Friday, November 27, 2009

De Novo Genes: What are the Chances?

We have been discussing the de novo gene T-urf13 found in the mitochondrial genome of certain varieties of corn. Readers have asked about the evolutionary claim that the gene arose via blind evolution, and in particular about the role of mutations. Let's have a look.

First, the time frame over which T-urf13 arose is too short for mutations to play a significant role. The evolutionary explanation is that existing, non proteins coding, DNA sequences in the corn mitochondrial genome provided the raw material for the new protein-coding gene, T-urf13.

Under evolution non protein coding DNA sequences are not supposed to carry a pre-planned protein coding information layer. Such information layers are common. For instance, in a communication system it is possible to transmit multiple messages simultaneously. Several telephone conversations or Internet users could, for example, share a single wire. In other words, there can be multiple messages superimposed on a signal.

And just as multiple messages can be transmitted in a wire, so too there can be multiple messages, or layers of information, in a DNA sequence. Of course evolution expected no such level of cleverness. Hence the surprise when overlapping genes were discovered in DNA. In recent years DNA has been found to contain several layers of information.

Could it be that there is another such layer of information that the cell uses to create new genes? Apparently so, for we now find de novo genes such as T-urf13. But because evolution dogmatically rejects any possibility of design, it does not allow non protein coding DNA sequences to carry a preplanned protein coding information layer.

Instead, such protein coding information must exist only by random chance. And when chance rearrangements occur, and a complete protein sequence just happens to form, then de novo genes can appear. A sophisticated protein such as URF13 may appear to be designed, but such an appearance must be deceptive. Such events must occur by chance, not by design.

And what are the chances of this occurring? The corn mitochondrial genome is about half a million base pairs in length. That is enough room to contain about 2,000 T-urf13 sized genes. This number can be increased by accounting for DNA's six different reading frames, and decreased by accounting for the fact that only part of the corn mitochondrial genome is available. It can also be increased by allowing for some overlap in the hidden genes.

Let's be conservative and say there is room for 100,000 such genes in the corn mitochondrial genome. Nonetheless this is seven orders of magnitude shy of the million million sequences needed for any hope of obtaining a functioning gene, as found in one experiment. And even that estimate was conservative because only the minor function of ATP binding was required.

In contrast, the URF13 protein is a far more complex, cleverly designed machine. It is designed such that several copies fit together to form a protein machine. And that machine fits into the inner mitochondria membrane, a very complex environment. And the machine provides a channel that allows only certain types of chemicals to pass through the membrane. And did I mention the channel is gated, with a molecular switch to open the gate?

The URF13 protein design dwarfs the experimentally screened function of ATP binding. And yet, even in that simple case, and with conservative assumptions, we find the probabilities of the T-urf13 de novo gene arising via blind evolution to be one in ten million (that is, 1 in 10,000,000). The real number undoubtedly has many more zeroes.

This is the story of evolutionary probabilities. Over and over we come up with so many zeros. In design after design, and species after species, evolution repeatedly draws upon unlikely events to create its marvels. Evolutionists now contemplate a multiverse--a super universe containing an untold number of unseen universes toiling away through the ages--in order to beat the odds. Sure these designs would never appear if there was just one universe, but what if there was a near infinity of universes? Surely one of them would get lucky.

Evolutionists do not worry that their story is unlikely. They insist de novo genes must, one way or another, be simply the result of blind processes. This is a good example of the absurdity of evolution. Like a robot that hits a wall but just keeps on trying, evolution cannot adjust to the scientific data. This is because evolution is not, at bottom, a scientific theory. It is driven by the metaphysical mandate that matter and motion must explain everything, regardless of the evidence. This mandate has arisen from a long history of thought in science, philosophy and theology. Religion drives science and it matters.

39 comments:

  1. The corn mitochondrial genome is about half a million base pairs in length.

    Why are you limiting this to a single corn mitochondrion? Every cell in a corn plant has multiple mitochondria, every corn plant has millions if not billions of cells, and there are more than a million million individual corn plants grown every year. Every time a corn cell divides (and its mitochondria divide along with it), we get the same chance for random mutation.

    Instead of thinking that "the" corn mitochondrion has "half a million base pairs", realize that each corn mitochondria has half a million base pairs.

    This is the story of evolutionary probabilities. Over and over we come up with so many zeros.

    Exactly! You understand now! If you look at how much stuff is out there, how many (in this instance) corn mitochondria actually exist in the real world, you see that your number, which seems so large from your armchair (a million million! Wow, that's a lot next to the number of people I can shake hands with) is actually minuscule next to the fecundity of the the biosphere.

    Your problem, if I may be so bold, is that you think you're dealing with a platonic ideal -- "The Corn Mitochondrion" -- that has a fixed and eternal structure, that is copied exactly from one cell to another, from one plant to another. This isn't how it works, and it seems to me like making a straw man.

    [Evolution] is driven by the metaphysical mandate that matter and motion must explain everything, regardless of the evidence.

    Well, we have evidence that matter and motion exist, so an explanation that involves only those things is preferable to one that invents other stuff willy-nilly.

    One minor problem with this line of argument; you've only eliminated one hypothesized materialistic mechanism for producing the novel protein. I think I've cleared up your misunderstanding and shown that your objection is wrong, but let me, for the sake of argument, grant your point. Let's say that random mutation and natural selection can't explain this novel protein. Why is your magical solution preferable to another materialistic one we haven't come up with?

    Wait, you aren't arguing for a God of the Gaps, are you? Just because we don't have the answer right now doesn't mean we get to make something up.

    Duke

    ReplyDelete
  2. Duke:

    "Why are you limiting this to a single corn mitochondrion? Every cell in a corn plant has multiple mitochondria, every corn plant has millions if not billions of cells, and there are more than a million million individual corn plants grown every year. Every time a corn cell divides (and its mitochondria divide along with it), we get the same chance for random mutation.

    Instead of thinking that "the" corn mitochondrion has "half a million base pairs", realize that each corn mitochondria has half a million base pairs."

    This sounds like evolutionary logic. You can have as many copies of the mitochondria as you like in that line of corn, they share the same genome. If you don't find your random protein coding sequences, then good luck on the next billion that you look at.

    "Just because we don't have the answer right now doesn't mean we get to make something up."

    Unbelievable! Evolutionists literally make things up all the time, and then turn around and blame the skeptics.

    ReplyDelete
  3. So, Cornelius, are you claiming that the maize mitochondrial genome is completely static and inert, that no recombination or re-arrangements can possibly occur?

    ReplyDelete
  4. You can have as many copies of the mitochondria as you like in that line of corn, they share the same genome.

    Um... Reference? You're saying that mitochondria copy themselves perfectly? With no mutation at all? You'll have to show your work on that, I'm afraid.

    If that's not what you're saying, and if the mitochondria do have random mutation (just like every other reproducing creature), then they don't share the same genome. Until you show otherwise, I will believe they share a similar genome, with room for variation.

    Just to give an idea of the scale of numbers we're talking about here. Let' say that every corn mitochondrion has the minimum number of genomes, 2 (although that number will probably vary, since having more or fewer genomes is a common genetic abnormality, even in humans), and let's say that every corn cell has 100 mitochondria in it (Google "How many mitochondria in a plant cell" and take the first result), which is the low estimate for plants. Let's also assume that each corn plant has a millions cells in it, which is ridiculously low, but I'm making a point here. That means we have

    2 genomes/mitochon. * 100 Mitochon/cell * 1e6 cell/plant

    That means 2e8 mitochondrial genomes per plant, and that's a ridiculously low estimate. Even so, that takes you to one five-thousandth of your legendary "million million" number.

    But wait! There are 4e12 corn plants grown every year. That means there are 8e20 (that's 800,000,000,000,000,000,000) separate mitochondria, every year, each one ready to randomly stumble upon a simple protein in that tiny 1e12 haystack.

    Unbelievable! Evolutionists literally make things up all the time, and then turn around and blame the skeptics.

    I haven't made anything up*. Perhaps you could engage with my argument, rather than the stereotypical "Evolutionist" you seem to have in your head. Seriously, you seem to think you know more about this than I do, so please, enlighten me. What have I missed? I've drawn data from publicly available sources and used high school math to show that your conception of numbers is off; "a million million", which seems so large that you keep invoking against evolution, is actually quite small. If you wish to challenge my numbers or my math, please do.

    Really, there's only two ways to save your argument. First, show that there is absolutely no change in the mitrochondrial DNA as they reproduce. This would make them effectively a platonic ideal, and if you could do that, I'd withdraw my argument and admit that this novel protein just couldn't evolve. The fact you don't understand that mitochondrial DNA -- all DNA -- mutates every time it reproduces might be why you don't accept evolution. We may have made real progress here.

    The second way is for you to find some egregious error in my numbers or my math, something that would put me off almost fifty orders of magnitude in the wrong direction. I don't think I've made such a mistake, but I've made mistakes in the past so I won't rule it out.

    But also, answer my question. Even if you manage to prove that this particular protein couldn't have evolved (despite the vast, vast number of chances it's had over the past 40 years), what conclusion will you draw from that? How are you going to go from ignorance (of the origin of this protein) to your desired conclusion?

    Duke

    * -- With the exception of the number of cells per corn plant, which I just couldn't find, but I think I picked something at least a thousand times smaller than the real value.

    ReplyDelete
  5. AG:

    "So, Cornelius, are you claiming that the maize mitochondrial genome is completely static and inert, that no recombination or re-arrangements can possibly occur?"

    No, I'm not saying that. In fact, my point is that it is, indeed, these mechanisms, more than mutation, that one might appeal to in order to explain how the de novo gene might have arisen.

    But to do that you need preexisting segments of DNA that constitute chunks of protein coding sequence, such that when combined they magically for a coding sequence. IOW, you need almost complete protein sequences that are split up into a few parts. Two parts in the case of T-urf13.

    ReplyDelete
  6. Oh, "magically." I see.

    ReplyDelete
  7. Duke:

    You are not explaining how you get your new proteins (from the looks of your numbers the corn mitochondria ought to be producing every possible protein there is). But you haven't explained the steps. What time span are you assuming for your mutations to accumulate?

    The problem is you are trying to explain how this de novo gene arose using an unlikely mechanism--mutations. You don't have a very big time window for starters. Also you would have to accumulate a great many mutations before you get a useful protein to be selected. And finally, the vast majority of the de novo gene we're talking about has high similarity to two existing segments in the genome. You're saying that unguided mutations just happened to create a new gene that mimics two existing sequences?

    Perhaps I'm misunderstanding what you are saying...

    ReplyDelete
  8. What time span are you assuming for your mutations to accumulate?

    I guess I wasn't clear enough in my explanations. All those numbers were per year. We have roughly 8e20 chances each year to stumble upon any particular protein that has some partial functionality. As for...

    [H]ow you get your new proteins

    I don't have to explain that. You provided that number, and I accepted it; every "million million" random combinations of base pairs will have some functionality. Once we hit upon something that has some functionality natural selection can work to improve the function (remember, you asked where the protein came from, not how it got better).

    The problem is you are trying to explain how this de novo gene arose using an unlikely mechanism--mutations. You don't have a very big time window for starters.

    Actually, I think I've shown that finding some new protein is actually quite likely, and the time window is actually forty times bigger than what I've assumed. (Honestly, we'll need several generations of corn to refine the protein, but we have that window of time).

    And finally, the vast majority of the de novo gene we're talking about has high similarity to two existing segments in the genome. You're saying that unguided mutations just happened to create a new gene that mimics two existing sequences?

    Um... Yes? This sort of thing is common and observed. Part of the genome is copied and added onto the genome twice. It's like some unguided copy-editor added two versions of Chapter Four to a book.

    The reason this is important to evolution is that genome now has two copies of the genes in question; that means one can retain its activity(and keep the organism or organelle alive) and the other can mutate at random (to produce new functionality). This has been observed both in the lab and in the field; the fact that the T-urf 13 gene is from that sort of change makes it much more likely to occur. The sort of accumulate-point-mutations-until-we-get-new-genes stuff isn't really the sort of de novo thing we see very much (since recombination and copying errors are much quicker and more effective). I only based my assumptions on it because you seemed most interested in it.

    [F]rom the looks of your numbers the corn mitochondria ought to be producing every possible protein there is

    But what of that? Most new proteins, if they were made, wouldn't have much of an effect on the activity of the mitochondria they were in; if anything, they'd be a slight detriment, since the organelle would waste resources producing things that didn't help it perform its fuction and might cause an immune response from the plant.

    The important thing is that we've now seen there is no problem finding new proteins; having too many new proteins is more of a problem than having too few. Since you haven't mentioned it, I assume we're OK with natural selection refining the functionality of any proteins that happened to occur.

    Duke

    ReplyDelete
  9. I also think we've been a bit one-sided in this conversation. Let's look at intelligent design for a moment. According to ID, the situation is as follows:

    1) The Designer creates teosinte, the wild precursor to maize, which looks completely different.

    2) The Designer change teosinte into maize right when the Native Americans are ready to harvest it.

    3) The Designer creates the fungus to prey on the maize, threatening the lives of everyone who relies on the maize for food.

    4) The Designer creates the T-urf 13 protein to protect his creation (maize) from his other creation (the fungus).

    Now, each of these steps requires the intelligent designer's handiwork if step 4) does; you can't say that the fungus is the result of random mutation if you maintain that T-urf 13 isn't as well. If you want to explain your theory, you have to tell us:

    1) How the designer detected the need for the protein.

    2) How the designer actually designed the protein.

    3) How the designer inserted the protein into the corn's mitochondria.

    4) Why the designer only inserted the protein into a few strains of corn.

    5) Why the designer went through the charade of creating fungus when it was only going to be stopped by the protein.

    Do you see why the rest of us think that the world looks much more like the random mutation that comes from rampant fecundity? The fungus happened to hit upon a good trick for preying on corn, stealing the energy it got from the sun to make more fungus. The corn hit upon its own good trick, driving off the fungus to continue making more corn rather than fungus. If this is the result of an intelligent designer, then the designer is at best inscrutable and at worst mad or obsessed with blood sport.

    As a possible solution, consider multiple designer theory. In this formulation, there is one designer responsible for maximizing the amount of fungus in the world, and one for maximizing the amount of corn. This would at least answer question 5) (although it wouldn't touch any of the others).

    And could you at least make a stab at discussing questions 1) through 4)?

    Duke

    ReplyDelete
  10. Duke - I think your questions are very good and appropriate. They echo my own suspicions about ID. Much is made of 'design' but very little can be said about how, why, and when. I don't think IDers even have many ideas of how they would even go about answering those questions.

    It makes me wonder that if there is a designer, that entity has done their very best to cover their tracks. Some would say there is a 'signature', but that of course is not at all a settled question (and quite likely is just a premature interpretation based on insufficient data); what we can see is that this designer has left little to no clues about the methods used and when they were used.

    Assuming that this designer is the God of the Bible (as I believe the vast majority of the Discovery Institute believe, including Cornelius) - you can't help wonder why the stories in the Bible seem then so ill-matched between what is actually observed in the natural world.

    ReplyDelete
  11. Duke:

    "I guess I wasn't clear enough in my explanations. All those numbers were per year. We have roughly 8e20 chances each year to stumble upon any particular protein that has some partial functionality."

    OK, so you are finding that a corn mitochondria generates roughly a new protein per year. And since you have a great many mitochondria, you therefore have a veritable factory for new proteins.

    The problem here is that theology is driving science. Evolutionists believe evolution must be true, there is no choice about that. So they must make the numbers work out, somehow.

    Each mitochondria generating new proteins every year is not something that is empirically observed, and of course it would impose an untenable mutational load, to put it mildly.

    As scientists, we have a duty to be faithful to science, not manipulate it for pre conceived ends.

    ReplyDelete
  12. Duke:

    "Now, each of these steps requires the intelligent designer's handiwork [...]

    Do you see why the rest of us think that the world looks much more like the random mutation that comes from rampant fecundity?"

    Yes, I certainly do. I did not intend to minimize these arguments--they are powerful. I will address these issues in an upcoming post.

    ReplyDelete
  13. OK, so you are finding that a corn mitochondria generates roughly a new protein per year.

    Not at all; a corn mitochondrion doesn't generate a new protein every year. Each mitochondrion has some random variation from its parent; that random variation may or may not enable it to produce a novel protein.

    And since you have a great many mitochondria, you therefore have a veritable factory for new proteins.

    Yes. Some of these novel proteins would be an advantage to either the mitochondria they're in, or to the corn plant itself, and there would be more of the mitochondria with that novel protein in the next generation.

    The problem here is that theology is driving science. Evolutionists believe evolution must be true, there is no choice about that. So they must make the numbers work out, somehow.

    Really? Please, just show me where I'm wrong, so I can see your point of view. Do mitochondria not vary during reproduction? Have I made some error with the math? Instead of accusing me of doing theology, show me where I'm doing science wrong. Show me my mistake, if you think I've made one.

    You say I'm doing theology. All I'm doing is saying that this situation doesn't justify hypothesizing an invisible, intangible intelligence that even its supporters (that is, you) can't make coherent statements about. If I said that Newton's Laws of Motion said that we no longer need to hypothesize drunken angels pushing the planets across the sky, would you accuse me of doing theology? That's the exact situation you've put yourself in, trying to find anthropomorphic angels in the equations of force and gravity. I'm not saying (in this instance) that god does or doesn't exist. I'm saying that this novel protein in mitochondria doesn't require a designer to explain it. You think that this particular novel protein does require a supernatural designer? Show me where the error in my thinking is; don't wave your hands, point your finger, and accuse me of practicing theology. Show me where I've inserted theology, at which step I've changed the facts of the universe to come up with my conclusion. Theologians accuse; scientists find error in argument. You want to claim the title of scientist and skeptic? Step up and show where my reasoning is wrong.

    Duke

    ReplyDelete
  14. Isn't this new protein bad for the corn? Doesn't it lead to sterility and vulnerability to a virus? So how can this mechanism be the one that leads to a new species? I'm curious.

    ReplyDelete
  15. Duke:

    "Really? Please, just show me where I'm wrong, so I can see your point of view."

    I thought you were saying you each mitochondria generates something like a new protein / year. If you're not saying that, then what are you saying? You haven't spelled out the details so I'm left trying to figure out what you are thinking.

    "a corn mitochondrion doesn't generate a new protein every year. Each mitochondrion has some random variation from its parent; that random variation may or may not enable it to produce a novel protein."

    OK, so sometimes the mutations generate new proteins. You'll still need more details. At about what frequency does this occur? Do we have any reason to think so many mutations can occur in so short a time? What you're saying seems to make no sense, but perhaps I just don't have all the details yet.

    ReplyDelete
  16. Cornelius, maybe you could support your claims about mutational loads as the concept pertains to plant mitochondria. I would be interested in specific studies and data that are consistent with your (otherwise unsupported) assertions.

    And please, try to point to studies of plant mitochondria. A general blanket reference to nuclear genes won't cut the mustard here.

    (We'll get to the unsupported claims that are behind your ideas about protein evolution later.)

    ReplyDelete
  17. Cornelius Hunter:

    I thought you were saying you each mitochondria generates something like a new protein / year. If you're not saying that, then what are you saying? You haven't spelled out the details so I'm left trying to figure out what you are thinking.

    Ah, yes. This is a common problem with me. I have such problems communicating what I think to others; I sometimes suspect I must as Asperger's Syndrome or something. Let me try a different tack.

    You've said that it takes "a million million" random arrangements to find a functional protein, right?

    What I'm trying to show is that that number, a million million, 1e12, is actually quite small next to the number chances to find it. That's what I was getting at with the whole 4e20 corn plants thing; that is far, far larger than 1e12. If you look at the number of corn plants that are created every year, you see that there are well over 4e8 times more corn genomes than your 1e12 number. And this is for the nuclear corn DNA, not for corn mitochondrial DNA or chloroplastic DNA.

    OK, so sometimes the mutations generate new proteins. You'll still need more details. At about what frequency does this occur?

    I'm terribly sorry, but I'm going to have to turn this back on you. Where did you get that "million million" number? I'm relying on you for frequency of useful proteins from random searches; that's the one thing I didn't look up. What's your reference?

    Do we have any reason to think so many mutations can occur in so short a time?

    Yes, we do. Every time DNA gets copied, errors are made. This is simple, brute fact. That's what we were getting at with that whole "Platonic ideal" thing.

    What you're saying seems to make no sense, but perhaps I just don't have all the details yet.

    I think the fact this doesn't seem to make sense is my fault for adding too many details. Let my try to simplify it.

    You said that you have to search 1e12 random arrangements to find a functional protein.

    I showed that there are far, far more random arrangements occurring in nature, since each corn plant, each mitochondria and each chloroplast is, in part, a random arrangement.

    To me, this answers your question about where the new protein comes from. You set a limit as how likely finding any protein with any function is (a million million) and I showed that there are far more chances to find a protein that that.

    Is that clearer?

    Duke

    ReplyDelete
  18. Isn't this new protein bad for the corn? Doesn't it lead to sterility and vulnerability to a virus? So how can this mechanism be the one that leads to a new species? I'm curious.


    Whether or not the protein is good or bad for corn is irrelevant; it's a new protein that has a new function, and it appears to have come about through unplanned, undirected, "natural" causes. Evolutionary theory requires this to happen at least occasionally, and if it did happen, it would invalidate intelligent design "theory".

    It seems to me that, at this point, mentioning forming a new species is a bit of a red herring; no one in this discussion is saying there was a speciation event, and no one in the more general community says that new proteins are sufficient or even necessary to make a speciation event (if I recall correctly, we don't have any new proteins when compared to chimpanzees, although I might be wrong on that).

    It's like coming in on two auto mechanics discussing a carburetor, and saying "You can't get to Philly, the bridge is washed out." We're discussing two completely different things that are only tangentially related.

    At least it seems that way to me. I could be wrong, or expressing myself unclearly.

    Duke

    ReplyDelete
  19. AG:

    "maybe you could support your claims about mutational loads as the concept pertains to plant mitochondria. I would be interested in specific studies and data that are consistent with your (otherwise unsupported) assertions."

    Hmm, I didn't think I would need justification that ~10^-3 mutations/site/year would be a big mutational load on the corn mitochondria genome. What new explanation do evolutionists have for this?

    ReplyDelete
  20. Duke:

    ID people, creationists, etc. have conceeded that microevolution does take place. Trivial changes in extant organisms happen. The point of controversy is whether this mechanism can explain how bacteria can turn into blue whales. That is an evolutionists claim, but they haven't demonstrated that it is possible.

    ReplyDelete
  21. The point of controversy is whether this mechanism can explain how bacteria can turn into blue whales. That is an evolutionists claim, but they haven't demonstrated that it is possible.

    And I feel that they have demonstrated it is possible, because I've looked at the evidence, whereas I suspect that you have not.

    What about the proposed evolution of the whale do you find untenable?

    Please, be specific. Where does the scientific consensus break down for you?

    Duke

    ReplyDelete
  22. Duke:

    The appearence of this new protein will not allow the corn to become a new species. It is more likely to make the corn become extinct. And evben if it didn't drive the corn to extinction, I;m not convinced that it could turn the corn into a new species. If I make small random changes to my car, I might get lucky, and oneof the changes will imporve my cars performance, but I don't think that it will ever turn my car into a truck.

    ReplyDelete
  23. If I make small random changes to my car, I might get lucky, and oneof the changes will imporve my cars performance, but I don't think that it will ever turn my car into a truck.

    Really? You can't see turning your car into a truck in small steps? Your car into a truck? Really?

    Are you serious?

    Of course, your thought experiment of turning a car into a truck is wildly wrong, betraying a gross misunderstanding of what your opponents believe. What you'd actually be doing is copying your car, repeatedly, thousands upon thousands of time, with minor variations. Those copies in the second generation that cease to function are destroyed; those that are better at performing whatever you need them to do are kept (in the case of the car-to-truck transition, this might be hauling cargo rather than carrying passengers) to breed the third generation. Repeat as necessary. This sort of thing does happen and has been demonstrated happening.

    And we have seen new species come about, both in the lab and in the field; check out Wikipedia and Google with "Speciation" to get your learning process started.

    I notice that you didn't say what, exactly, you find untenable in the bacteria-whale transition. Do you not have any specifics? What have you learned about what biologists say about that process?

    Duke

    Duke

    ReplyDelete
  24. Duke:

    "You've said that it takes "a million million" random arrangements to find a functional protein, right? What I'm trying to show is that that number, a million million, 1e12, is actually quite small next to the number chances to find it. That's what I was getting at with the whole 4e20 corn plants thing; that is far, far larger than 1e12. If you look at the number of corn plants that are created every year, you see that there are well over 4e8 times more corn genomes than your 1e12 number."

    You are comparing apples and oranges and not making sense. Let me suggest you wait for AG to make his points.

    ReplyDelete
  25. Duke:

    All the cases mentioned may very well be just part of the normal variation within one species.
    The literature often calls theses possible incipient species not clearly a new species.


    Now, one problem with the car to truck scenario is that some the random changes will not work unless anyother corresponding change happens at the same time. An how many changes do you need to effect some real change that is an advantage? And the majority of copies won't have any change. So you need and lots and lots of copies. For organisms that reporduce quickly, like bacteria that might work. But for whales and humans it becomes a problem.

    And again, I'm not convinced that the case mentioned could turn corn into a new species.

    ReplyDelete
  26. I undertand that a single amino acid in a protein chnages on the average once per every 1,000,000 organisms in a species. That means that if a beneficial chnage requires that two amino acids change, then you need a trillion individuals to get it. Bacteria can do that. Blue whales can't.

    ReplyDelete
  27. According to this model, it would take a really long time for a two amino acid change to happen in humans.



    http://www.genetics.org/cgi/content/full/180/3/1501

    ReplyDelete
  28. I undertand that a single amino acid in a protein chnages on the average once per every 1,000,000 organisms in a species. That means that if a beneficial chnage requires that two amino acids change, then you need a trillion individuals to get it. Bacteria can do that. Blue whales can't.

    Yes! You're exactly right! You've made the correct evolutionary prediction. Blue whales don't create new proteins as they evolve, and this goes for all "higher" animals.

    Unfortunately, that doesn't have anything to do with their speciation. Evolution and speciation in metazoans is a matter of chromosomal re-arrangements and re-shaping of existing proteins, not creation of new proteins. (Rather, I should say, that evolutionary theory predicts that speciation in metazoans comes from re-shuffling of genetic material and not from de novo proteins. If you wanted to disprove neo-darwinian evolutionary theory, you could do it by finding lots and lots of new proteins in human that don't exist in chimpanzees!)

    Do you see why I say bringing up speciation was red herring? It doesn't touch on this article.

    Duke

    ReplyDelete
  29. According to this model, it would take a really long time for a two amino acid change to happen in humans.

    You're being careless in your terms; a two amino acid change is trivially easy. Each of us has proteins that are two amino acids different from our parents due to random mutation. (I have the math for that, if you want to see it).

    Read the article again; this is about the creation, from nothing, of a completely new protein that didn't exist previously, not the modification of an existing protein. (To use creationist terms, we're talking about an observed instance of protein macro-evolution, where as the "two amino acid change" you're denying with this comment is micro-evolution.)

    Duke

    ReplyDelete
  30. Duke,

    I appreciate your willingness to engage and answer questions. On other parts of the Internet I've been, evolution proponents mock and insult those who ask questions about it. Thank you for your civility and willingness to explain things repeatedly.

    -jimbo

    ReplyDelete
  31. Duke:

    First of all, large organisms do have unique proteins, so evolution happened at the level of the protein as well. And evolution of forms requires changes in the genes that control development, which is probably much harder to get right than changing proteins.

    What I should have said about the article was that two specific amino acids necessary to create a new adaptation would take along time.

    And I must say I agree with jimbo. Your civility is refreshing.

    ReplyDelete
  32. First of all, large organisms do have unique proteins, so evolution happened at the level of the protein as well.

    Reference? How many novel proteins are there between chimpanzees and humans? Gorillas and humans? Macaques and humans?

    And evolution of forms requires changes in the genes that control development, which is probably much harder to get right than changing proteins.

    Reference?

    This is actually completely backwards from the case as I understand it. If you look at a list of the proteins used to construct us, we are virtually identical to chimpanzees, gorillas, cats, dogs, mice (of course the exact structure of the protein varies, but it's the same protein in a slightly different shape). Can you find a record of a de novo protein evolving in mammals? The only instance I can think of (ignoring the immune system, which is evolved to produce and react to novel proteins in the form of anti-bodies) is the venom in platypuses' spurs (and I might be wrong about that). Can you find a reference of a new protein evolving in mammals? If you find a reference, I'll discuss it, but until then...

    Current evolutionary theory holds that most metazoan evolution happens through reproductive isolation, which causes the populations to lose their ability interbreed; we can see this played out in real time in ring species such as the Herring Gull/Lesser Black-Backed Bull. Novel proteins don't play any major part at all in animal speciation (although if you find a reference, I'll be happy to change my story.) I'll joyously defend the scientific consensus, but don't ask me to defend your mis-interpretation of the consensus.

    What I should have said about the article was that two specific amino acids necessary to create a new adaptation would take along time.

    Well, first off, you're assuming that the species you're talking about is asexual, where they first have to find one amino acid change, and then find the other after the first moves to fixity in the species. In a sexual species (like almost every animal in the world), things don't work that way. The two amino acid changes can arrive separately and then be mixed together in some fortuitous offspring. This is a real, observed effect.

    Second, how long? Let me give you a back-of-the-envelope calculation. According to current thinking among people who've actually studied the physical world, whales separated from the rest of the hoofed mammals around 53 million years ago. Let's just say, for argument's sake, that throughout that time whales had a stable population of one million individuals, each with a lifespan of 50 years. I'm making these numbers up for illustration, using a ridiculously small population (considering the number of whales who existed up to historical times) and a ridiculously long life (considering that many of the early whales were small meat eaters like hoofed wolves). That is, I'm fudging the numbers in your favor by several orders of magnitude.

    There have been

    5.3e7 years * (5e1 years/whale)^-1 * 1e6 whales = 1.06e12 whales

    since the lineage began.

    There's your trillion whales. Of course (unless you show otherwise), I'm going to say there are no new proteins in whales, just refinements in existing proteins inherited from their common mammalian ancestors. The numbers are there, though.

    And I must say I agree with jimbo. Your civility is refreshing.

    Thank you very much. I'm really seeing this as a two part exercise; explaining the ideas as clearly and succinctly as possible and (more difficult and more important) not losing my temper doing it ^_^.

    Duke

    ReplyDelete
  33. What I ment by unique was unique, not completely unique, but having a unique configuration of amino acids. It does happen in individual species. Human proteins are different than chimp protein. And the article I sited that it is not just the time it takes for the new configuration of amino acids to appear, but for them to work through the population. It says >100,000,000 years for human like species. And this is assuming it is neutral, and not bad for rthe species. If it is negative, it will take longer. Something like the sickle cell mutation that gives resistance to malaria by a change in one amino acid in the hemoglobin molecule is what they have in mind, that happens.

    Now, if evolution has nothing to do with new proteins, then it has to do with contol genes. Reproductive isolation means that changes in the control genes are not overwhelmed by the old, unchanged genes, and have a chance to spread through a population.

    Now, 1 trillion wahles might be enough to get an adaptation that requires 2 amino acid cahnges, but for 3 you need 1 trillion times 1 million whales. That's harder to come by.

    ReplyDelete
  34. The herring gull and the lesser black backed gull have been observed interbreeding, along with the yellow legged gull. So they may very well be all one species.

    And recent events concerning global warming have led to think that scientific consensus, doesn't mean very much.

    ReplyDelete
  35. And there are orphan genes, genes that have no known homologues, in every group of organism.

    ReplyDelete
  36. natschuster : If I make small random changes to my car, I might get lucky, and oneof the changes will imporve my cars performance, but I don't think that it will ever turn my car into a truck.

    The El Camino is a transitional fossil. ;)

    ReplyDelete
  37. Duke, the T-urf13 protein is a trait selected for by human (read: intelligent) breeders for it's primary trait: Cytoplasmic Male Sterility (CMS). An unfortunate side affect was that it makes the plant susceptible to a fungal toxin.

    Meaning plants which have this gene will get killed preferentially by a certain fungus. Also plants that have this gene are male sterile. That's two bad things. And not only that, both of them were obtained in an intelligently directed breeding program.

    ReplyDelete
  38. Nat:
    "Now, 1 trillion wahles might be enough to get an adaptation that requires 2 amino acid cahnges, but for 3 you need 1 trillion times 1 million whales. That's harder to come by."

    There seems to be an overriding yet unacknowledged assumption that many such sequential mutations would have been necessary.
    As has been alluded to, it is more likley that 'new' proteins are less of an issue in terms of evolution than simply altering the timing or order of expression during development; this also goes for the sequential acquisition of a pre-specified series of amino-acid altering mutations.


    Oh, and I, too, would like some justification for th e'million million' assertion. Thanks.

    ReplyDelete
  39. "Duke, the T-urf13 protein is a trait selected for by human (read: intelligent) ... an intelligently directed breeding program."

    I read Randy Wysong's "The Creation Controversey" (1976) about 20 years ago. In it, he claimed that life had been created in a lab (news to me) but that this did not mean abiogenesis was real or that evolution was true, since the people added 'KNOW-HOW' (caps in original) and that the experiment was designed, the materials were designed, etc. He was essentially finding a way to pre-reject any sort of origin of life research and by erroneous extention, to reject any evidence for evolution.

    The more things change, the more they don't.

    ReplyDelete