What is the evidence for “purposeful intervention”?

FMM: Purposeful intervention is pretty much the opposite of random mutation.

FMM notes in the same comment:

 If there in nothing about an idea that distinguishes it from it’s alternative it seems to be superfluous.

So the idea is “non designed mutations” and the alternative is “purposeful intervention”.

Give that, and given FMM has not discarded the idea of purposeful intervention there must be something that distinguishes it from non designed mutations.

What is that distinguishing factor? What is the actual evidence for “purposeful intervention” regarding mutations?

And, more broadly, what is the evidence for “purposeful intervention” in any area of biology? Apart from, of course, wishful thinking.

603 thoughts on “What is the evidence for “purposeful intervention”?

  1. John Harshman,

    Why are you talking about IC unless you think it argues against evolution? What is your argument that IC argues against evolution?

    Many reasons but because a flagellum motor has 30 parts that need to fit together to perform the function of mobility, a detailed evolutionary explanation of its origin is problematic . A large amount of DNA has to store the information to build these proteins. Proteins that need to bind to several proteins need more sequence integrity then ones binding to a single molecule.

    The proteins have to fit together to perform a function that is different then the individual protein function if cooption is part of the process.

    The proteins need to be expressed in the right order. Assembly proteins are required to reliably build the complex.

    Building the complex needs to be integrated into the cell division process.

    All this needs to get done before the advantage of mobility is achieved.

    What “last protein”? Hmmm…apparently calponin, found in smooth muscle, is a troponin homolog.

    Do you think it is easy to evolve on from the other? The application is calcium binding and muscle control. Tropomyosin is also involved with troponin in calcium binding muscle control in skeletal muscles and does not appear to be involved muscle control with sooth muscles.

    What do you mean by “most of the parts are different”? I would like you to make some explicit claims. Do you think titin didn’t evolve? Do you think its only function is in skeletal muscle? Do you think it has no homologs? What is the relevance of its size?

    The structure of these muscle cells are very different and involve different parts including different versions of actin and myosin.

    I think titin was designed to work with modified versions of both actin and myosin and are animal specific. Titin needed to be designed specifically for each animal type. For example chimp and human skeletal titan are over 2000 AA difference in length. This difference is problematic for the chimp and human ancestral connection hypothesis. Humans largest titin isoform is around 34000 AA in length where chimps is around 32000 AA in length.

    Titin is also found in cardiac muscles. Its size makes its evolution very challenging although somewhat simplified given it has repeated amino acid sequences. Titin contains a single exon of almost 6000 AA.

  2. Entropy: Wouldn’t you agree that the mere existence of so many protein families, each containing lots and lots of variants, at the very least suggests that “cliffs” are not that much of a problem?

    A sane person might note that the existence of alleles demonstrates the possibility of gradual change. Behe understands this, which is why he had to work so hard to assert a counterexample. Wrote a whole book arguing for cliffs, unsuccessfully.

  3. colewd,

    You really don’t know anything about this, do you? You can google, you can pull out little details, many of them wrong, but you can’t assemble it all into any coherent argument.

  4. Corneel: Definitely, neither you or I are very representative, I fear.

    I would also like to see how folks on the spectrum would choose as well as those with schizophrenia.

    I think you should come up with something like 20 sets of three and after an observer is done choosing which one is most likely to be from an intentional process. He could see how his choices measured up against the rest of the participants.

    A normal person’s choices should be similar to everyone else’s.

    Corneel: 2 is also random and messy, but can only take 10 values. This is why you need to control the range.

    perhaps those ten values come to the surface on their own as part of the process being recorded?

    regardless I do agree that controlling the range is important. It’s analogous to choosing how closely to look at a phenomena in “real life”.

    What looks like a circle from one resolution might appear to be just roughly circular from another. So it’s important that you use similar resolution when comparing different phenomena.

    That is an important lesson to learn when inferring intent in everyday life as well as this exercise.

    Corneel: Rhythmic waves? hey,what about music?

    Most rhythmic waves are non-random rhythmic and algorithmic.

    This kind of “Rhythmic waves” on the other hand is non-random rhythmic and and non-algorithmic 😉

    do you see the difference?

    peace

  5. OMagain: For all FMM knows, tinnitus is how the universe creator has picked to talk to his chosen people.

    little hints like this one let me know you are not paying attention 😉

    peace

  6. Entropy,

    Wouldn’t you agree that the mere existence of so many protein families, each containing lots and lots of variants, at the very least suggests that “cliffs” are not that much of a problem?

    I agree based on your claim they are not a problem and this is quite a mystery. I read a credible paper today and will attach the abstract that the chance of a selected nuclear protein of becoming non functional with one AA random substitution is greater then 30%.

    This number is larger then I expected and if it is common then even subtle natural evolution of nuclear proteins appears highly unlikely.

    Protein tolerance to random amino acid change
    Haiwei H. Guo*, Juno Choe†, and Lawrence A. Loeb*‡
    *Joseph Gottstein Memorial Cancer Laboratory, Departments of Pathology and Biochemistry, University of Washington School of Medicine, Seattle, WA 98195-7705; and †Institute for Systems Biology, Seattle, WA 98103

    Mutagenesis of protein-encoding sequences occurs ubiquitously; it enables evolution, accumulates during aging, and is associated with disease. Many biotechnological methods exploit random mutations to evolve novel proteins. To quantitate protein toler- ance to random change, it is vital to understand the probability that a random amino acid replacement will lead to a protein’s functional inactivation. We define this probability as the ‘‘x fac- tor.’’ Here, we develop a broadly applicable approach to calculate x factors and demonstrate this method using the human DNA repair enzyme 3-methyladenine DNA glycosylase (AAG). Three gene-wide mutagenesis libraries were created, each with 105 diversity and averaging 2.2, 4.6, and 6.2 random amino acid changes per mutant. After determining the percentage of func- tional mutants in each library using high-stringency selection (>19,000-fold), the x factor was found to be 34% 􏰀 6%. Remark- ably, reanalysis of data from studies of diverse proteins reveals similar inactivation probabilities. To delineate the nature of toler- ated amino acid substitutions, we sequenced 244 surviving AAG mutants. The 920 tolerated substitutions were characterized by substitutability index and mapped onto the AAG primary, second- ary, and known tertiary structures. Evolutionarily conserved resi- dues show low substitutability indices. In AAG, 􏰁 strands are on average less substitutable than 􏰂 helices; and surface loops that are not involved in DNA binding are the most substitutable. Our results are relevant to such diverse topics as applied molecular evolution,

  7. colewd:
    I agree based on your claim they are not a problem and this is quite a mystery.

    I wish it was a mystery. I’d have something more interesting to investigate. Anyway, this is where I tell you that when we examine the sequences and their variation we can see some clues for purifying selection, as we expect since they have diverged from a common ancestral sequence.

    We see, for example, that the genes are more conserved than the sequences around them, showing that the coding regions are kept more conserved by purifying selection than the sequences surrounding them (introns, for example). We also see higher rates of synonymous substitutions (substitutions in the DNA that don’t change the encoded amino-acid), than of non-synonymous substitutions (substitutions in the DNA that change the encoded amino-acid). Both of these are evidence for purifying selection. So, yes, there’s evidence that “cliffs” are “avoided” by purifying selection.

    As you might notice, I have no option but to be convinced. Surely you can make some excuse as to why this is not indicative of what it is indicative: that these sequences diverge from common ancestral sequences, that “cliffs” are “avoidable.” You might even find a way to justify the “appearance” of purifying selection in the examination of these sequences. What you cannot deny, is that I have very good reasons to be convinced.

    colewd:
    I read a credible paper today and will attach the abstract that the chance of a selected nuclear protein on becoming non functional with one AA random substitution is greater then 30%.

    Thanks. the paper looks interesting. I don’t know which numbers might be right, but, as I said, the data say that it’s possible for purifying selection to keep problem at bay, and that sequences can diverge a lot and still be functional. Of course, the evidence for purifying selection also means that a good number of amino-acids would not work, which is to be expected. Do you see how interesting this is?

    colewd:
    This number is larger then I expected and if it is common then even subtle natural evolution of nuclear proteins appears highly unlikely.

    I don’t see why would this make protein evolution unlikely, when we have evidence showing exactly what would be expected from the evolution of proteins, like diverse sequences within a family, with their genes showing signs of purifying selection.

  8. fifthmonarchyman: Most rhythmic waves are non-random rhythmic and algorithmic.

    This kind of “Rhythmic waves” on the other hand is non-random rhythmic and and non-algorithmic 😉

    do you see the difference?

    Heh, you don’t listen to house music, do you? Modern music often involves programming loops of musical sequences, which sounds pretty algorithmic to my ears. You are moving into a grey area here.

    This reminds me of the darwintunes project. Are you familiar with that? Several researchers from the Imperial College London developed an algorithm that would generate short looping polyphonic sound sequences. These were rated for aesthetic appeal by a population of consumers on the web, and the most appealing sequences were used as progenitors for a new generation. The point is that no human-derived sounds, rhythms, or melodies were provided as input to the algorithm, but that the evolution of the sequences was guided by humans (and the result sounds pretty neat). So how does that square with your statement in the OP that “Purposeful intervention is pretty much the opposite of random mutation.”? I am also really curious where this goes in your classification scheme of random / algorithmic / intentional patterns.

  9. fifthmonarchyman: I think you should come up with something like 20 sets of three and after an observer is done choosing which one is most likely to be from an intentional process.

    The permuting of original sequences to generate random pattern seems straightforward enough. But what do you intend to use as intentional sequences? And how do you generate matched algorithmic sequences?

  10. Corneel: Modern music often involves programming loops of musical sequences, which sounds pretty algorithmic to my ears. You are moving into a grey area here.

    Perhaps, I think that what makes music interesting is not “the loops” themselves but the chosen connections between them.

    Corneel: The point is that no human-derived sounds, rhythms, or melodies were provided as input to the algorithm, but that the evolution of the sequences was guided by humans (and the result sounds pretty neat).

    That sounds about right.

    It’s that “guidance” that I’m interested in. I think that much of human behavior is merely blind habit and repetition but there are punctuations of intention now and then. Those are the moments that make us persons instead of animatrons.

    Corneel: So how does that square with your statement in the OP that “Purposeful intervention is pretty much the opposite of random mutation.”?

    I don’t see any conflict. Perhaps I’m missing your point

    Corneel: I am also really curious where this goes in your classification scheme of random / algorithmic / intentional patterns.

    The “guidance” part of the process is where the intentional comes in.

    Also keep in mid that it’s always possible to hide your intentions if you try hard enough.

    Corneel: But what do you intend to use as intentional sequences?

    non-algorithmic non-random ones. Like the one above.

    It’s not like I’m declaring these sorts of patterns to be intentional it’s that I hypothesize that most folks will be more likely to deem them to be so.

    Corneel: And how do you generate matched algorithmic sequences?

    When it comes to my reported weather hobby I compare things like actual reported data with the output of forecasting algorithms.

    Before that I used an EA on the randomized data that targeted an r squared value of 80% or so.

    If we are just looking to verify what sort of pattern that we are more likely to attribute to intentional processes any sequence that is the result of an algorithm would do as comparison.

    peace

  11. fifthmonarchyman: Corneel: So how does that square with your statement in the OP that “Purposeful intervention is pretty much the opposite of random mutation.”?

    I don’t see any conflict. Perhaps I’m missing your point

    You were trying to infer intent at the level of individual mutations and DNA sequences, right?

    In this particular situation of evolving musical sequences, the mutations were random, but the subset that ended up in the program were not. So the “design” was not in the creation of mutations, but in the way they were combined in the musical genome.

  12. Corneel: You were trying to infer intent at the level of individual mutations and DNA sequences, right?

    no, I’m not trying to infer intent anywhere.

    I’m hypothesizing that we are hardwired to infer intent when presented with certain types of patterns. That is all.

    Corneel: So the “design” was not in the creation of mutations, but in the way they were combined in the musical genome.

    OK, I think that is interesting enough.

    There is no rule about where we choose to tweak the algorithm only that the pattern as a whole is not explained by an algorithm

    I’m curious about the different places that we infer intention for me it’s usually associated with something that’s unexpected.

    peace

  13. fifthmonarchyman: no, I’m not trying to infer intent anywhere.

    I’m hypothesizing that we are hardwired to infer intent when presented with certain types of patterns. That is all.

    Are humans being hardwired to infer intent one of those certain types of patterns?

  14. fifthmonarchyman:

    I’m curious about the different places that we infer intention for me it’s usually associated with something that’s unexpected.

    In other words, it’s a function of your ignorance?

  15. Corneel: In this particular situation of evolving musical sequences, the mutations were random, but the subset that ended up in the program were not. So the “design” was not in the creation of mutations, but in the way they were combined in the musical genome.

    That’s a non sequitur. Randomly generated doesn’t mean not designed.

    There absolutely was “design” in the generation of random sequences, for what would have been the point of generating the SAME sequences over and over?

  16. newton: Are humans being hardwired to infer intent one of those certain types of patterns?

    Interesting question.

    I’d say It depends on if animals have a similar instinct.

    Is there any compelling evolutionary reason that it’s better to begin with an inference that the rustle in the bushes is the result of intent rather than something to be investigated further or avoided as the case may be?

    peace

  17. Fair Witness: In other words, it’s a function of your ignorance?

    I’d say yes it begins as a function of our ignorance when presented with certain patterns.

    If we already knew a phenomena was intentional then there is no reason to infer it.

    Mung: Where would be be without ignorance?

    Exactly.

    If we already knew everything then things like inference would be nonsensical.

    peace

  18. Mung: There absolutely was “design” in the generation of random sequences, for what would have been the point of generating the SAME sequences over and over?

    So we have the choice to vary the sequences and the choice of which sequences to string together and the choice of what order.

    Seems like a whole lot of designing going on

    peace

  19. fifthmonarchyman: I’d say yes it begins as a function of our ignorance when presented with certain patterns.


    If we already knew a phenomena was intentional then there is no reason to infer it.

    If it is born from ignorance, then it can be completely wrong.

    What’s the next step needed to confirm the inference?

  20. Fair Witness: If it is born from ignorance, then it can be completely wrong.

    Of course, just as it is with any observation or conclusion.
    I might be a brain in a vat or in the matrix after all.

    Fair Witness: What’s the next step needed to confirm the inference?

    What do you mean by “needed”?

    What is the next step “needed” to confirm my inference that I’m not a brain in a vat?

    peace

  21. Entropy,

    . Anyway, this is where I tell you that when we examine the sequences and their variation we can see some clues for purifying selection, as we expect since they have diverged from a common ancestral sequence.

    I agree that purifying selection is real based on your evidence. How do you think this explains the large divergence in the protein family if we use a 30% chance of a single substitution eliminating gene function as a working hypothesis?

    Purifying selection is another way of eliminating the gene from the population. What we need to explain is how several divergent functional sequences were found.

  22. colewd:
    I agree that purifying selection is real based on your evidence. How do you think this explains the large divergence in the protein family if we use a 30% chance of a single substitution eliminating gene function as a working hypothesis?

    In the simplest terms, the large divergence is explained by the few changes in amino-acids that were not eliminated by purifying selection. More time means more chances that an “allowable” non-synonymous substitutions will occur. Allowable substitutions are not eliminated by purifying selection. We can also measure the strength of purifying selection and thus try and estimate the amount of substitutions that would be “rejected” by purifying selection. They authors of the paper you quoted talk about using such methods to estimate the proportion of “intolerable” mutations in some protein families.

    colewd:
    Purifying selection is another way of eliminating the gene from the population. What we need to explain is how several divergent functional sequences were found.

    As I said, more time means more opportunities for “allowable” changes in amino-acids to occur.

  23. Mung: That’s a non sequitur. Randomly generated doesn’t mean not designed.

    There absolutely was “design” in the generation of random sequences, for what would have been the point of generating the SAME sequences over and over?

    Hi Mung, nice to see you back.

    Not listening to house music either, I take it? I love hearing the same sequences over and over.

    Still, that is a peculiar type of design you propose. As I understand it, “randomly generated” just meant that the mutations in the darwintunes program were random with respect to aesthetic appeal, and most of them resulted in awful music. But if somebody were to compose a song, why would he/she introduce anything but aesthetically pleasing arrangements?

  24. Entropy: In the simplest terms, the large divergence is explained by the few changes in amino-acids that were not eliminated by purifying selection. More time means more chances that an “allowable” non-synonymous substitutions will occur. Allowable substitutions are not eliminated by purifying selection.

    Also, positions in the sequence are not independent of one another. If an “allowable” non-synonymous substitution happens, that may cause another previously non-allowable substitution at another site to become allowable. (Of course it may also cause a previously allowable substitution to become non-allowable, but that doesn’t affect the point of increasing exploration of sequence space without significant change of function.)

  25. fifthmonarchyman: Is there any compelling evolutionary reason that it’s better to begin with an inference that the rustle in the bushes is the result of intent rather than something to be investigated further or avoided as the case may be?

    I think there is a evolutionary advantage to be aware of the environment for clues of possible dangers or prey. Matching patterns to known patterns.

    Do we agree that it is likely the signal( leaves rustling ) is not designed, since it likely lacks intent to create the signal?

  26. Corneel: But if somebody were to compose a song, why would he/she introduce anything but aesthetically pleasing arrangements?

    You would need to know something about the designer.

  27. newton: Do we agree that it is likely the signal( leaves rustling ) is not designed, since it likely lacks intent to create the signal?

    Assuming occasionalism is false I would say the signal is not (directly) designed but the behavior it illuminates could very well be.

    newton: I think there is a evolutionary advantage to be aware of the environment for clues of possible dangers or prey. Matching patterns to known patterns.

    Do you think that other animals have the same sort of feeling of teleology that we do when presented with certain rustlings in the bushes?

    peace

  28. Corneel: if somebody were to compose a song, why would he/she introduce anything but aesthetically pleasing arrangements?

    perhaps she believes that random sequences eventually lead to something that is aesthetically pleasing.

    peace

  29. newton: You would need to know something about the designer.

    True, but I am still entitled to find that peculiar, am I not? 🙂

    fifthmonarchyman: perhaps she believes that random sequences eventually lead to something that is aesthetically pleasing.

    But if that is the case, then purposeful intervention is not the opposite of random mutation, right?

  30. fifthmonarchyman: perhaps she believes that random sequences eventually lead to something that is aesthetically pleasing.

    peace

    I agree, differentiation between the intentional and the unintentional often requires more information than an isolated output.

  31. fifthmonarchyman: Do you think that other animals have the same sort of feeling of teleology that we do when presented with certain rustlings in the bushes?

    If one wishes to engage in anthropomorphism. What makes a rustling “certain”?

  32. Corneel: But if that is the case, then purposeful intervention is not the opposite of random mutation, right?

    No, in that case what happened was not random but purposeful mutation.

  33. newton: I agree, differentiation between the intentional and the unintentional often requires more information than an isolated output.

    In this thread we are not talking about the final conclusive determination we might make.

    We are talking about the initial innate inference and that often happens with an isolated output. A rustle in the bushes for example

    peace

  34. Just for the record, I think it’s a little bit problematic to say that an inference is being made when an animal (even a human animal) detects a sound that it takes to be a probable indicator of another animal hiding in the bushes or whatever. What I’m concerned with is whether that cognitive act — responding to a stimulus as probably indicating the kind of object that generally causes stimuli of that kind — is really an inference. What’s the difference between a genuine inference (“if p, then q”) and a mere association (“p and q”)?

  35. fifthmonarchyman: Corneel: But if that is the case, then purposeful intervention is not the opposite of random mutation, right?

    No, in that case what happened was not random but purposeful mutation.

    Eh? A purposeful mutation resulting in a random sequence? I am losing track of this discussion, I am afraid.

  36. Corneel: Hi Mung, nice to see you back.

    Thank you!

    Do you agree that there can be randomness by design, that the two are not mutually exclusive?

  37. newton: What makes a rustling “certain”?

    In Texas it’s when your cows have gone missing and you find them in the possession of a bunch of complete strangers who cannot explain how they came to have them in their possession.

  38. fifthmonarchyman: That it can be distinguished from other rustlings

    I’ve always thought that what distinguishes the kinds of rustlings is the size of their wings, but I’ve never actually seen one. They are tiny angels whose purpose is to move the leaves if plants and trees when the wind blows.

  39. Mung: Do you agree that there can be randomness by design, that the two are not mutually exclusive?

    In principle yes, but as I indicated in a previous comment, that strikes me as an odd way of designing things. Also it seems to go against the way design is portrayed here usually.

    The “randomness” of the mutations in the darwintunes set-up (and by analogy, in biological organisms) is with respect to aesthetic appeal (= fitness). We can envisage a scenario where a designer on purpose introduces a collection of random changes as substrate for a selection process. After all, this is exactly what happened in the darwintunes example. But, given the usual resistance of the resident IDers here to the idea that random (lucky, blind, mindless, bladibla) processes can ever generate something useful, I am rather surprised that Fifth and you do not immediately reject that idea.

    So tell me, are all beneficial mutations purposefully designed or not?

  40. We haven’t even discussed one of the most useful implications of my idea. It allows for a better Turing Test.

    Simply take two copies of the same software and run them on the same hardware with the same inputs and if one of the two produces a pattern that is distinguishable (ie non-random) and non-algroythimic then we can say that that system has intent. In other words it is exhibiting intelligent behavior.

    peace

  41. newton: If one wishes to engage in anthropomorphism.

    Cool, then I think we can say that the tendency to infer design from patterns that are non-algorithmic and non-random is not itself a pattern that is conclusively non-algorithmic and non-random

    peace

Leave a Reply