DNA Takes Square Roots

Around these parts we’ve been known to discuss whether it makes any sense to say that the universe is a computer. There’s little doubt, of course, that parts of the universe are computer-like. And in case you are wondering, you can now officially remove DNA from your personal list of “things I suspect are not computers.”

Caltech researchers Lulu Qian and Erik Winfree have managed to coax 130 strands of DNA into performing what is unquestionably a calculation: taking the square root of a number. (Ars Technica post; Science paper behind paywall; open-access background paper.) Not a big number: we’re talking about four-digit binary numbers, so 15 at the biggest. And not very efficiently: with prodding, the calculation took eight hours. Moore’s Law isn’t really in danger here.

Still, pretty cool stuff. Mostly it’s interesting because it seems scalable: the authors claim that this kind of circuit architecture could be made much larger. It’s not the first biochemical circuit; RNA and bacterial colonies have been made into logic gates. But it’s the first to do something as elaborate as taking a square root.

Best of all, the authors decided to illustrate their method for a wide audience by means of a … whimsical YouTube video! Let’s hope this idea catches on.

The seesaw magic book: the computational power of DNA molecules
15 Comments

15 thoughts on “DNA Takes Square Roots”

  1. That was very clever and did a good job of explaining what’s going on. The music is a bit overbearing but other than that it was good.

  2. “with prodding, the calculation took eight hours.”

    Ah, they must have used teenager DNA.

  3. And the Earth is actually a huge computer performing an experiment for our overlords, the Mice…

  4. Sean: Back to the question, “Is the universe a computer?”, you took proper pains to arrive at a general definition of ‘computer’, coming up with an excellent one: “a system that takes a set of input and deterministically produces a set of output”.

    After re-reading that post from January 2008 I was struck again by what Tony Legget pointed out in the form of a question, which you identified as getting “right to the heart of the matter”: Legget asked, “What kind of PROCESS would NOT count as a computer?” [Emphases mine].

    It’s ultimately about process(ing). It occurred to me that your recent efforts to understand time within this context offers an even more general (or, at least, more abbreviated) definition: “any system that operates in time”.

    That ‘definition’ however seems excessively vague, since we have not arrived at an explicit definition, let alone an understanding, of ‘time’.

    But it in turn begs the question: What ‘system’ isn’t critically dependent on time? If system depends on process, and process is an inevitable consequence of time (whatever that may be), any CHANGE in the form of an arrived ‘output’ from precursory ‘input’ inevitably demonstrates a ‘processing’ and therefore a ‘computation’.

    On that basis it seems to me that there is nothing that ISN’T a ‘computer’.

    Existence is predicated on output from input. ANY change is an output from ‘previous’ input. ANY evolutionary dynamic is an artifact of ‘computation’ – or, as you put it in your 2008 post, a ‘calculation’.

    Without time, nothing ‘happens’: No time, then no change, no process, no system, no input or output, no determinism, no evolution, no relation, no “calculation”…and no computer. Even no room for concept. No time = no dynamic. ‘Information’ itself can have no meaning without the relations and comparisons offered by delta change in time. (There goes also the concepts of order/disorder, entropy and the 2nd Law, etc). Even ‘meaning’ and ‘relation’ make no sense without time. Curiously, however, the theoretical CONCEPTUAL relation between time and timelessness we pose to ourselves while we are necessarily obliged to brood about it WITHIN time at least points out the conundrum if not quite yet any serviceable solution…

    In any case there is obviously something crucial about time that is essential to ‘existence’.

    As some anonymous sage(s) put it in a nutshell: “[Only] time will tell”.

    Without time all is mute. That much seems to have been fairly strongly established. The understanding seems to have become fairly common knowledge amongst theorists wrestling with the issue, yet progress is nevertheless blocked by impasse as stubborn as the stasis of timelessness.

    Now back to tackling the problem of arriving at a general and explicit definition of time…which, if successful, would seem to simultaneously define existence. Indeed a mighty tall ‘order’. Man oh man. Good luck!

    Forgive me if I have strayed off the topic at hand. To induce molecules (DNA) to compute square roots is certainly interesting as a techno-toy, to be sure, but I find that what DNA/RNA has been up to over the last 3.5+ billion years as an agent of naturally-selected self-recursive evolution on this planet is a ‘computation’ that blows it away as a triviality.

  5. At the ~56 second mark the video dramatically declares (in elegantly ornate font), “Maybe one day synthetic molecular systems will be robust enough to scale up to complexities comparable to life itself…”

    Followed by the narrator’s compelling voice-over, “Once upon a time there was a magic book”…with a graphic showing a book entitled, “The Computational Power of DNA Molecules”

    Oh, wow. Holy friggin’ cow. EGAD.

    See, that’s what bothers me. Pardon me for expressing my annoyance at such pathetic declarations (besides the incredibly foul solicitation of the woo contingent in the target audience) but I will just point out here in this ostensibly physics-oriented forum that there is NOTHING ‘synthetic’ under the sun of physics.

    Not when genetic molecules have been doing their thing in splendid fashion without any ‘conscious’ (there’s another ridiculous concept) guidance for several billions of years. And not when upstart engineers pretend (as the handsomely-produced video emminently implies) to have developed something profound out of coaxing DNA molecules into achieving a desired computation.

    The chronic technologist conceit (wherein we have, for example, historically and quite frequently bragged about ‘conquering’ nature, as if we have discovered some means of circumventing the laws of physics rather than opportunistically utilizing them) is as if that mindset regards nature as an opponent to defeat or circumvent nature rather then treat ‘her’ as an ally to cooperate with: according to the inventor spiel, if ANY human contraption works at all, it MUST be because they have found a way to flout natural laws, the better to lard their claim with the appearance of ingenuity. It is rarely pointed out that the laws of nature permit their contraptions to operate in the first place. Without those laws, they could not even enjoy their contraptions.

    There can not be anything remotely ‘synthetic’ about it. The very use of the word automatically constitutes a lie.

    As ‘applied science’ technologists, we are all plagiarizers: we pretend to ‘create’ circumstances that are somehow ‘unnatural’ or our efforts somehow render the laws of nature impotent, and we frequently brag about it as a matter of habit. How many times have we read about how we have ‘conquered’ everything from disease to space? Yet the truth is that ANYTHING we come up with to meet a challenge posed by nature must ultimately defer to the laws of nature. We are nowhere near clever enough to generate our own set of laws which can somehow replace those which nature already provides.

    MAYBE one day synthetic molecular technologists (and technologists and explorers in general) will be ethically “robust enough” to scale up to something comparable to honesty and an admission of humility befitting their inescapable deference to the laws of nature, no matter how cute their contraptions are.

    I suppose now that a great noise might arise over a proper definition of the terms “synthetic” or “artificial”, in an effort to clarify such terms we are used to. Might as well throw in that dilapidated strain of concepts like ‘consciousness’ or ‘sentience’ or ‘spirit’ or ‘soul’ while we’re at it. Defining such is a fool’s errand. Save your breaths. False concepts don’t require definitional refinement. If anybody thinks they have found any that are at once CONSISTENT WITH as well as INDEPENDENT OF nature, a refutation of the necessity for any such particular definitions is immediately available.

  6. @ David #1 who says, “The music is a bit overbearing but other than that it was good.”

    No, not just a “bit overbearing”. It’s ALOT overbearing. This video is the mark of an enormous ego who found an opportunity to exhibit his or her imagined pianistic wherewithal. Unfortunately for the listener – whether trying to understand what the narrator is saying OR paying attention to the ‘music’ – it is an insipid excuse of a composition that has no business whatsover being there – along with the fancy BOOK graphics. Absolutely unlistenable. “Other than that”? NO, it is NOT good. Its absolutely unlistenable. Utterly unwatchable. Its horrible.

  7. The experiment sounds interesting, but the “whimsy” is downright annoying.

    Yes, Youtube can be used for good, but this was not an example of. Too dull; didn’t watch.

  8. This seems to be DNA using DNA to arrange DNA to perform a calca. A lot like a circus, only less fun.

  9. I think the basic problem starts with the assumption that computers need to use binary, instead of another core operating system. Remember that DNA has 4 letters, which can be use to create 26 ideograms used to concoct a virtually infinite number of paragraphs, incorporating different methods of construction etc.

    By making DNA imitate a binary computer’s method of operation we are limiting DNA and limiting our thinking.

  10. They aren’t really taking a square root as much showing that they can implement an arbitrary binary function of 4-bit input and 2-bit output, almost as just a look-up table. They did almost all the work for it – the actual machinery is not really as sophisticated as they made it sound. What would be much more impressive is if it actually used the same (or better) algorithm that regular computers would use.

  11. What is interesting isn’t that they computed a square root. What’s interesting is that they demonstrated computing any arbitrary logic.

    DNA has already been shown to have the ability to search for solutions to problems, and with reasonable performance.

  12. Pingback: Intelligence behind the nature? Scientists found that DNA can take square roots | News-Worthy Information

  13. “So the biologists aren’t real? What are they made of? Synthetic Biologists”

    Is it worse being a theoretical physicist? (Perhaps it sounds better if one can claim to be a high-energy physicist.)

Comments are closed.

Scroll to Top