Slashdot Mirror

← Back to Stories (view on slashdot.org)

Complex Logic Circuit Made From Bacterial Genes

Posted by samzenpus on from the feeding-the-machine dept.

another random user writes "Just as electronic circuits are made from resistors, capacitors and transistors, biological circuits can be made from genes and regulatory proteins. Engineer Tae Seok Moon's dream is to design modular 'genetic parts' that can be used to build logic controllers inside microbes that will program them to make fuel, clean up pollutants, or kill infectious bacteria or cancerous cells. The circuit Moon eventually built consisted of four sensors for four different molecules that fed into three two-input AND gates. If all four molecules were present, all three AND gates turned on and the last one produced a reporter protein that fluoresced red, so that the operation of the circuit could be easily monitored."

37 comments

  1. 99.9% OF GERMS ARE GOOD FOR YOU !! by Anonymous Coward · · Score: 0

    And the rest, bad !!

  2. Um... by Type44Q · · Score: 1
    ...correct me if I'm wrong, but doesn't life, by definition, do this already... but in a nearly-infinitely more elegant, efficient fashion?

    This research strikes me as comparable (scrambling around for a suitable analogy here...) to welding a bunch of pairs of vise-grips into a shape vaguely reminiscent of a pair of pliers and then loudly proclaiming that one has achieved the ability to manufacture impressive tools...

    1. Re:Um... by Anonymous Coward · · Score: 0

      Maybe he wants to patent his new forms of life before Apple or God does?

    2. Re:Um... by Anonymous Coward · · Score: 0

      But you don't know how to make Vise-grips(tm), you can only find them in your environment. So this is as close as you've gotten.

      PS - Just because the submitter is loudly proclaiming something doesn't mean anyone with a clue is doing the same thing.

    3. Re:Um... by Anonymous Coward · · Score: 2, Insightful

      Sure, but we don't know how to program life to do what we want. I'm not convinced that implementing digital logic circuits is the way to accomplish his goal, but I'm not convinced that it isn't either. This may be a dead end. Perhaps we will have a much better method later. However, this seems to be an approach that we can investigate now.

    4. Re:Um... by Belial6 · · Score: 1

      No, it is more like finding a bunch of hammers around, welding them together in the shape of a pair of pliers in a land where no pliers previously existed, and loudly proclaiming that one has achieved the ability to manufacture impressive tools.

    5. Re:Um... by drinkypoo · · Score: 1

      correct me if I'm wrong, but doesn't life, by definition, do this already

      It's not clear whether anything in our heads actually operates like a classic logic gate. Our brains are analog. There may be digital logic in there somewhere, but it's probably all emulated.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    6. Re:Um... by Gorobei · · Score: 1

      ...correct me if I'm wrong, but doesn't life, by definition, do this already... but in a nearly-infinitely more elegant, efficient fashion?

      This research strikes me as comparable (scrambling around for a suitable analogy here...) to welding a bunch of pairs of vise-grips into a shape vaguely reminiscent of a pair of pliers and then loudly proclaiming that one has achieved the ability to manufacture impressive tools...

      Yes, but that is what building abstractions to support higher order systems is all about. The computer I'm typing this post on does about 12 billion ops/sec, many more in the GPU,) and probably trillions of nand gate state changes per second. All that to underline "nand" as a misspelt word to me. Every abstraction layer in a computer (general purpose CPU, VM model, garbage collection, multi-threading, communication protocols, ACID file systems, HLL representations, etc) costs about 10x in terms of terms of providing the abstraction versus "coding to the metal" in the layer below. It's absurd that the system I work on can do about 10K if/then/else checks a second when the hardware can do a million times that, but that is the price I paid for an abstraction that lets a thousand people work together in a shared cognitive environment.

      Hoisting bio systems up to a few gates may seem useless now, but it'll seem common sense and obvious in 10 years time.

    7. Re:Um... by ColdWetDog · · Score: 2

      Stop with the analogies already. That's like trying to fix a radiator leak with WD-40.

      Yes, life does infinitely more complex logical operations than these simple logic circuits. But we can't control that stuff well and we have figured out how to do fairly elaborate things with simple logic gates in hardware. So if you can create such 'easy to program' devices into cells, you can hopefully use them as an interface to the more complex machinery.

      Imagine a Beowulf cluster of E. coli ....

      --
      Faster! Faster! Faster would be better!
    8. Re:Um... by Anonymous Coward · · Score: 0

      Too late! Apple already patented it!

      God could better watch out, because he will be sued into oblivion if his creations have round corners...

    9. Re:Um... by Anonymous Coward · · Score: 0

      you're an irritating douche

    10. Re:Um... by pepty · · Score: 1
      We're already pretty good at programming life to do what we want. It's what molecular biologists have been spending their time doing for 40 years: getting bacteria to express new proteins or express/not express not express proteins in response to certain stimuli. You can buy the modular 'genetic parts' from catalogs like New England Biolabs, complete with digital or analog on/off switches that react to the presence of stimuli such as particular sugars, antibiotics, peptides, or small molecules.

      There are kits you use to insert the gene for the protein(s) you're interested in into those modules, and other kits with bacteria primed and ready to receive the modules.

      What Moon did was add more switches to the process. This really isn't useful for the making fuels or getting rid of pollutants. If you want to do that you'd use a bacterial strain that is highly optimized for the process, not one that's burdened down with general purpose or overly complicated modules. I can't really see this as useful in cancer either - there are better ways of localizing and delivering chemotherapeutics than bacteria. Similarly if you want to kill off infectious bacteria you don't need a bunch of sensors in bacteria to tell you the patient is septic; you already know that. Targeted antibiotics would be nice, but again they don't need a bacterium to do that.

      Where this type of technique will be useful is in molecular biology research - specifically where they are trying to deconvolve or reconfigure complicated pathways in real time. The rest is hype.

    11. Re:Um... by tylikcat · · Score: 1

      I think it is still up in the air how useful this will be - and not just in the research side of things, though certainly it will be used there first. It doesn't seem that unlikely to me that you want to have finer control over what is expressed when, that you want different things expressed at different times, or that you want a series of things expressed in a particular order under different circumstances.

      This as it stands it pretty rudimentary, but it could be the first step in allowing us to program cells in a much more defined and complex way than anything we can do yet.

    12. Re:Um... by Anonymous Coward · · Score: 0

      If you can make small digital circuits that you program on demand (instead of spending a lot of time in R&D labs) to find a target and the just paint it so that the immune system kills it, that would be useful I'd think. Cancer? Tell the immune system it's, I don't know, flu, and hope it will damange the cancer just enough for it to die and be recycled like normal dead cells. Gene robot finds cancer markers, emits paint, moves on. Immune system sees paint, starts chewing away. Just make sure your little robot doesn't paint the wrong targets and that it dies after a while :)

      Agile medicine anyone? :D

    13. Re:Um... by Anonymous Coward · · Score: 0

      welding a bunch of pairs of vise-grips into a shape vaguely reminiscent of a pair of pliers

      Where is Xzibit when you need him?

    14. Re:Um... by Samantha+Wright · · Score: 1

      Pfft, forget heads. Every cell does this all the time; it's how signalling cascades work. More embarrassingly, projects like this are routinely done by undergraduate students as part of the iGEM competition. To have this story publicised verges on humiliation.

      --
      Bio questions? Ask me to start a Q&A journal. Computer analogies available for most topics!
    15. Re:Um... by TapeCutter · · Score: 1

      Yep, it's like inventing a crappy uneven wheel and proclaiming it's more impressive than the rolling log it came from

      --
      And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
    16. Re:Um... by TapeCutter · · Score: 2

      This as it stands it pretty rudimentary, but it could be the first step in allowing us to program cells in a much more defined and complex way than anything we can do yet.

      It's all relative, my 1970's HS biology teacher would have viewed this as "Star Trek" science. This level of detailed understanding would have made his jaw drop in awe, as it did mine. :).

      --
      And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
    17. Re:Um... by TapeCutter · · Score: 1

      Imagine a Beowulf cluster of E. coli ....

      No imagination required, I have a tub of yoghurt in the fridge. I don't actually use the API myself, it gives my tongue a furry feeling.

      --
      And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
    18. Re:Um... by Anonymous Coward · · Score: 0

      Really? Obviously you have no inkling of what was done. This is an equivalent to nanotechnology. Granted it's not a machine, but it performs the same task. To claim it's a clumsily built equivalent, is a misunderstanding of what has been done.
      This is the first step towards freely operating in a world that we have only limited access to right now. This generation of tools will help us build the next.

  3. Inductors by Anonymous Coward · · Score: 0

    You forgot inductors. I guess it's time to coil up and fly back to wherever you came from.

  4. List of applications sounds like funding BS ... by perpenso · · Score: 1

    ... make fuel, clean up pollutants, or kill infectious bacteria or cancerous cells ...

    This list of applications sounds like generic funding BS found in grant applications. As others have pointed out this is a bit nonsensical since we already have organisms that do some of these things. We are unlikely to outperform nature's solution, the organism eating oil in the gulf of mexico for example, its hard to beat millions of years of evolution.

    That said this research could be useful. Perhaps there would be an advantage to organic circuitry. Size, performance, cleaner manufacturing process, ...? Perhaps mentioning goals such as these could help the project to be taken more seriously.

    1. Re:List of applications sounds like funding BS ... by pepty · · Score: 3, Interesting
      Actually, we are very likely to outperform nature when it comes to specialized jobs like eating crude oil in the gulf. We would start off with the organisms that are there naturally and optimize them for better metabolism in that specific environment, or hybridize them with other bacteria to make them more efficient in other environments. The trade off is the resulting microbes are crap at living under other conditions, but we don't care about that. Nature makes horses fast but evolution balances speed against fragility. People breeding horses make horses much faster - and then watch in horror as their legs break like matchsticks. When it comes to industrial bacteria, people turn mustangs into the equivalent of eight legged quarterhorses on steroids that keel over just after finishing their first race.

      I don't see Moon's research conferring much of an advantage in industrial, remediation, or medical uses; those are best off with a single control mechanism. Most of the decision making process for those situations will continue to be most efficient if it is done outside of the bacteria. For an analogy: a factory could have four sensors hooked up to four smartphones to tell it how much it has in stock of four diffeerent parts and automatically order more when the levels get too low. Or instead a factory could have a sensor and a smart phone phone attached to each and every individual part to do that job. Each part would have to carry that sensor and phone which would inflate its cost without contributing to its final use. Ditto for distributing the control architecture to the bacteria.

      I think the use for Moon's control architecture will be in research; the rest (like you said) will be hype.

    2. Re:List of applications sounds like funding BS ... by Anonymous Coward · · Score: 0

      when I build a machine I put multiple E-Stops on it and all have to be deactivated (actually they are NC so the output is on...) for the machine to run.

      Were someone to ask me to design a bacteria for release into say the Gulf of Mexico... I would want to put in multiple ways to shut it down and have those conditions be linked with nested ANDs... make the organism even MORE fragile so it didnt run away from me.

  5. oh my by kenorland · · Score: 0

    Engineer discovers biology, builds elementary genetic circuit, and thinks it's a great breakthrough!

  6. Grey goo by Msdose · · Score: 0

    So the nanotechnological "grey goo" takes its first wobbly steps.

  7. odd comments by Anonymous Coward · · Score: 0

    I find it odd that several posters seem to thing this isn't a big deal. What would you rather have, a computer that you can program, or a computer that you can only use code-snippets to paste together something close to what you want?

    1. Re:odd comments by pepty · · Score: 2
      Well if you were already used to pasting together molecular biology code snippets, you'd be impressed that someone found a way to combine that many in one system without something going wrong. You wouldn't see it as something that different from what has been going on for the past ~25 years. If you want a direct ancestor they were doing three logic gates back in 1999.

      If you were interested in using this to do something industrial or medical (not just research in a research lab) you would focus on the "something close to what you want" part and realize that you would end up going with a specialized solution that gives you what you want, not something close to it.

  8. This is really interesting. by nickol · · Score: 1

    This was really interesting. And now I am waiting .... for a first bug report.

  9. DuctTape is for leaks. WD-40 is for squeaks. by Anonymous Coward · · Score: 0

    A: Is it supposed to move?
    B: Does it move?

    If A & B = "yes" then no problem
    If A & B = "no" then no problem
    If A = "yes" & B = "no" then apply WD-40
    If A = "no" & B = "yes" then apply DuctTape

  10. I can see the advertisement now by VicVegas · · Score: 1

    --- Have a Green Thumb? You too can have the latest in organic computing! --- I will have to stick with silicon based computing, if my basil and rosemary plants are any indication.

  11. Huge Accomplishment by qbel · · Score: 2

    Very interesting article.. For those that didn't read the article, I really like how Moon emphasizes the difference between what has been done before and what he has done. What was accomplished earlier was the construction of gates, circuits and complex systems from non-living material, silicon... But what he has accomplished is the intellectual breakdown of an already living system, and the use of that knowledge to manipulate and prove that he can control it by reproducing the gates and circuits we use for modern technology. I say it while biting my cheek, but hopefully this will also lead eventually to complex, controlled biological systems. Looks like there are still hurdles to go over, but definitely bravo so far.

    Kind of raises a pretty important question, though: if and when complex systems start growing, and if AI and robots are created from them.. Will they still be just robots, or will they be living organisms? Man what I wouldn't give to hear Isaac Asimov talk about that for 5 minutes.

  12. shunkygirl by Anonymous Coward · · Score: 0

    http://www.crusher007.com
    http://www.sand-making-machine.com
    http://www.china-impact-crusher.com

  13. This guy by Mashdar · · Score: 1

    Wins the award for most meta implementation of a genetic heuristic.
    woooaaah

  14. I misread the title by Anonymous Coward · · Score: 0

    My feed reader truncated the title to "Complex Logic Circuit Made from Bac...", which I optimistically assumed would end with "on."

    I was somewhat dissappointed to discover that it actually ends with "terial Genes."

    One day, maybe, we will build computers out of delicious pieces of dead pig, but it seems that day is not today.

  15. Great by TheSkepticalOptimist · · Score: 1

    So in Little Alchemy, bacteria + bacteria = computer.

    --
    I haven't thought of anything clever to put here, but then again most of you haven't either.