Scientists Build Possibly The First Man-Made Genome

hackingbear writes "Wired is reporting that researchers have created the longest synthetic genome to date by threading together four long strands of DNA. 'Leading synthetic biologists said with the new work, published Thursday in the journal Science, the first synthetic life could be just months away — if it hasn't been created already. [...] The ability to synthesize longer DNA strands for less money parallels the history of genetic sequencing, where the price of sequencing a human genome has dropped from hundreds of millions of dollars to about $10,000. Just a few years ago, synthesizing a piece of DNA with 5,000 rungs in its helix, known as base-pairs, was impossible. Venter's new synthetic genome is 582,000 base-pairs.' As a programmer, I'm most excited by the possibility of a new platform and the programming jobs that will be created by it."

Thanks for the SuperFlu, Craig!

[nebrshugyo]

If Venter and company royally screw-up, and create some bug that kills us all, or turns the biosphere to a pile of gray goo, nobody's going to make any money off of dandy, new, commoditized designer life forms. Where do I complain?

Impossible?

[XxtraLarGe]

Just a few years ago, synthesizing a piece of DNA with 5,000 rungs in its helix, known as base-pairs, was impossible.

Yet, somehow we've managed to have life on earth...

Re:Thanks for the SuperFlu, Craig!

[Fallingcow]

I'd be more worried about the tech becoming common enough and easy enough to use that anyone with $100,000 and some spare time can make a super-virus, or a bacterium that is extremely hardy and destroys wheat or rice crops, or any number of other nasty things.

Wonderful

[pnewhook]

As a programmer, I'm most excited by the possibility of a new platform and the programming jobs that will be created by it.

Geez. The LAST thing society needs is a bunch of synthesized clones running around with hacked up spaghetti code for genes.

An omission

[leob]

The article does not say if it's methylated in the right places.

Re:Thanks for the SuperFlu, Craig!

[KublaiKhan]

If 'grey goo' could happen from nanotech or biotech, then bacteria would have done it already.

So far, all that's happened is some assorted earthtone sludge.

Biology as the next Programming language

[DFDumont]

The final line of the paragraph scares me to death - I haven't met a programmer whom I'd turn loose on a DNA construction. It would be like handing a loaded, fully-automatic weapon, with the safety ground off, to a three-year-old; or asking them to complete a fully distributed ERP system written in assembler.
Just because we CAN do something doesn't mean we SHOULD. Perhaps if we constructed a complete corpus of biological effects, and dependencies of all currently known sequences (yeah right, like we're going to sequence every living organism on the planet) we could at least reasonably predict what the effect of NEW sequences might be. Until then the human race is the three-year-old. The gun is loaded. (waiting for the bang...)

Dennis Dumont

Procedural Abstraction

[Aram Fingal]

In the article, Venter says that they will need something similar to high level programming tools in order to accomplish useful modifications. I think that there is already plenty of evidence that genetic systems have procedural abstraction. In talking about gene activation, Biologists often use the term "ordered cascade" to describe what's happening when one gene activates a few more and those genes, in turn, activate other genes. If you think about it, it's exactly like subroutines of a program. Construction of the bacterial flagellum, for example, starts with the activation of one gene, which activates others, leading to the contribution of about 25 genes. These genes contribute various parts of the flagellum and activation of the cellular machinery to put it together and attach it to the cell wall.

Re:Thanks for the SuperFlu, Craig!

[QuantumG]

Well, the thing about nukes is that you need a good delivery device.. oh, and they're pretty conspicuous, so you'd need a secret silo too. That's, umm, really a lot of capital investment. That said, if more concerned citizens had access to "the button" then we'd hardly have any threat of military coup would we?

Monster Debugging

[MOBE2001]

As a programmer, I'm most excited by the possibility of a new platform and the programming jobs that will be created by it.

And who's going to debug all the billions of self-reproducing monsters you unleash into the world, pray tell?

Re:Thanks for the SuperFlu, Craig!

[bodan]

Actually, the nasty thing about nukes is that you don't need a very good delivery device. It doesn't matter that much where one goes off. Pretty much anywhere in an inhabited or industrialized place one could do a big mess. If anyone could have one, that's a lot of potential messes.

So a car or a backpack could be very good delivery devices for small enough nukes. You don't even need to be suicidal, you can just leave one somewhere. And suicidal people are not that rare; they weren't even before the current Muslim craze.

9/11 was such a big deal because it's hard to cause that much damage. You need a good plan and a lot of dedicated people to hijack plane and fly it into a building, even if you have access to guns or normal bombs.

Imagine for the sake of argument that a nuke was as easy to own as a gun is now. Of course, for rational people with something to loose they would be a good deterrent against aggression. Nukes work (almost) as an anti-aggression deterrent amongst the countries that have them, because the complexities of government tend to average out the crazies. But just barely. They don't stop anyone who doesn't care if they die, though, and there are plenty of crazies in the world.

Re:Procedural Abstraction -- Prolog?

[ChrisA90278]

will need something similar to high level programming tools in order to accomplish useful modifications. I think that there is already plenty of evidence that genetic systems have procedural abstraction.

Sounds to me like programming in Prolog.

For those who don't know... A Prolog program is a set of patterns and actions. When a pattern is "matched" it action occures. The set is unordered. A more modern and more widely used version of this is the language "Erlang". I think Erlang points to the way we will write very large systems in the future. For one thing it scales well to systems that have many, many cores. Procedural languages just don't scale so well. Also I think this style of programming could be adapted to formal methods, proof of correctness and so on.

Back to DNA. I think DNA simply reacts to patterns in it's environment with all of the DNA "looking" for these patterns pretty much in parallel

Re:Thanks for the SuperFlu, Craig!

[mc moss]

That is incorrect. Just b/c natural evolution didn't produce something yet doesn't mean it is not possible. A common fallacy is thinking that whatever species exist today are the pinnacle of evolution. It doesn't work that way. There is always some mutation that can happen that can produce something new.

Re:Ewww!

[namgge]

As a regular guy, I am NOT excited by the thought of thousands of fat, greasy programmers drooling over a test tube and a well worn copy of "Weird Science."

As a programmer, I can assure you that the first code implemented this platform that says "Hello World!" is going to excite you witless.

Namgge

Re:Thanks for the SuperFlu, Craig!

[Cheesey]

You forgot about all the other life on Earth.

We are the grey goo. The plants and the bacteria had a good go at spreading all over Earth, but we spread further and faster than any previous life. The "grey goo scenario" is limited by the assumption that energy is abundant, and indeed energy (food) shortages are all that stops us covering every inch of the world.

I, for one, welcome my fellow grey goo overlords.