as methylation of DNA is not preserved when the DNA is duplicated, any effects of these additional bases are only localised. AIUI, different cells in your body will have different methylation states. This means that the information about which precise bases are methylated is not especially useful for most things you would want a gene sequence for. It's important for those studying internal cell processes, especially DNA repair, but otherwise not really interesting.

23andMe

[tbird81]

Not sequencing of course, but checks a whole lot of SNPs. I've been quite happy with the information you can get, and you can download all the data yourself.

Problem is, once you get the sequence, it's hard to know what to do with it.

Re:23andMe

[Anonymous Coward]

Make them noted in public records and license them under terms similar to GPL? After all, why should only a few profit from genetics related technology and knowledge when all the actual parts are available for free in nature?

If the tech is cheap enough, I'm sure there could be at least one group put together for an OpenDNA project.

Re:23andMe

[ColdWetDog]

Not sequencing of course, but checks a whole lot of SNPs. I've been quite happy with the information you can get, and you can download all the data yourself.

Problem is, once you get the sequence, it's hard to know what to do with it.

Why are you happy with the information if you don't know what to do with it?

Re:23andMe

[Scubaraf]

There is an open source DNA sequencing project out there: http://www.polonator.org/

But the fact is that it is still expensive as a hobby.

Re:23andMe

[Gmooron]

Check out http://www.snpedia.com/

Re:Norway

[JustOK]

no computers died, but many user sessions abended.

already covered with Citizens United

[decora]

gotcha! check and mate my friend.

Not particularly exciting

[Anonymous Coward]

It is hard for those of us in genomics to even fathom how this paper was published. It cheapens all of our publications. The sequencing mechanism is exciting- however it is not competitive with any existing technology, has a cumbersome library preparation protocol, produces very small amounts of noisy reads and they charge way too much for the chips and reagents. It will be very surprising if many labs adopt this, it seems like most people are buying them because they are so inexpensive rather than having some use for them. Perhaps with improved lib techniques, better accuracy, and longer reads, it may be viable.

Re:G, A, T, C

[DigiShaman]

Uhuh, sure. That will be enforced. And Social Security Numbers aren't supposed to be used for public identification of citizens either. Financial institutions often require the number to run a credit check or to reset the password at the myfico.com website. It's often used to verify account holders when speaking to a TSR or CSR of your local cable/phone provider. Though they don't require it as long as you have your account number handy.

No, this law is one of those wink-wink-nod-nod kind of things. You can genetically profile anyone just as long as you don't admit to doing it.

Sorry.

[Anonymous Coward]

Posting anonymously because Amy Winehouse's death reminded me.

My ex girlfriend used to claim addiction was a choice, and anyone suffering from an addiction had chosen their own path and could get out of it if they wanted to.

She also occasionally used cocaine, and used her (and my) ability to go on and off it to back up her claims.

So I introduced her to meth. I wouldn't touch the stuff, seen what it can do.

She's been in and out of rehab for four years now, and meth has fucked up her life.

I feel guilty, all for an "I told you so" that didn't feel all that good in the end.

Sorry Lisa.

Re:Sorry.

[Anonymous Coward]

Now maybe next you try saying that to her.

the dawn of hype itself!

[tverbeek]

"...may unlock the door to ...life itself"

Well, it's about time! The universe has been sitting lifeless for so long, and here we sit, unable to make any!

Personal DNA Machine

[hyades1]

For the love of god, don't let one of those machines find its way into the southern United States. Can you imagine what damage it would do to the family trees if they had ironclad proof of how many hillbillies didn't understand that even if they got divorced, they're still brother and sister?

Re:Personal DNA Machine

[DNS-and-BIND]

Here it is, bigotry and hatred, right out on display in public and modded funny. Disgraceful.

Re:Personal DNA Machine

[maxwell demon]

For the love of god, don't let one of those machines find its way into the southern United States. Can you imagine what damage it would do to the family trees if they had ironclad proof of how many hillbillies didn't understand that even if they got divorced, they're still brother and sister?

After divorce, your wife becomes your sister?

This is actually LIFE Technologies.

[Anonymous Coward]

NASDAQ: LIFE

Re:G, A, T, C

[Anonymous Coward]

Did you see Gattaca ? It talked about an act there too... Only nobody cared.

Re:G, A, T, C

[Anonymous Coward]

Stop thinking so small. This is actually going to be used to replace the REAL ID. I am sure once they get closer the DHS and DoJ will come in with real money and make it happen. In reference to GATTACA, remember how everyone gained access to everything? DNA sequencing via blood sample.

This will make all forms of physical ID cards obsolete. You carry this around with you all of the time and it cannot be changed or stolen (except how the movie did it). One master, federated DB across all 50 states with ties into an International system and they can instantly identify you all over the world.

Remember 3 billion base pairs, 99.9% are identical across all humans (twins = 100%) on the planet (or so they say). That leaves you with only having to determine the differences in the remaining 3 million pairs. With only 2 options (AT or GC) at 3 million positions, I believe that comes to only 6 billion unique combinations. Thank goodness our worlds populations is less than that. Oh wait...

Gattaca

[Chuby007]

Gattaca, Gattaca, Gattaca !

Re:G, A, T, C

[thrich81]

That brings up what the real purpose of Insurance is. Is its purpose to level out the risk for everyone no matter what their predisposition to the risk of a loss is? Or is its purpose to level out the risk of a catastrophic loss to an acceptable level among a population with about the same predisposition to suffering the loss? I don't think anyone would argue that in the case of automobile insurance the amount you pay for insurance should reflect the risk and severity of the claims you will probably make on the policy -- otherwise the good drivers will be subsidizing the bad ones. In a pure insurance situation the "good risks" should not generally subsidize the "bad risks". Now for medical insurance we get all squeamish about this principle, especially because some risks are out of the individual's control (heredity) and some lifestyle risks are well within their control. If I have reliable knowledge that I am not a high medical risk for anything at this time, then from a personal financial point of view I would like a medical insurance policy which only includes similar individuals. Do we want to do this as a society -- I don't know. I know for a fact that I don't want my medical insurance premiums subsidizing someone else's risky lifestyle choices. Is it desirable for the "good genetic risks" to subsidize the "bad risks"? -- that is question which needs to be honestly laid out and discussed, no matter how uncomfortable. Clearly some individuals are a priori higher medical risks than others. Saying, "everyone could be a bad risk" is not an honest way to approach it.

I predict

[Anonymous Coward]

I predict that the very first use of this new technology will be a police surveillance device.

Confusing summary

[ClickOnThis]

The summary says:

A team led by Jonathan Rothberg of Ion Torrent in Guilford, Conn is working on a system which uses semiconductors to decode DNA, dramatically reducing costs and taking them closer to being able to reach the goal of a $1000 human genome test. The current optical based system costs around $49000 and is already on the market and being used in over 40 countries.

Unfortunately this summary is missing two crucial sentences (between the above two) from the article:

Typical DNA sequencing machines use optical technology instead of semi-conductors. While fast, optical technology is expensive and complex.

Without the above sentence, the summary sounds like Ion Torrent sells "[t]he current optical based system" which they do not. (However, it is true that Ion Torrent is part of another company (Life Technologies) that does in fact sell an optics-based system called SOLiD.)