Forgot your password?
typodupeerror
Science

Scientists Developed Artificial Structures That Can Self-Replicate 127

Posted by samzenpus
from the what-could-possibly-go-wrong? dept.
First time accepted submitter mphall21 writes "New York University scientists have developed artificial structures that can self-replicate, a process that has the potential to yield new types of materials. In the natural world, self-replication is ubiquitous in all living entities, but artificial self-replication has been elusive. The new discovery is the first steps toward a general process for self-replication of a wide variety of arbitrarily designed seeds."
This discussion has been archived. No new comments can be posted.

Scientists Developed Artificial Structures That Can Self-Replicate

Comments Filter:
  • Another step (Score:4, Interesting)

    by mysidia (191772) * on Thursday October 13, 2011 @07:11PM (#37708364)

    Towards Grey goo. [wikipedia.org]

    Or big nations making mechanical viruses as weapons, and ultimately... those creations at risk of being turned against their creator through malfunction, hackers, or worse.

    • Re:Another step (Score:4, Insightful)

      by ihaveamo (989662) on Thursday October 13, 2011 @08:57PM (#37709404)

      Already happened. Except it's pink goo. And it's us.

      • by EdZ (755139)
        Or rather, it's green goo (with life by volume being predominantly chlorophyll-using). Note how the entire planet has not become a single lump of homogeneous cells. Considering why this is will quickly tell you why the Grey Goo scenario is rather silly. If there is any danger from unrestricted replication, it would be more akin to the introduction of a foreign species into an ecosystem. However, unless whoever builds aforementioned unrestricted (and pointless: unless the replicator itself is useful, why wou
      • Ugly bags of mostly water

    • Re:Another step (Score:4, Insightful)

      by interkin3tic (1469267) on Thursday October 13, 2011 @09:29PM (#37709614)
      And you getting a better computer is just one more step toward skynet [wikipedia.org].

      Or big nations making artificial intelligence as weapons, and ultimately... those creations at risk of being turned against their creator through malfunction, hackers, or worse.

      Slashdot: news for technophobes. Lay off the LSD. Every technology can be abused. You're suggesting we shouldn't look into self-replicating structures because one day far down the road, some evil government agency MIGHT use it to unleash a horde of nanobots which will destroy the world? That's absurd.
      • Re:Another step (Score:4, Insightful)

        by Oxford_Comma_Lover (1679530) on Thursday October 13, 2011 @11:58PM (#37710472)

        The greater concern is that the technology will be used without understanding of the consequences. The Replicators in Stargate, for example, emerged from an experiment in which a childlike intellect taught its toys to make more of themselves. Research into self-replication, while reasonable, is not without nightmare scenarios or significant potential drawbacks.

        The cockroach is one example of such an experiment. Who is to say that in time, we will not create an example capable of out-competing us for some natural resource? So it is not without risk to experiment in self-replication. You can limit the risk, of course. Until someone makes the wrong kind of mistake at the wrong time. Kind of like researching Level 4 biohazards in a major population zone. If nobody does something dumb or protocols require fifty dumb things to happen at once for a problem and no massively unexplained events occur, it works just fine.

        • Except that StarGate is fiction.
          It'd be a pretty shitty story that went "Thousands of years ago, a civilisation created a self-replicating machine. It escaped into the environment. And was promptly turned back into raw silicon dioxide by the first bacteria that found it tasty."

        • It is always amazing to me when folks are willing to hold up a piece of fictional art to contest a15,000 year old (how long have modern humans been around exactly?) historical trend. We've been developing earth-shattering technologies that could be used to royally obliterate ourselves for awhile now. Think about it, designing a metallic blade that could, literally, break every other blade wielded against it back during the various transitions from stone-age to bronze-age probably convinced many of the folks
          • Technology breakthroughs have been occuring for thousands of years. The nuclear bomb, dynamite, machine guns, rifles, muskets, long bows, hell, even something as simple as putting a rotten corpse on a catapult and flinging it at your enemies could be considered technology. We've managed not to kill ourselves yet.

            Well sure, but keep in mind, we only have to end the Earth once in order for it to all be over, I am sure that at some point that will happen. SOMEONE has to get "Red button that kills everyone"-happy eventually, it happens all the time in movies.

            I am sure everyone in charge of anything that could possibly escalate into "Grey goo" is keeping a close eye on not letting it go too far.

          • I didn't say Stargate was evidence. The evidence was the cockroach. The possibility of outcompeting for resources is the most likely concern. The fiction merely illustrates it.

            And yes, all of those things were technology. Alexander the Great (the corpses) invented biowarfare in that way, but nobody thought it would destroy the world--it didn't have the potential to yet. Today, it can be pretty gruesome. Machine guns had a huge cost--world war I had many more casualties because of them. Other technolo

        • The Replicators in Stargate, for example

          Going to stop you right there. That's not an example of anything other than how technology destroying the earth is a common THEME IN SCI FI.

          The cockroach is one example of such an experiment.

          What.

          Who is to say that in time, we will not create an example capable of out-competing us for some natural resource? So it is not without risk to experiment in self-replication.

          You can't assume all possibilities to be likely just because of how dire they would be. Who is to say that by getting out of bed, you will not be exposed to a flu, and it will recombine with some bits of your genes, and create a world-ending superflu? No one, because it's possible. Extremely unlikely, but possible. With replicators, how about we wait until the

      • This is the same as the story of the Golems [wikipedia.org], so nothing new.

    • by satuon (1822492)
      Hate to disappoint you, but they're using DNA for the self-replication part, so no mechanical viruses, only organic.
  • And no doubt it will be called (wait for it) Project Genesis. [wikipedia.org]
  • by GuJiaXian (455569) on Thursday October 13, 2011 @07:12PM (#37708390) Homepage

    ...they've created an artificial structure that can self-replicate. The bad news is that it's Ice-9.

  • If this thing self-replicates to resemble a Robert Patrick, we're all screwed.
    • I for one would like to welcome our self-replicating overlords.
      • Re:Uh oh (Score:4, Funny)

        by Fnord666 (889225) on Thursday October 13, 2011 @07:54PM (#37708788) Journal

        I for one would like to welcome our self-replicating overlords.

        Especially if they look like Kristanna Loken [wikipedia.org]

        • by PopeRatzo (965947) *

          Especially if they look like Kristanna Loken

          The Academy snubbing her for an Oscar for her amazing performance in Uwe Boll's 2006 epic In the Name of the King (Dungeon Seige) is a travesty that I still have not gotten over. They gave it to that boring Helen Mirren for The Queen that year, and while Helen Mirren is decent, and certainly bangable, she doesn't have Kristanna Loken's acting chops. However, I understand Mirren can pick up quarters with her...you know, so I can see how the Academy might be swaye

          • by tehcyder (746570)

            The Academy snubbing her for an Oscar for her amazing performance in Uwe Boll's 2006 epic In the Name of the King (Dungeon Seige) is a travesty that I still have not gotten over.

            That film's got Jason Stathan, Ray Liotta and Burt Reynolds in it too. Seldom can so much ham have been gathered together in one place.

  • by catbutt (469582) on Thursday October 13, 2011 @07:20PM (#37708488)
    ...if you are allowed to have complex raw materials.

    Fire self replicates. Fallen-down dominoes self-replicate. The line between "chain reaction" and "self replication" is very blurry.
    • by Co0Ps (1539395) on Thursday October 13, 2011 @07:46PM (#37708720)
      Yes.. your argument applies to literally everything though... so dismissing anything as "just a chain reaction" is basically saying that "this is just a subset of the universe." In other words your argument is true but pointless. Disclaimer: I assume that the universe is a deterministic state machine.
    • by Kozz (7764) on Thursday October 13, 2011 @08:36PM (#37709226)

      ...if you are allowed to have complex raw materials.

      Fire self replicates. Fallen-down dominoes self-replicate. The line between "chain reaction" and "self replication" is very blurry.

      I don't think it's as blurry as you'd make it out to be. Fire and falling dominoes are instances of entropy , quite the opposite of what these scientists are after, I believe.

      • by artor3 (1344997)

        Everything is entropy. Self-replicating machines included.

        • by Kozz (7764)

          Everything is entropy. Self-replicating machines included.

          Really? Everything? So then, order and disorder are just a matter of opinion [wikipedia.org]? Well, let's put down our books and get in line for the local megachurch.

          • Even self-replicating machines can only represent local order. So what you view as "self-replicating" or not depends on the scale at which you examine it.

            • by Chris Burke (6130)

              So what you view as "self-replicating" or not depends on the scale at which you examine it.

              Yeah, at the scale at which the self-replication is occurring, it's self-replicating. :P

              These molecules are self-replicating. The parent they are present on is not. This is not exactly a deep insight.

          • by artor3 (1344997)

            I don't know what church has to do with anything. The laws of thermodynamics are abso-fucking-lutely clear that no process can decrease the total entropy in a system. That includes self-replication. The machines will inevitably give off waste heat or break down complex materials. That's grade school level stuff. Perhaps next you'd like to bitch at me for asserting that the world is round?

            • delta(G) = delta(h) - Tdelta(S) is the equation you are talking about I guess.

              Clearly the entropy of a system can be decreased during a spontaneous process if it is exothermic enough. I see you do refer to it giving off waste heat, which is necessary, but that's not the same. As far as I know cosmologists have not determined the source of order in the universe, and so you cannot say that heat is just an abstraction of entropy.

            • by Anonymous Coward

              Yes, but self-replicating machines (and life) increase the _local_ order of the system (e.g. the number of states in which a self-replicating machine exists is much smaller than all possible states of components) while increasing the entropy of the surrounding environment.

              The end-products of fire, falling dominoes, etc., are _locally_ less ordered than the fuel consumed. E.g. carbon ash and smoke are less ordered than pieces of wood or not-fallen dominoes. That's easy to see -- there are more ways to arrang

              • by astar (203020)

                Contemporary science has various sources of truth. A common one is experiment. Now the way thermodynamics is demonstrated experimentally, well, half a century ago, is to put the test system in a box, let it get to stable equilibrium and measure whatever goes in and out. So far so good?

                Okay, the other thing science gets its truth from is some sort of religious thingy. If you say the universe is going to some sort of heat death hell, then just how are you going to talk about putting the universe in a box

          • re: order/disorder/opinion

            I've often wondered about this: If a pile of stones are randomly thrown on the ground, they have a high entropy yes? What I place them carefully in a pre-determined (but identical to the 'random' version earlier) configuration directly on the ground? Surely one has high entropy and the other has low entropy - yet they're identical!

            Replace stones with some tiny molocules if you want.

            I'm obviously missing something obvious.

            • by Chris Burke (6130)

              Surely one has high entropy and the other has low entropy - yet they're identical!

              The entropy is the same. It doesn't matter how they got in that state. Deliberately placing stones (or molecules) in a disorded state is still a disordered, and thus high entropy, state.

              One useful way to think of the entropy of a system is in terms of information. The amount of information needed to describe a system represents its entropy (in a real and quantifiable way according to QM and Information Theory). A smooth rock wall requires less information to describe than randomly scattered rocks. A cr

        • by julesh (229690)

          Life is an example of negative entropy, i.e. a process that absorbs free energy from its environment and uses it to work against entropy by making a small section of that environment more ordered. That the absorbtion of energy creates more entropy than the localised reduction is given, but it doesn't detract from the usefulness of this observation. I imagine any realistic self-replicating machinery will have the same attributes. Fire and falling dominoes, however, don't.

      • What the other poster said. Exactly. Self-replication, the seeming creation of greater complexity out of less complexity, is only possible in open systems, that can get information or energy from elsewhere.

        The universe, being (as far as we know) a closed system, therefore, can only allow it in relatively small and isolated regions, precisely because of entropy. Local entropy can go down (self-replication) but inevitably it adds to the overall entropy of the universe.
        • The universe, being (as far as we know) a closed system

          The problem is that we don't know very far. There is convincing evidence that the cosmos is a house of mirrors and thus finite and closed. See: 'topological lens effect.'

          • My words:

            "The universe, being (as far as we know) a closed system..."

            Your words:

            "There is convincing evidence that the cosmos is a house of mirrors and thus finite and closed."

            I fail to see where there is disagreement here.

            • Sorry, I did confuse matters a bit. My point was merely that there are, and will always be, people arguing both sides of this. There is considerable evidence for the 'mainstream' view of an infinite universe too (which is why the 'topological lens effect' is so remarkable). Open vs closed and finite vs infinite are really two orthogonal concepts and I shouldn't have conflated them.

              But it is the words 'as far as we know' that are the real issue here. We do not know very far and we likely never will.
              • I see what you are saying. And that is true, as far as it goes.

                However, we do know some things, to within certain limits of accuracy, about the observable universe. And in our observable universe, overall entropy always increases, even though localized areas of "complexity building" can occur.
    • Those examples increase entropy. 'Self-replication of materials' must DEecrease enrtopy. Ergo, you need an energy input to a process designed or evolved to do the replication. It won't just 'go out there and do it'.
  • Artificial? (Score:5, Insightful)

    by Hentes (2461350) on Thursday October 13, 2011 @07:34PM (#37708608)

    They took DNA, a natural structure that can replicate, and modified it without breking that property. I wouldn't call it artificial self-replication.

    • At what point will we have a text based programming language that will compile the results into a DNA sequence? Coding the next plague wouldn’t be such good idea. Because you know, there are assholes in this world that would do just that.

      • At what point will we have a text based programming language that will compile the results into a DNA sequence?

        Automated production of short sequences is a well-established technology; Google on "custom oligonucleotide synthesis" and "custom gene synthesis" and you'll get links to a bunch of companies that will be happy to manufacture just about any sequence you want. Assembling an entire genome is harder, but not that much. So the answer to your question is pretty much "we're already there."

        Nobody's built any superplagues base pair by base pair yet, and honestly, I think it's not particularly worth worrying about

    • by Chris Burke (6130)

      Fair enough. That's true, but they modified it a lot, creating a structure that as far as we know doesn't exist in nature. It replicates without any of the normal cellular chemistry that makes it happen naturally including enzymes, which separates this from the Polymerase Chain Reaction technique for DNA amplification someone mentioned below. It also allows them to replicate structures other than proteins composed of amino acids defined by the normal base pairs.

      They even fabricated the DNA they used, so i

    • by julesh (229690)

      Erm... no. They took DNA, a natural structure which is half of a system that can replicate (the other half being a collection of enzymes that can transcribe the DNA to RNA, and a ribosome that can take RNA templates and make enzymes, some of which can produce more DNA or ribosomes) and rearranged it into an entirely new structure that doesn't require the assistance of a ribosome to replicate itself.

  • Look, it's a watch that self-assembles!

  • by RobinEggs (1453925) on Thursday October 13, 2011 @07:36PM (#37708628)
    I see the first five responses were about science fiction scenarios in which nanomachines destroyed human life.

    All that's really necessary to prevent the machines from getting out of control, however, is to design them with some chemical dependencies. If it needs gold or it can only incorporate carbon from certain uncommon molecules to grow then it can't get very far. Plus, natural selection will be true in part with any self-replicating thing. If they get out they'll have to struggle for resources just like any other form of life. There isn't any reason to automatically assume they'll be better at it simply because they're artificial.

    There are even scenarios in which it might be nice to design nanomechanical organisms with the express purpose of setting them free; I'd sure like an organism that got along by fixing the carbon in carbon monoxide, the ozone in smog, and the nitrogen in nitrogen dioxide to replicate itself. It could make Los Angeles habitable again, and its reproduction would be limited to the rate at which we produce pollutants.
    • by russotto (537200)

      All that's really necessary to prevent the machines from getting out of control, however, is to design them with some chemical dependencies.

      That assumes your machines don't mutate, ala Jurassic Park.

    • by erktrek (473476) on Thursday October 13, 2011 @08:21PM (#37709088)

      I've also heard that the "grey goo" scenario is a bit overstated given that:

      Organisms have already evolved optimal survival strategies over the millennia and if nanobots were made of organic material they would be "prey" to some of these.
      - and -
      The energy requirements for taking on such a task is unlikely to be satisfied in the current environment (especially if made of non organic materials)

      • Remember though, there is no animal on earth that is superior to a simple rifle. We're quite capable of building a bionuke, intelligence goes to places evolution never would or perhaps could.

        • That actually depends on the rifle and the animal. I'm pretty sure a simple .32 wouldn't bring down a full grown bull elephant... or, for that matter, even an octopus. And I'd still be interested in seeing the rifle that can take out a large ant colony.
      • Nonsense. Evolution excels at finding local optima, but there is nothing in evolution that suggests it will find global optima. With regards to energy requirements, consider for a moment that plants are green and not black. That means, they are not converting the green spectrum into energy at all. Which means, that an artificial plant, which absorbed all natural solar produced light spectrums (sp) would have more energy to work with than a natural plant does, which means it could definitely out compete the

    • Until the organism gets tired of smog and goes to the ozone layer. or the carbon monoxide eating one mutate to dioxide and the plants starve then when all suffocate. "but it would be limited by the amounts of pollutants we create"? no it would go after that small but very necessary amount needed. humans are notorious for screwing up there own environment because it seemed like a good idea at the time.
    • by geekmux (1040042)

      ...Plus, natural selection will be true in part with any self-replicating thing. If they get out they'll have to struggle for resources just like any other form of life. There isn't any reason to automatically assume they'll be better at it simply because they're artificial.

      You're absolutely right, except you're not taking into account the very mechanism that has allowed almost all current species to survive over the eons; mutation. And unless that is kept in check, then any new self-replicating "organism" will likely follow those same evolutionary lines. Chaos theorists will have a field day with this.

      Personally, I'm a little more concerned at whatever targeted resource is identified to "feed" these...our planet isn't exactly thriving these days with options...perhaps we'

    • by cstacy (534252)

      Make sure the machines have a lysine dependency (but cannot manufacture it themselves).
      Also, just make them all female.
      That should do it!

    • by Kjella (173770)

      Until it mutates into something that's not so restricted, and if it then has huge food sources it can devour alone it'll spread like wildfire. On a much less sci-fi note, a true global pandemic is still one of those really scary scenarios despite all the hype. If it first spirals out of control and you have people fleeing everywhere breaking quarantine it could get really, really nasty.

    • by Alef (605149)
      You mean they will struggle like rabbits and foxes did in Australia when Europeans brought them there? Introducing new species into an existing ecosystem quite commonly screw things up, and then we are still talking about life forms that are rather similar to what is already there. I imagine that some sort of artificial life form, that the existing species have no "evolutionary experience" defending themselves against, could do a lot more harm.
    • All that's really necessary to prevent the machines from getting out of control, however, is to design them with some chemical dependencies.

      It's going to be so hard to build machines without some chemical dependencies, that having them get out of control really won't be a problem. Nature has been trying to do that for millions of years and molds and bacteria do quite well but all exist pretty much only in a specific environment for whatever type they are. I doubt if we will ever be able to build a robust s

  • They're basically doing PCR-like reactions to clone DNA-like polymers. Same as DNA amplification has been done for years.
    • by julesh (229690)

      Except, at least as I read the press release, it appears to be self-catalyzing.

  • Does General Hammond know about this?
  • Wow, totally didn't see that coming :)
  • Welcome our Replicator overlords.
  • by ConsistentChaos (594109) on Thursday October 13, 2011 @08:10PM (#37708986)
    The Remote Self-Replicating Robot Explorer Probe. Be afraid.
  • Professor Farnsworth: "Bad news, everyone! Look at this infinite series representing the mass of successive generations of Benders. It's nonconvergent!"
  • Moya?

    Self replication of artifices (self repairs..) is what is going to be needed to long term voyages.

  • Alright, grey goo has the first comment, so how about evolution [imdb.com]?

  • by ThorGod (456163) on Thursday October 13, 2011 @09:08PM (#37709482) Journal

    Just when I thought I couldn't get any lower as a bachelor...machines go and gain the ability to replicate - I can't even do that!

  • Scientists have developed...

    OK, I have something that's been bugging me for a while.

    Scientists discover things. Engineers develop things. Frequently someone can do both, but they're two different processes.

    There's this terrible societal misconception that "scientist" means someone who works with technology. It leads to the mistrust of scientists because they're perceived as some ivory tower loonies who're lording technology over the populace.

    Scientists are people who apply the scientific method to acquire knowledge. I don't expect

    • by artor3 (1344997)

      I think you're thinking of theorists. Experimental scientists run experiments. Those experiments involve developing prototypes and running tests on them. Engineers also develop prototypes and run tests. The difference is the goal: scientists do it to learn, engineers do it to create a product to sell.

    • by Chris Burke (6130)

      You're drawing a line where none exists, and in the wrong place to boot.

      Engineers develop useful, practical things. Things that serve a particular need, that can be made economically, etc.

      Scientists develop things, too, but they aren't as often concerned with the direct utility or practicality of what they develop. For instance what these scientists developed is an ingenious proof-of-concept that could eventually be developed into a practical procedure for replicating materials, possibly by a chemical eng

  • From the article:

    DNA replication process involves complementary matches between bases -- adenine (A) pairs with thymine (T) and guanine (G) pairs with cytosine (C) -- to form its familiar double helix. By contrast, the NYU researchers developed an artificial tile or motif, called BTX (bent triple helix molecules containing three DNA double helices)

    In order to achieve self-replication of the BTX tile arrays, a seed word is needed to catalyze multiple generations of identical arrays. BTX's seed consists of a

    • by julesh (229690)

      Also FTFA:

      no biological components, particularly enzymes, are used in its execution

      So no, not PCR or PCR-like, as such processes require enzymes.

  • Run-away replication has already happened over forty years ago when the Starship Enterprise was overrun with tribbles. All it takes is a great intellect like that of Capt. James T. Kirk to deal with the problem.

    What's that? Star Trek was a work of fiction?

    That's different.

    Never mind.

    ----NR

  • The original article states.... "While our replication method requires multiple chemical and thermal processing cycles, we have demonstrated that it is possible to replicate not just molecules like cellular DNA or RNA, but discrete structures that could in principle assume many different shapes, have many different functional features, and be associated with many different types of chemical species," added Nadrian Seeman, a professor in NYU's Department of Chemistry and a co-author of the study. Yes... ma
  • Oh no, they have created the Seed.
    Quick, call Protocol Enforcement.

Everyone can be taught to sculpt: Michelangelo would have had to be taught how not to. So it is with the great programmers.

Working...