Cool Science

[Shadow]

Tractor Beams!

[Gen. Volkov]

No way. That is beyond cool.

[Durza]

Yay science, you are so cool

[windhound]

http://science.slashdot.org/story/12/10/29/0241252/designing-dna-specific-bio-weapons

Bio-warfare is scary stuff

[Shadow]

"A few years" is a very optimistic estimate given the current state of sequencing tech. While it may be that we beat the NIH goal of $1000/24h genome in the next decade or so (Oxford Nanopore claims to have beaten it but biological nanopore methods are severely limited by shelf-life and their press-release is likely ridiculously over-optimistic), that level of engineering is still in the power-point stage.

I would measure time to that capability in decades if I were being optimistic.

Not that that makes it much less scary. But in all honesty, single-target weapons are a lot less scary than bioWMDs, which in all likelihood exist already.

[Gen. Volkov]

"A few years" is a very optimistic estimate given the current state of sequencing tech. While it may be that we beat the NIH goal of $1000/24h genome in the next decade or so (Oxford Nanopore claims to have beaten it but biological nanopore methods are severely limited by shelf-life and their press-release is likely ridiculously over-optimistic), that level of engineering is still in the power-point stage.

How is it ridiculously over-optimistic? They have the machines built already, they sequence a virus genome in seconds, and the human genome in hours.

[Shadow]

They have a very high error rate, relative to other methods (~4%, last talk I attended), and biological pores have a useful lifetime of a few hours and a shelf life of days if you are lucky. It's a lovely platform, but it will be obsolete the moment someone manages to build a similar device with a solid state pore, which will very likely be in the next decade.

They also aren't clear on the cost of their re-usable cartridges - if the bulk of that $900 cost is the cartridge, then over time the cost won't be that much lower than conventional instruments which cost hundreds of thousands of dollars upfront but can be used repeatedly after that, especially if the error rate is such that you need to use multiple cartridges per genome to reduce it. Where they win big is the prep time (low labour costs), but again, that has nothing to do with their sequencing platform, it's just some onboard standard microfluidics which could be used equivalently in any other pore-based platform.

Now, if someone could incorporate that lovely little enzyme they used to unzip their DNA into a solid state pore along with an appropriate read-head (in principle there's no reason they can't, it's mostly a matter of making the pores cheaply and consistently that it getting in the way) then I'd say they nailed it.

Don't get me wrong, I am ridiculously impressed at the device they've created. It proves conclusively that nanopore based sequencing approaches are heading in the right direction, and their microfluidic lab-on-a-chip system is nothing short of art (running untreated blood through a microfluidic system and extracting purified genomic DNA in-situ, holy crap). I just know that we can do a hell of a lot better with solid state pores, and it won't be long before we do, so I doubt that those devices will be widely used for very long.

[Gen. Volkov]

They have a very high error rate, relative to other methods (~4%, last talk I attended), and biological pores have a useful lifetime of a few hours and a shelf life of days if you are lucky. It's a lovely platform, but it will be obsolete the moment someone manages to build a similar device with a solid state pore, which will very likely be in the next decade.

Don't get me wrong, I am ridiculously impressed at the device they've created. It proves conclusively that nanopore based sequencing approaches are heading in the right direction, and their microfluidic lab-on-a-chip system is nothing short of art (running untreated blood through a microfluidic system and extracting purified genomic DNA in-situ, holy crap). I just know that we can do a hell of a lot better with solid state pores, and it won't be long before we do, so I doubt that those devices will be widely used for very long.

Well yeah, obviously its a transitional device, but what they have created has impressed the hell out of me as well, and it does make fast sequencing of the human genome a viable thing. The side where targeted bioweapons fail isn't the sequencing side, it's the targeting side. Viruses and bacteria mutate, if you create an infectious organism that is going to be effective, then you have to deal with mutation. There is no way to predict what will happen once you release that weapon. It may kill the person you want to assassinate, or it may kill everyone beyond the initial hosts, because the restrictor mechanism got lost in a mutation. If you can obtain a person's DNA, you can now sequence it quickly and with the right equipment and enough expertise, you could probably make a virus that does what it says in the article, at least in the lab. But like I said, the issue is, if you make a bug that is super-deadly to one person, there is a good chance that it will mutate and become super-deadly to everyone.

[Shadow]

http://www.youtube.com/watch?gl=GB&hl=en-GB&v=mcAq9bmCeR0

This is quite frankly the most interesting video that currently exists on youtube. Genetic algorithms are the coolest thing out there.

[Gen. Volkov]

Indeed, cool video Shadow.

Also very cool:

http://www.youtube.com/watch?annotation_id=annotation_467006&feature=iv&src_vid=mcAq9bmCeR0&v=SeTssvexa9s

[Kilkenne]

They should just make that a really huge .png that you can scroll down at your own pace rather than bother making a video at all.

[Gen. Volkov]

Hmm, yeah, that's a good idea.

[Shadow]

Apparently we can now turn pee into stem cells

[Gen. Volkov]

Well there you go.

[Durza]

That might stop some of the arguments against stem cell research