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

A question about Natural Selection

  1. Mar 15, 2014 #1
    Hi, I just want some explications about the nature of "Natural Selection" in term of physical laws acting on molecules. In other words, how exactly "Natural Selection" acts on molecules and atoms.
  2. jcsd
  3. Mar 15, 2014 #2


    User Avatar
    Gold Member

    Natural selection doesn't directly act on molecules and atoms. It acts on the whole organism, often by killing them, therefore preventing them from reproducing.

    The molecules and atoms (such as DNA) are affected by mutation. Natural selection occurs when the environment acts on functional results of those mutations, reducing the reproductive capability of that DNA line, preventing it from propagating. Or, seen another way, some mutations promote reproductive success and increase the spread of that DNA line.
  4. Mar 15, 2014 #3


    User Avatar
    Science Advisor

    As pythagorean said, natural selection works at the level of populations of organisms. Basically, you have a set of individuals in an ecosystem with different traits (the traits are for the most part determined by genetics). The interaction between these traits and the environment determines the "fitness" of an organism which determines the likelihood that each individual will pass on its traits to the next generation. This process, by which traits leading to higher fitness get passed on to the next generation more frequently than traits that do not improve fitness, is natural selection.

    How do molecules and atoms play into this? The traits of an organism are defined by the molecules that make up the organism (which are, mostly, encoded in the organism's DNA). Therefore, in principle, if we look at an organism's genetic material, we should be able to predict its traits, and, if we know enough about the other individuals in its population and its environment, we can infer its fitness and how evolution will play out. Indeed, this is a major goal of research in evolutionary biology and in some limited settings, researchers have had some success.

    For example, in the study cited below, researchers are able to look at the RNA from the population of circulating influenza strains in one year then predict with fairly good accuracy the distribution of strains in the flu season the following year. This, of course, is very useful as it gives vaccine makers information on which strains they should include in the flu shots for the upcomming flu season.
    Łuksza & Lässig. 2014. A predictive fitness model for influenza. Nature, 507: 57. http://dx.doi.org/10.1038/nature13087 [Broken].

    Where do physical laws play into this? Well, in order to be able to predict the fitness of an organism from its DNA sequence, we need the following information:
    1) How do variations in an organism's DNA affect the properties of the biomolecules it encodes?
    2) How do the interactions between biomolecules determine the traits of an organism?
    3) How do the interactions between an organisms traits and its environment determine its fitness?

    The answer to #1 falls squarely in the field of biophysics, while the answer to #2 falls into the field of systems biology and #3 falls to ecology and population genetics. Of course, this simplifies the issue a lot as the interactions are a lot more complex than portrayed above, but that's the general idea. The molecules encoded by an organism define the its traits and its fitness, and the "fitness landscape," in turn, determine which molecules get passed to subsequent generations and which get weeded out of the population.

    If you're interested in reading further about the topic, here are two nice recent reviews exploring the interplay between biochemistry/biophysics and evolution:
    Liberles et al. 2012 The interface of protein structure, protein biophysics, and molecular evolution. Protein Sci, 21: 769. doi:10.1002/pro.2071 PMC3403413.

    Harms & Thornton. 2013. Evolutionary biochemistry: revealing the historical and physical causes of protein properties. Nat Rev Genet. 14: 559. http://dx.doi.org/10.1038/nrg3540 [Broken]
    Last edited by a moderator: May 6, 2017
  5. Mar 16, 2014 #4


    User Avatar
    Gold Member

    The mutations are in the DNA, the DNA codes for proteins, the proteins participate in processes that form all the tissues and molecular machines that make up organisms. Those tissues and molecular machines are the functional results. There is no decision-making required anywhere here.

    I'm not sure why you think decision making must follow from function. Or what you mean by biased action.
  6. Mar 16, 2014 #5


    User Avatar
    Science Advisor

    Ah, okay, I think I understand your question now. Essentially, you are asking how do molecules and atoms form life, right? As pythagorean said, the process by which a collection of chemicals can become a self-replicating, evolving, living system is called abiogenesis. How abiogenesis occured on Earth is not well understood, and although there are many theories about what steps may have been involved and how those steps could have happened, there is no general conensus around the issue yet. Here is a nice, simplified description of what could have happened: http://evolution.berkeley.edu/evosite/evo101/IIE2bDetailsoforigin.shtml [Broken]

    Now, in my previous post, when I said that natural selection operated on the level of population of organisms, I was discussing that in the context of modern day natural selection. In the early prebiotic Earth, before life had emerged, it is likely that natural selection did act on molecules. The basic rerequisites for natural selection to occur are:
    1) A self-replicating system that,
    2) has traits that can be passed on to its descendents, and
    3) lives in an environment where certain traits allow some individuals to reproduce more than others

    So, in order for a molecule to undergo natural selection, it must be able to replicate itself and its ability to replicate must be heritable. Researchers have indeed been able to engineer molecules with these properties and, when placed in the right environment, they undergo natural selection in the lab:
    Joyce and Lincoln sought to evolve their molecule by natural selection. They did this by mutating sequences of the RNA building blocks, so that 288 possible ribozymes could be built by mixing and matching different pairs of shorter RNAs.

    What came out bore an eerie resemblance to Darwin's theory of natural selection: a few sequences proved winners, most losers. The victors emerged because they could replicate fastest while surrounded by competition, Joyce says.​
    See also Lincoln and Joyce. 2009. Self-Sustained Replication of an RNA Enzyme. Science 323: 1229. doi:10.1126/science.1167856.
    For more on efforts to build living systems in the laboratory see http://www.nytimes.com/2011/07/28/science/28life.html

    Now, are these self-replicating RNA molecules alive? That very much depends on how one defines life (usually, the ability to metabolize – to harvest energy and building blocks from the environment – is considered a halmark of living systems and these RNA enzymes lack this ability). But this brings up a broader question: what makes "living" molecules different from "non-living" molecules.

    Here, pythagorean again has stated the answer. Life is a property that comes about not from the molecules themselves, but from the interactions between the molecules – it is an emergent property of a collection of molecules. If you are mathematically inclined, here's one way to think about it. Consider a system of coupled non-linear differential equations. These differential equations might describe different interactions between molecules in a living system. Now, depending on the parameters of that set of differential equations, the system will behave differently. For some sets of parameters, the system will tend toward attractor points whereas for other parameters, the system will diverge away from repellor points. The system might enter a limit cycle and display oscillatory behavior for some parameter, while other parameters will lead to a chaotic system. These behaviors all depend on the initial values of the parameters of the system, the point in phase space at which these systems begin. Life, in this somewhat simplified view, is the region of phase space that allows the system to exist as a self-replicating system that can extract energy and building blocks from its surroundings. All you need to do is to find the right collection of molecules (i.e. the right set of differential equations) and set them up at the correct point in phase space, and you can create a living system.
    Last edited by a moderator: May 6, 2017
  7. Mar 16, 2014 #6


    User Avatar

    Staff: Mentor

    Yes, and the reason why was answered already: the selection happens by the mutation either making the organism better or worse at reproducing.
    There is only one set of laws of the universe and everything has to follow them.
    Correct: natural selection is what makes organisms evolve - so it isn't what created the first organism.
  8. Mar 16, 2014 #7
    The problem is the correct initial values is almost impossible in a stochastic point of view, and even if we assume that the correct first spark exists, and that a good soup of amino acids also exists at the good place at the good time, it would be so unstable, that a tiny tiny perturbation, a simple photon can destroy the whole machinery of the forming cell, by knocking off an electron and collapses everything. The need of an intervention of natural selection at this stage is a must to guide this sensitive unstable process. But here again, the nature of natural selection strikes me, I see it as a ghost effect that no one can explain it to me. I see it as a huge unanswered question, it does not make sense for me. I am working on the field of Evolution Equations and Random Processes, that's why I am interested by this phenomena.

    I read the other day about a theory that makes a link between physical laws and natural selection, it is called "Cosmological natural selection" by Lee Smolin. But it was so inconsistent, and seemed only to try to interpolate natural selection in a cosmological scale.
    Last edited by a moderator: May 6, 2017
  9. Mar 16, 2014 #8
    This is what I call biased action, why selecting the mutations that help survival over other mutations ?
  10. Mar 16, 2014 #9
    populations of organisms who do not adapt to their environment, they won't survive. So where's the question natural selection if they die of.

    You may call it biased, is it biased for organisms to live and adapt to the environment?. Accordingly mutations that give a slight advantage for populations to survive and adapt will be called beneficial(even if you disagree).
  11. Mar 16, 2014 #10


    User Avatar
    Gold Member

    Why do you believe it would be so unstable?

    What do you mean by intervention?
  12. Mar 17, 2014 #11


    User Avatar
    Science Advisor
    Education Advisor

    I think you're right.

    In fact if you're talking about abiogenesis, it's quite possible that many of the first complex molecules that came together with the potential to self-replicate were destroyed before they ever did. A self-sustaining chain may have been highly improbable. But the thing is, it only had to happen once. And current thinking is that it happened in stages:
    (a) prebiotic synthesis of nucleotides
    (b) prebiotic synthesis of polynucleotides
    (c) emergence of RNA molecules capable of catalyzing their own replication
    (d) evolution of the replicating RNA favouring efficiency
    (e) emergence of other catalytic RNA molecules
    At some point all of this gets enveloped in a lipid envelope allowing for some insulation from the outside world and from what I understand there is a kind of chicken or egg debate on when this happened.

    I'm not sure what you mean here. Natural selection at the molecular level is really just a case of continuing to self-replicate or not. If a mutation comes along that makes the process more efficient, then the more efficient self-replicators are more likely to do so and so you end up with more of them.
  13. Mar 17, 2014 #12

    That's good, I think I get this step. I understood that it is a gradual random cumulative process, RNA molecules which had bad mutations collapses, and other which had mutations that do not stop the replication power continue to replicate. But I see a lot of obstacles here:

    1-The need of a huge soup of amino acids in a very short time, a time shorter then the disintegration of the RNA molecule. Kind of giving food (nucleic acids) to RNAs quickly before they lose the replication power by nasty mutations.

    2-A lot of mutations will destroy a lot of RNAs or stop their replication power. That needs the healthy RNA to stay in the soup.

    3-Assuming the conditions 1 and 2 remains, lot of RNAs with different mutations survived and replicating well, after time this will create a random cumulative tree, but how the complex molecular machines appeared from this random cumulative tree.

    4- Is there any computer simulation that shows this formation over time, like we do in physics to simulate evolution processes over time ?

    5- The probabilistic problem, in the last points, I made huge generous assumption which has very low probability existence, the product of very low probabilities will give a far low probability.

    6-The machine-machine dependance, machines need other machines to do their job. Questions like how the replication passed from simple nucleic acids binding to other nucleic acids in an RNA molecules to something sophisticated like this:


    Is there again a computer simulation to show this process of evolution ?
    This is so sensitive and complex that even scientists can't construct it in labs in purpose, let alone by chance. I see that the more time goes, the more complex the cells appears.

    7- The probabilistic problem of a replicating RNA molecule to suddenly appear:

    8- How the DNA repair system that fights random mutations appeared ?

    9-The information stored in DNA, how it was made ?
    Last edited: Mar 17, 2014
  14. Mar 17, 2014 #13


    User Avatar

    Staff: Mentor

    Sorry, too many questions and too broad. You need to post a link to a study or article, then ask a specific question about what you do not understand. Posting what you read that you don't understand helps us to understand if you have simply misunderstood what you read or if you are needing a simpler explanation. One or two questions per post. You must post the specific paragraph(s) in the paper, do not expect us to read the entire article. We are not here to debate or teach evolution. Abiogenesis is not evolution.

    I have deleted several of your off topic posts on abiogenesis, that has nothing to do with this thread. If you wish to discuss abiogenesis, please start a new thread following the guidelines I have posted above.

    Members, please do not respond to this member unless their posts meet our guidelines to prevent the thread from going off topic.
    Last edited: Mar 18, 2014
  15. Mar 18, 2014 #14


    User Avatar
    Science Advisor

    Most mutations occur during replication from errors in the replication process. Thus, rates of replication and mutation are, in general, coupled. The important parameter here is thus the fidelity of replication (i.e. the error rate).

    Although we know that such molecular machines came about because of evolution, we currently don't understand the exact mechanisms by which such machines evolved. Here is one example of a recent study addressing one aspect of this question: Finnigan et al. 2012. Evolution of increased complexity in a molecular machine. Nature 481: 360. http://dx.doi.org/10.1038/nature10724 [Broken]
    Last edited by a moderator: May 6, 2017
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Similar Discussions: A question about Natural Selection