Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Simplest self-replicating molecule?

  1. Oct 18, 2012 #1
    Can you help me identify simplest naturally occurring self-replicating molecule?

    Or even better, let's make one artificially... where "self-replicating molecule" is such that contains at least two different atoms, it reacts with other atoms and molecules so that it accumulates duplicates of those atoms it is composed of, and then at some point reaches "critical mass" (length, or something) where it splits into two identical molecules.

    This question was inspired by http://organicbuilder.sourceforge.net/
    - "This freeware demonstrates an artificial chemistry. Each circle is an atom, that floats around and can become bonded to others to form a molecule... Perhaps the coolest thing you can do is template replication - the process DNA uses to make copies of itself. If you choose the right rules you can actually get this working - follow through the challenges to find out how."

    I find that little simulation is amazing, and pretty cool as a game too! Surely then we should be able to do it with real atoms as well, right? What software would be best for exploring solutions to this puzzle?
  2. jcsd
  3. Oct 19, 2012 #2

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

  4. Oct 19, 2012 #3
    Ok, thank you. Unfortunately those paper seem all to be talking just about complex molecules with thousands of atoms.

    I was hoping for solution with like 3 to 5 atoms molecule. Something like this imaginary example:

    "H2C302" + 2H20 + 2C2 -> H4C6O4 -split-> 2x "H2C3O2"

    We can surround our molecule with whatever other atoms and molecules we want, perhaps some of them to act as catalyst, and we can chose any arbitrary conditions we want, like pressure/temperature. My goal is not replicate what happens in reality, but to understand the underlying necessary principle for this to occur in minimalistic terms, so it only needs to be theoretically possible, and the simpler the better.
  5. Oct 19, 2012 #4

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    Part of the presentation shows a template for replicating, and some of the molecules are of the order of a score or so atoms. Hexadeoxynucleotide looks like 23 atoms. The thing at slide 24 looks like 24 atoms... could be miscounting.

    But you are right - they do like the massive molecules - I pointed you at it as a starting point for the underlying theory.

    Define "self replicating".
  6. Oct 19, 2012 #5
    Great, that's what I was looking for, just missed it the first time. I was looking at all the other papers by reading the text under the presentation window, which lacks formatting so it's a bit unreadable.

    Basically "self replicating" I call any molecule that accumulates atoms it is made of, and then it splits in two, starts all over again. In principle it's a virus, it should continue to make copies of itself until there is no more molecules that can take necessary atoms from to reproduce itself. So naturally it's better the more stable molecule is in relation to environment and "food" molecules it reacts with, yet still reactive enough so it can "eat" or "assimilate" other molecules.

    I guess the simplest polymer molecule could be a good template for the start.
  7. Oct 19, 2012 #6
    polymers are mostly inert.

    you're basically asking for a self catalyzing molecule. Except you're not. If the molecule could form in the first place it would not split since it must be thermodynamically stable. If it was not, it would not form, unless you input energy. And that's something catalysis just can't do.
  8. Oct 19, 2012 #7
    Awww. Obviously this is far out from my understanding of chemistry.

    Yes, I think that's good description. My point is, it's possible. We have DNA and RNA to prove it, and there is this "hexadeoxynucleotide" with only 23 atoms they say it's self-replicating too. The question is only what is the minimum number of atoms to construct something like that, can we make it replicate with less than 23 atoms? By making very nice environment for it to live in, for example, or something, whatever trick, whatever works.

    So yes, I agree with you that one of the main problems is to figure out when and how the molecule should split considering it has to be pretty stable in the same time. I have no idea. How DNA does it? I expect there must be some "critical mass" or "critical length", or "critical number of something" with certain molecules in certain configurations so that kind of stuff happens... but I wouldn't know where to even start, so figuring out those type of things is exactly what I need help with. I think this is very nice puzzle.
  9. Oct 19, 2012 #8

    Simon Bridge

    User Avatar
    Science Advisor
    Homework Helper
    Gold Member
    2016 Award

    The presence of the molecule catalyses the production of more of the molecule ... the slide presentation I linked to has the more usual definition for a self-replicating molecule.

    I'm pretty sure some mathematician will have worked on the question of the smallest, in principle, self replicator ... it seems like what they do, but I am not familiar enough with the lit to do the search.

    Any number on self-replicators in cellular automata.
  10. Oct 20, 2012 #9
    If catalyst molecule is part of our self-replicating molecule, we could call it "mouth" or "stomach".

    I've googled a lot and your link is by far the best. I am yet to start reading it though.

    Just recently some guy discovered, or constructed, first self-replicating entity in the game of life, I think, but I'm not really sure what software they used or what they actually did. In any case it was apparently a big deal.

    The guy who made that artificial chemistry simulation is linked from Wikipedia and has published quite a bit on the subject - cellular automata mixed with chemistry, in search for underlying principle behind mechanics of self-replicating molecules, that is life. It's exactly what I am interested in, that is to simulate abiogenesis and hopefully observe evolution in such virtual environment. But unlike his simulation mine would be based on real chemistry, still approximation though as my real goal is to simplify it as much as possible so I can simulate bigger world, and in shorter time. Obviously I have lots to learn first, so any help and info is welcome.
  11. Oct 20, 2012 #10
    I'm not *entirely* sure what you're after here, but one thing to consider is the issue of scale. If, for example, you have a strand of nucleic acid template, and there are a bunch of nucleotides floating around, getting that existing templating strand to autocatalyze the formation of another strand of template is far less challenging than trying to synthesize the nucleotides from - say - carbon dioxide, dinitrogen, and such. With the strand of nucleic acid template, there is a means of exerting some geometric/structural control, as any daughter product should interact favorably with the parent strand at multiple points of contact. There are also thermodynamic considerations - for example, the dinitrogen triple bond is a notoriously robust one.

    P.S. - Most chemists will consider anything with ~ 25 atoms to be a "small" molecule.
  12. Oct 20, 2012 #11
    Can they "feed" from wide variety of molecules or just from some narrow group of compounds?

    What you just said is exactly kind of thing I'm after. Thank you.

    So many things, so little brain. How in the world am I gonna learn all that stuff, uh. Can you explain how these strands can be reactive and yet so stable? Or what is it that makes catalyst molecule be able to react and yet stay complete itself?

    It must be about bonding/breaking energies, so if we take some catalyst and list of all the reactions and sort them according to energy there should be some pattern in that sequence that if we can replicate with some other molecules of similar bonding/breaking energies the interaction should play about the same. If you know what I mean? And if that was true that would make sense to me and I would know how to get on from there without even knowing much of anything else, or so I hallucinate.
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: Simplest self-replicating molecule?
  1. DNA Replication (Replies: 2)

  2. He2 molecule (Replies: 18)

  3. Polarity of a molecule (Replies: 3)