The Fridge-o-matic Challenge

I once offered nonlin.org the ‘fridge-o-matic’ challenge. The idea was to give me a number from 1 to 9 (a shelf in my fridge/freezer), and a lateral location Left, Middle or Right, and a depth locator Front, Middle, Back. That gives 81 sectors. I go to the named sector and fumble for the nearest organism (he’d have to trust me not to cheat). If it’s a stew or pizza I’d need more info. Then I’d need a chromosome number in that organism, and a gene number on the chromosome. We could shortcut the back’n’forth by giving numbers 1-50 and 1-10,000 which I’d then normalise to the actual counts.

Armed with this target of a random gene in a random organism in my fridge, I’d go to a public database and (if it’s in) run a BLAST on the gene to find the sequence alignments, arranged by closeness of match. I offered high confidence that this would recover a reasonable approximation of the Linneaean hierarchy – closest matches in species, then genus, then Family etc. This, I feel, nicely fits with the hypothesis of Common Descent of taxonomic ranks.

Now, the Linnaean hierarchy was based on morphology. He did not have the technology to ‘see’ genes directly. So we’d have to wonder why I could stick my random thumb in and pull out a Linnaean plum pretty much every time, on genes invisible to him.

Nonlin declined to play (he is, frankly, no fun at all). “It’s expected”, he said. “Like organisms will have like genes”. But for that to be a relevant objection, every single genetic difference between every single species pair must be involved in every morphological distinction between them. That requires quite a lot of heavy lifting from a gene. Its job might be to … I dunno, ligate a strand break in DNA let’s say. But it’s also got to do it in a squid way, and a dandelion way, and a yeast way that feeds through to morphology for every gene. We’re always being told how brittly unchangeable genes are, yet we seem to have a huge range of latitude in genes that are notionally pinned in place both by primary function and by multiple morphological roles across species.

I bring this up now for two reasons related to nonlin’s ‘Sexual selection’ thread.

1) If every difference between every gene really were morphologically significant, there’d be a constant stream of mutants not possessed of the vital characteristics that render the sexes mutually attractive to each other but not to members of other species.

2) Sexual dimorphism is a particularly striking morphological difference, which, on the ‘genes=morphology’ view, would require similar genetic differences between the sexes to those observed between like species. But most genes, even those involved in sexual differences, reside on the autosomes, which pass through both genders of offspring.

The answer, of course, is that genetic differences on the broad scale are not the key to morphological difference. Those differences are under the control of relatively few genes, and a lot of regulation (itself genetic in origin). This is as true between species as it is within (don’t fight me on this; it’s an established fact!).

Anyway, anyone fancy taking the Fridge-o-matic for a spin?

+2

166 thoughts on “The Fridge-o-matic Challenge

  1. Whilst I’m reasonably familiar in principle how sequence comparison works my practical experience is fifty years out of date. Can anyone with reasonable computer skills, a terminal and an internet connection perform a BLAST analysis? It would be great to work an example with hand-holding – a bit like a Gordon Ramsay cookalong.

    0
  2. Alan Fox,

    It’s dead easy – I just picked it up from fiddling round a bit (although John Harshman would probably have a fit at my amateurishness). There are numerous ways into the program – most protein and genome browsers have a BLAST button, once you’ve found a sequence or gene of interest, but you can dive straight in if you already have a sequence, in nucleotide or amino acid form, or a gene ID. I quite like Uniprot.

    Paste the sequence below into the box, click ‘Run BLAST’ and sit back. I picked this gene deliberately; it’s not something I keep in my fridge! I’m setting myself up for a fall with the ‘fridge-o-matic’, for reasons many biologists will spot straight away, but I’m confident it will do the job most of the time.

    MTNMRKTHPLFKIINHSFIDLPAPSNISSWWNFGSLLGVCLMVQIITGLFLAMHYTSDTM TAFSSVTHICRDVNYGWLIRYMHANGASMFFICLFLHVGRGLYYGSYTFMETWNIGVLLL FAVMATAFMGYVLPWGQMSFWGATVITNLLSAIPYIGTTLVEWIWGGFSVDKATLTRFFA FHFILPFIIAALAIVHLLFLHETGSNNPTGLNSDADKIPFHPYYTIKDILGILIMFLILM TLVLFFPDMLGDPDNYMPANPLNTPPHIKPEWYFLFAYAILRSIPNKLGGVLALILSILI LALMPFLHTSKQRSLMFRPITQILYWILVANLLILTWIGGQPVEHPFIIIGQLASISYFS IILILMPISGIIEDKMLKLYP

    0
  3. Doing blast searches is fun and, as Allan notes, dead easy. The tricky bit may come in interpreting the significance of the results obtained, particularly the meaning of those apparent exceptions to the Linnaean hierarchy. The databases are not error-free.

    0
  4. DNA_Jock:
    Doing blast searches is fun and, as Allan notes, dead easy. The tricky bit may come in interpreting the significance of the results obtained, particularly the meaning of those apparent exceptions to the Linnaean hierarchy. The databases are not error-free.

    Indeed. And I may be given a gene that’s an ORFan, or too invariant, or subject to LGT or homoplasy. But still, I’m prepared to put my virtual money where my virtual mouth is, provided people don’t think I am offering a guarantee of success. Let the chips fall where they may!

    0
  5. DNA_Jock: Doing blast searches is fun and, as Allan notes, dead easy. The tricky bit may come in interpreting the significance of the results obtained, particularly the meaning of those apparent exceptions to the Linnaean hierarchy.

    And that’s the crux of the problem. Why would anyone think this supports “evolution”? You still don’t get the “affirming the consequent” fallacy? And it’s even worse given “evolution” never ever predicted anything genetics. Much less an obvious expected match between genetics and morphology.

    Genetics goes its merry way, and “evolution” always comes behind “explaining” what needs no further explanation than what geneticists discovered. IOW, who cares of an explanation of genetics in terms of “evolution” when “evolution” cannot forecast anything genetics.

    Also, reusing code is a standard practice of design. In fact, you can use the same (similar) tool to track what portion of the code was written by who knows what “ancestor”. And of course it’s all Intelligently Designed.

    0
  6. Nonlin.org,

    In fact, this whole thing goes from Linnaeus to genetics. A direct link from a Christian’s classification to the most modern science, right over the Darwinist nonsense. Love it!

    0
  7. Nonlin.org:
    Nonlin.org,

    In fact, this whole thing goes from Linnaeus to genetics. A direct link from a Christian’s classification to the most modern science, right over the Darwinist nonsense. Love it!

    Genetics provides a fundamental explanation of what Linnaeus observed. The mistake you make is in thinking that every last genetic difference underlies every last morphological one.

    0
  8. Nonlin.org: And that’s the crux of the problem. Why would anyone think this supports “evolution”? You still don’t get the “affirming the consequent” fallacy? And it’s even worse given “evolution” never ever predicted anything genetics. Much less an obvious expected match between genetics and morphology.

    “I expect all genetic differences to track morphological differences. They do”. How is this not an affirming the consequent fallacy? And why would all differences be involved in morphology anyway? A DNA ligase, for example?

    Genetics goes its merry way, and “evolution” always comes behind “explaining” what needs no further explanation than what geneticists discovered. IOW, who cares of an explanation of genetics in terms of “evolution” when “evolution” cannot forecast anything genetics.

    Is firecasting important in this realm? What can genetics forecast? Or ID for that matter?

    Also, reusing code is a standard practice of design.

    This is a logical fallacy known as ‘argument from analogy’. A stretch of a genome is not a piece of computer code, nor anything like.

    0
  9. Allan Miller: Genetics provides a fundamental explanation of what Linnaeus observed. The mistake you make is in thinking that every last genetic difference underlies every last morphological one.

    That’s evolutionary genetics provides an explanation of what Linnaeus observed.

    But it’s obvious that Nonlin missed the whole point, and in the same way you had already explained in the OP.

    0
  10. No takers? Ok, here’s a gene I chose pretty much at random. Methyl-malonyl CoA epimerase, a mitochondrial protein involved in fatty acid metabolism. And also, according to Nonlin Genetics, involved in the specific differences between humans, gorillas, chimps, colobus monkeys … not only the protein itself, but also the 3 bits that are cut out (the introns). All of these variations are directly involved in making humans distinct from chimps. Along with almost every other gene I could randomly pick, doing all manner of esoteric molecular rearrangements. This seems something of a stretch. All epimerases do is flip between isomers – they swap the groups sticking out of a chiral carbon in a substrate molecule – in this instance, a methyl group is swapped with hydrogen. But Nonlin ‘expects’ this to follow morphology. The way gorillas flip these groups has to be different from chimps, which in turn differ more from colobus monkeys than they do from us … They are, however, all doing the exact same thing as far as their substrate is concerned.

    MARVLKAAAANAVGLFSRLQAPIPTVRASSTSQPLDQVTGSVWNLGRLNHVAIAVPDLEK AAAFYKNILGAQVSEAVPLPEHGVSVVFVNLGNTKMELLHPLGRDSPIAGFLQKNKAGGM HHICIEVDNINAAVMDLKKKKIRSLSEEVKIGAHGKPVIFLHPKDCGGVLVELEQA

    0
  11. Here’s methyl thingywotsit in a biochemical pathway chart. You can see the rearrangement it does, methyl group flipped (in a distorted projection).

    All the blue writing refers to the various enzymes that accomplish the various transformations in the vicinity. I could pick any of ’em. This is but a zoomed-in part of a massively bigger whole; there are loads to go at. Are all of these involved in morphology?

    +2
  12. Allan Miller:
    They are, however, all doing the exact same thing as far as their substrate is concerned.

    One of the classic experiments on homeobox proteins, involved in embryogenesis, is one where the scientists exchange a drosophila homeobox gene with its equivalent in humans, and lo and behold, the little flies develop all right.

    0
  13. Entropy,

    Yes. We could point to within-population polymorphisms too, which are of exactly the same order as between-species polymorphisms. Despite a change in underlying genetics, which nonlin fondly imagines means a change in morphology, ‘they’re still chimps’.

    +1
  14. Says Allan:

    I offered high confidence that this would recover a reasonable approximation of the Linneaean hierarchy – closest matches in species, then genus, then Family etc. This, I feel, nicely fits with the hypothesis of Common Descent of taxonomic ranks.
    On sample of one this seems to check out. 🙂

    0
  15. Alan Fox:
    Mammalian cytochrome B?

    Yep – specifically, mouse, hence you should be seeing the taxonomy unroll from its position outwards. Cyt B is frequently used as a mitochondrial phylogenetic marker because it varies on approximately the right scale. As I say, this was a deliberate choice; methylmalonyl CoA epimerase somewhat less so.

    It does need care in interpretation, though, as Jock says. In the latter case, for instance, gorillas are closer than chimps and bonobos on the aligned segment, but there’s a deletion/insertion at one end, so even though human/gorilla gives “100%”, it isn’t. But it does illustrate a remarkable general thing, which does need an explanation: “I expect genes to follow morphology” isn’t it, especially for low-level cellular function..

    0
  16. This will probably expire in the next few days – it can be repeated by reBLASTing the methylmalonyl CoA epimerase sequence above, then pick ‘MSA viewer’ on the results, and change ‘Coloring’ to ‘Show Differences’. The display can be zoomed to specific residues.

    This small 176-acid protein has a domain starting in position 47, and it’s visually striking how variation tends to concentrate mainly in residues to the left of that start point, particularly when one scrolls to more distant relatives. Even where variation occurs inside the domain, it tends to be biased towards conservative differences – the Alanine-Valine difference between us and chimps, for instance, in position 76, differs only by a 1-carbon extension and forked methyl group in the second. Both are hydrophobic.

    It would seem pretty foolish to try and control morphological distinctions using such a protein under clear functional constraints for its primary chemical role. And, there’s simply no evidence that it happens – that chimps and a couple of gibbon genera (and, incidentally, pumas and leopards) depend morphologically on this genetic distinction from gorillas and people, with a vital valine at position 76 instead of alanine.

    0
  17. DNA_Jock: Oh nonlin, I gave you a clear example of a useful genetic prediction from evolution a while ago.

    I’m looking at the “theory” of “evolution” and just don’t see anything of that nature there.

    Help me out, are you proving “gradualism”, “divergence of character”, “struggle for survival”, “fitness”? What exactly do you call “a useful genetic prediction” – presumably of “evolution”?!?

    Allan Miller: And also, according to Nonlin Genetics, involved in the specific differences between humans, gorillas, chimps, colobus monkeys … not only the protein itself, but also the 3 bits that are cut out (the introns).

    Allan Miller: But Nonlin ‘expects’ this to follow morphology.

    Will you stop making up stuff?

    Allan Miller: Despite a change in underlying genetics, which nonlin fondly imagines means a change in morphology, ‘they’re still chimps’.

    I do? Either you produce a quote, or admit you don’t have a clue about what others “imagine” or “expect”.

    0
  18. Nonlin.org to Allan:
    I do? Either you produce a quote, or admit you don’t have a clue about what others “imagine” or “expect”.

    You’re a bit slow Nonlin. Allan’s interpretation is the most charitable one to your idiotic “explanation” for the fact that any shared genes will show differences that, when displayed in a tree, closely follow Linnaean taxonomy. Your non-explanation is that “like organisms will have like genes.” For that to work, the phenotypic differences would have to be a product of every change in every protein/gene. For that to be “reused code” they’d have to be identical, rather than follow the taxonomy with their differences.

    Unlike Allan, I don’t give you the benefit of the doubt, and I think you don’t understand the problem at all. I seriously doubt that you understand what Allan means by the differences in the protein, or in the gene, sequences. I think you only understood the word “differences” and you thought, “of course differences between organisms follow taxonomy! Haha.”

    I won’t explain what those differences are to you, mostly because you neither care, nor have any aptitude to understand. Heck. I doubt you understood these explanations. You’re just too stupid.

    0
  19. Nonlin.org:

    What exactly do you call “a useful genetic prediction” – presumably of “evolution”?!?

    Well, it would be a ‘prediction’ about ‘genetics’ that is of use. Which of these words is giving you trouble?

    I’m looking at the “theory” of “evolution” and just don’t see anything of that nature there.

    Well recognizing you have a problem is the first step to fixing it.

    Help me out, are you proving “gradualism”, “divergence of character”, “struggle for survival”, “fitness”?

    Minor aside: “proving” is the wrong word; science doesn’t do proof.
    The useful prediction arises out of common descent, random mutation and natural selection. “Gradualism”, not so much.
    Now, I explained all of this to you the last time around (in 2018), and you showed no sign of comprehension. So, the next step will be for you to describe to me, accurately, what my argument is. You seen rather scared of attempting this.

    +2
  20. DNA_Jock: Well, it would be a ‘prediction’ about ‘genetics’ that is of use. Which of these words is giving you trouble?

    Your prediction about genetics follows from your general knowledge of genetics and Linnaean classification. You only imagine it follows your belief in “evolution”. There’s nothing in the “mechanism” of “evolution” (the elements of which I listed) related to anything genetics.

    DNA_Jock: The useful prediction arises out of common descent, random mutation and natural selection.

    But it doesn’t. The Linnaean classification preceded “common descent”. What has random mutation to do with anything in this case? Same for “natural selection”?

    DNA_Jock: So, the next step will be for you to describe to me, accurately, what my argument is.

    I don’t entirely know what your argument is. That’s why I’m asking those questions. Btw, when was the last time you reciprocated by repeating my arguments?

    Allan Miller: “an obvious expected match between genetics and morphology”

    That’s not exactly the same, is it? Yes, I expect a match. Not:
    “‘expects’ this to follow morphology”. [genetics to follow morphology?!?]
    or
    “a change in underlying genetics, which nonlin fondly imagines means a change in morphology” [every genetic change to be reflected in morphology?!?]
    Your equivalence seems tendentious.

    And stop pretending we’re debating genetics. We’re not. We’re debating “evolution”. If you didn’t know…

    0
  21. Nonlin.org,

    That’s not exactly the same, is it? Yes, I expect a match. Not:
    “‘expects’ this to follow morphology”. [genetics to follow morphology?!?]
    or
    “a change in underlying genetics, which nonlin fondly imagines means a change in morphology” [every genetic change to be reflected in morphology?!?]
    Your equivalence seems tendentious.

    It is quite unclear what you are saying here. Your obfuscation seems tendentious.

    And stop pretending we’re debating genetics. We’re not. We’re debating “evolution”. If you didn’t know…

    The Fridge-o-matic is about genetics, in the first instance. An interesting pattern emerges for a large number of genes, even those chosen entirely at random. When you do sequence alignments from any start point, the matches tend to descend in order of closeness according to (approximately) Linnaeus’ morphological hierarchy. Now, one explanation of that could be that all the genetic differences between two organisms are involved in their morphological differences, irrespective of gene.

    You seem to be denying that this is your position. If so, this leaves a conundrum: what causes genes with no involvement in morphology to correlate closely with Linnaeus’ classification based on morphology alone? You ‘expect a match’. Why, for a metabolic gene?

    0
  22. Allan Miller,

    I admire your confidence that Nonlin understands more than half the words you wrote.

    ETA: I suspect that, in order to help Nonlin with genetics, we’d have to start by teaching the poor lad the sounds that each letter of the alphabet is supposed to represent.

    0
  23. Nonlin.org: I don’t entirely know what your argument is.

    Clearly you don’t have the first clue what it is. You keep bringing up the Linnaean classification when it is completely irrelevant: it is no use whatsoever in formulating my useful genetical prediction… So you admit your profound ignorance. That’s refreshing, I guess.

    Btw, when was the last time you reciprocated by repeating my arguments?

    I do it often enough . See pages six through nine of
    this thread for half a dozen examples.
    You never reciprocate. I know why.

    0
  24. DNA_Jock: it is no use whatsoever in formulating my useful genetical prediction

    Nonlin didn’t follow the link you provided. He just assumed you were talking about the same thing that’s in Allan’s OP. “Illiteracy” is too kind for describing Nonlin’s problems.

    0
  25. Allan Miller: It is quite unclear what you are saying here.

    That’s your problem. You only understand what you want to.

    Allan Miller: An interesting pattern emerges for a large number of genes, even those chosen entirely at random. When you do sequence alignments from any start point, the matches tend to descend in order of closeness according to (approximately) Linnaeus’ morphological hierarchy.

    Again, as expected based on examples of design. Look at two designs and they will match even where they don’t have to. Because “why reinvent the wheel?” IOW, reinventing the wheel is costly, potentially dangerous, and generally not worth it unless absolutely necessary.

    Allan Miller: Now, one explanation of that could be that all the genetic differences between two organisms are involved in their morphological differences, irrespective of gene.

    You seem to be denying that this is your position.

    There could be a lot more of that than you expect. Especially since:
    “In practice, however, the two terms [morphology and anatomy] are used almost synonymously.”
    But as I explained, the carried design is good enough explanation.

    And what is “evolution” predicting? “Continuous (gradualism) improvement (survival of the “fittest”) at a constant speed (gradualism) in random directions (divergence)”. Which is exactly what we’re NOT seeing. Not continuous, not improvement, not constant speed, and not random directions.

    DNA_Jock: Clearly you don’t have the first clue what it is.

    Maybe it’s you that’s not making any sense.

    DNA_Jock: You keep bringing up the Linnaean classification when it is completely irrelevant: it is no use whatsoever in formulating my useful genetical prediction…

    So you say. But why? You didn’t prove your point back then and are not doing it now.

    You claimed that whatever common between several organisms will also be common with the extra one and whatever different will also be different in the extra one somehow makes sense on the account of “evolution” but not on the account of ID.

    What am I to do with your unsubstantiated claim:

    DNA_Jock: The useful prediction arises out of common descent, random mutation and natural selection.

    …or this one from back then:
    “ID thinking, tellingly, makes no such prediction.”

    And how is your point different than Allan’s?

    DNA_Jock: I do it often enough .

    Those are mostly misrepresentations.

    0
  26. Nonlin.org: That’s your problem. You only understand what you want to.

    Some people with a point to get across attempt to clarify it. Others do that ^^^^

    Again, as expected based on examples of design. Look at two designs and they will match even where they don’t have to. Because “why reinvent the wheel?” IOW, reinventing the wheel is costly, potentially dangerous, and generally not worth it unless absolutely necessary.

    In the cases we’re looking at, it’s not ‘reinventing the wheel’, it’s ‘making small changes to the wheel’. So your design reuse example does not cover the case.

    This particular wheel sits deep inside the organism, invisible to 18th Century naturalists. So the correlation of variations in it with those in things he could see is striking, and, to me, unexpected on a strict Design view. I don’t ‘expect’ a 3mm change in a brake pipe attachment point to correlate with a change in the hubcaps.

    There could be a lot more of that than you expect.

    There could, but the challenge as offered is attempting to find random genes, which may or may not be involved in morphology, without prejudice. If you think all functional genes are involved in morphology, we could look at ‘genes between genes’, so to speak.

    Especially since:
    “In practice, however, the two terms [morphology and anatomy] are used almost synonymously.”

    ?

    But as I explained, the carried design is good enough explanation.

    This does not account for the alanine-valine difference in human and chimp methylmalonyl epimerase. The enzyme’s only function is to flip a methyl group on its substrate – a very important task, but not one that would be expected to influence a feature Linnaeus would use to classify. This has (as far as we can tell) no involvement in morphological differences.

    You may say ‘you never know’, but then you’re explicitly denying a position that claims every genetic difference as a morphological (anatomical?) one, then arguing as if that is what you think. At the same time, you ‘expect’ both human and gorilla MCEE to be the same, and to differ between them and the two chimp species in that one position. Seems like you ‘expect’ whatever turns up.

    +1
  27. Nonlin.org,

    And what is “evolution” predicting? “Continuous (gradualism)

    Changes in the BLAST results above look pretty gradual to me. Unless an amino acid substitution is a big jump in your world.

    improvement (survival of the “fittest”)

    There is no such expectation. Changes can also be random. The alanine-valine substitution above is likely one such. There appears to be no functional reason for it (which is a bit of a pisser for a Design enthusiast).

    at a constant speed (gradualism)

    No, that’s ‘constant-speedism’. You may be thinking of the molecular clock, but one would not expect changes in the genes the Fridge-o-matic picks to be very clock-like.

    in random directions (divergence)”.

    Yep. Hence the pattern of differences as we ascend the Linnaean hierarchy from (almost) any random start point. That’s divergence, writ large. A branching hierarchy at the genetic level.

    Which is exactly what we’re NOT seeing. Not continuous, not improvement, not constant speed, and not random directions.

    Suuuuure … 🤔

    +1
  28. Entropy:
    Allan Miller,
    I admire your confidence that Nonlin understands more than half the words you wrote.

    Haha! It’s more an exercise in articulation than any expectation of persuasion.

    +1
  29. Nonlin.org: Which is exactly what we’re NOT seeing. Not continuous, not improvement, not constant speed, and not random directions.

    Well, what are we seeing then?

    Can you give a specific example of, say, a non-continuous change? And perhaps an explanation of how that unambiguously could only be the product of design?

    0
  30. Also, nonlin, if evolution has a direction what is it and where is it going?

    Logically if it has such a direction and you know what it is you should easily be able to make some predictions. Can you?

    0
  31. Nonlin.org to DNA_Jock:
    And how is your point different than Allan’s?

    If anybody doubted that Nonlin has no clue.

    I rest my case.

    0
  32. OMagain,

    Nonlin means “continuous” in the mathematical sense. Thus, one of Nonlin’s “proofs” that evolution is “false” consists on talking about nature not being mathematically continuous, as in, for example, quantum mechanics (though I don’t remember if he talks about quantum mechanics specifically). So, if evolution does happen, but it happens in discrete steps, no matter how small, then it’s still false because “gradual,” to Nonlin, means continuous in the mathematical sense, and if it’s not continuous, it’s not gradual, and it cannot be evolution. Yep. Nonlin is convolutedly stupid.

    0
  33. Nonlin.org: You claimed that whatever common between several organisms will also be common with the extra one and whatever different will also be different in the extra one somehow makes sense on the account of “evolution” but not on the account of ID.

    That’s better. This is a pretty good effort to describe my argument. And immediately, we see a potential source of confusion: nonlin’s vague phrasing “whatever common” and “whatever different”. The argument is much more specific: it rests on the fact that there are regions of the GH gene where we see a lot of variation, and regions where we see very little. The evolutionary explanation is that the different GH genes descend from a common ancestor, and that the subsequent random mutation has caused the sequences to diverge, unless natural selection is restricting the allowable variation. Some parts of the gene are under strong selective pressure; some are not. Therefore, we predict that those regions of the GH gene that vary between human and mouse and cow will also vary between human and mouse and rabbit. That’s where to make the PCR primers.
    Now, the Design explanation makes no such prediction (for example, if the rabbit sequence were identical to the pelican sequence, this would not constitute evidence against ID). As I noted back in 2018, IDists would like to tailor their Design “predictions” to align as closely as possible with evolutionary predictions. So nonlin comes up with the idea that the kind of results that Allan sees with the Fridge-O-Matic challenge are expected, based off the Linnaean morphology tree. He is claiming (although he oscillates between realizing and not realizing that this is his claim) that the observed differences are design features, designed into these various beasts to make them different, and this design requirement is WHY each of these thousands of genes conform to the hierarchy.
    Allan is making fun of this with housekeeping genes, Entropy with cross-phylum gene transfer experiments.
    My argument stuffs the Design thesis from an entirely different direction.
    The gene I was working on is (purely by chance) actually a gene where some of the differences between species probably ARE tied to morphology. The Design argument would be that my prediction about which regions of the rabbit GH gene would differ from human and mouse is actually a Design prediction, to wit — those regions that differ between human and mouse differ because humans and mice are quite different in size. Design predicts that these regions would also differ in rabbits, because of Design requirements.
    Is that the reasoning behind the Design prediction, nonlin?
    [and yes, it is a trap]

    0
  34. Allan Miller: In the cases we’re looking at, it’s not ‘reinventing the wheel’, it’s ‘making small changes to the wheel’. So your design reuse example does not cover the case.

    What are you talking about? The comparison is across different organisms. Each one of those is ‘reinventing the wheel’. Why don’t you combine whatever parts and make chimeras? Because you can’t. Because going from organism A to organism B is not “small changes”.

    Allan Miller: I don’t ‘expect’ a 3mm change in a brake pipe attachment point to correlate with a change in the hubcaps.

    You mean like when comparing a Jeep to a Ford? What exactly would you expect and where’s the big surprise?

    Allan Miller: This does not account for the alanine-valine difference in human and chimp methylmalonyl epimerase.

    It’s not about the function or size of the component. It’s about the macro difference between organisms. This is likely the root error in your collective thinking. If I look at human vs monkey – whatever subcomponent – I would expect smaller differences than between human and lizard – same subcomponent. Same thing as Jeep to Ford (smaller differences) as opposed to Jeep to a bicycle (more differences).

    Still, this is not a strong generic rule. And it has exceptions. Honda for instance makes cars, motorcycles, lawn mowers, etc. Those might be more similar to each other than to other manufacturer.

    So anyway, what has any of these to do with “evolution”?

    Allan Miller: Changes in the BLAST results above look pretty gradual to me.

    When you’re looking at little monkey vs big monkey? As opposed to little monkey vs snake? Aren’t you just getting confirmation of what you see with your own eyes? Where’s the surprise?

    OMagain: Can you give a specific example of, say, a non-continuous change?

    Climb a mountain and observe the biology change. You see transition to different organisms, not same organisms with slightly different characteristics. This disproved “gradualism”.

    OMagain: Also, nonlin, if evolution has a direction what is it and where is it going?

    What “evolution”?

    DNA_Jock: The evolutionary explanation is that the different GH genes descend from a common ancestor, and that the subsequent random mutation has caused the sequences to diverge, unless natural selection is restricting the allowable variation.

    And you don’t see a problem with your “explanation”? Maybe the fact that it’s presupposing “evolution”? You know, the very thing you want to prove? Ever heard of “circular logic”?

    DNA_Jock: Some parts of the gene are under strong selective pressure; some are not.

    Why would that be? What’s a “strong selective pressure”? What’s your “fitness” anyway? And how do you know if you’re under “strong selective pressure”?

    DNA_Jock: Now, the Design explanation makes no such prediction (for example, if the rabbit sequence were identical to the pelican sequence, this would not constitute evidence against ID).

    What does that even mean?!? What exactly would be “evidence against ID”?!? Are you trying to “prove” randomness again? When will you learn you can’t do that?

    DNA_Jock: He is claiming (although he oscillates between realizing and not realizing that this is his claim) that the observed differences are design features, designed into these various beasts to make them different, and this design requirement is WHY each of these thousands of genes conform to the hierarchy.

    Why are you making up stuff like this? How did you come up with “designed into these various beasts to make them different”?!? How would you or I know? I’m not the Designer.

    DNA_Jock: Design predicts that these regions would also differ in rabbits, because of Design requirements.

    I don’t know what you’re smoking, but you’re not making any sense. What exactly are the “design requirements” and how would you know them?!?

    0
  35. Nonlin.org:
    And you don’t see a problem with your “explanation”? Maybe the fact that it’s presupposing “evolution”? You know, the very thing you want to prove? Ever heard of “circular logic”?

    Yes, but this is actually not circular. A scientific theory is a proposed explanation for a body of observations. In science, most theories are at least partially true, but it’s assumed that all of them are in some respects incorrect and/or incomplete.

    The important point here is, a proposed explanation is not circular. A proposed explanation, to be of any use, must make testable predictions. So people go test, and sometimes the predictions are correct, sometimes not. There is an important difference between making a prediction and presupposing.

    As a simple example, let’s say my model contends that the earth is rotating with respect to the sun. IF my model is correct in this respect, then we on the surface of the earth should see the sun appear to rise and set. But if we do, we didn’t presuppose the correctness of my model, we TESTED it.

    Why would that be? What’s a “strong selective pressure”? What’s your “fitness” anyway? And how do you know if you’re under “strong selective pressure”?

    The proposal here is, any significant change to certain parts of the gene would be fatal to the organism. (Note here that this doesn’t mean these parts of the gene DO not mutate, it means that WHEN this happens, the organism does not survive to pass along that mutation.)

    The selective pressure is to deselect mutations in those locations, and the result is that the gene is conserved without significant change over evolutionary timescales.

    What does that even mean?!? What exactly would be “evidence against ID”?!?

    The common argument against ID is exactly as you imply – that there IS NO evidence against ID, and can’t be. Because no matter what we find, we can always say that it was intelligently designed that way! And therefore, if anything and everything is consistent with ID, there is no test for ID that could ever fail.

    How did you come up with “designed into these various beasts to make them different”?!? How would you or I know? I’m not the Designer.

    You have put your finger on the real problem here. Since nobody is the designer, nobody (not you or I or DNA_Jock) can know the intentions or the methods of the designer. So we either presume a designer and attribute all the various forms of life to Its efforts, or we look for some other process that could produce the observed result. That process we call “evolution.” But since you deny there’s any such thing as evolution, you seem to be saying that the Designer is responsible. That is, the Designer designed differences into the various beasts.

    What exactly are the “design requirements” and how would you know them?!?

    The design requirements are that there are internal differences among the various beasts, which make them morphologically different. If they were all identical on the inside, they’d be identical on the outside.

    But “predicting” that different beasts are different isn’t much use.

    +2
  36. Nonlin.org,

    What are you talking about?

    I’m talking about the differences between genes in different organisms, and the way they map onto the morphological hierarchy of Linnaeues.

    The comparison is across different organisms.

    Yes. Of specific genes.

    Each one of those is ‘reinventing the wheel’. Why don’t you combine whatever parts and make chimeras? Because you can’t. Because going from organism A to organism B is not “small changes”.

    We aren’t talking of the entirety of the changes, we are looking at specific genes and how they change.

    Allan Miller: I don’t ‘expect’ a 3mm change in a brake pipe attachment point to correlate with a change in the hubcaps.

    Nonlin: You mean like when comparing a Jeep to a Ford? What exactly would you expect and where’s the big surprise?

    I knew it was a bad idea to pursue analogy! 🤣 The big surprise is that a gene that is not involved in morphology nonetheless follows the hierarchy built solely on morphology.

    Allan Miller: This does not account for the alanine-valine difference in human and chimp methylmalonyl epimerase.

    Nonlin: It’s not about the function or size of the component. It’s about the macro difference between organisms.

    Nope. It’s about the correlation of micro difference to macro difference. You’d only expect genes involved in forming the features that 18th Century naturalists could arrange on to vary in line with those features. There is no such expectation with the way they stick a methyl group on a substrate.

    This is likely the root error in your collective thinking. If I look at human vs monkey – whatever subcomponent – I would expect smaller differences than between human and lizard – same subcomponent. Same thing as Jeep to Ford (smaller differences) as opposed to Jeep to a bicycle (more differences).

    This expectation is only sustainable through an ignorance of genetics. In the case of methylmalonyl CoA epimerase, there is actually no difference between the human and gorilla sequence. There is a difference between them and the chimps. So this confounds your ‘expectation’ anyway.

    All MCEE does is flip a methyl group. Imagine it as a little tappet that knocks a sticky-out bit through 180 degrees. There is no expectation that this has to be done an increasingly different way as we ascend the hierarchy from a given point, in step with gross morphology, in every organism in my fridge.

    So anyway, what has any of these to do with “evolution”?

    Nothing, if you’d rather.

    +2
  37. Nonlin.org,

    Me: Changes in the BLAST results above look pretty gradual to me.

    Nonlin: When you’re looking at little monkey vs big monkey? As opposed to little monkey vs snake?

    Interesting that you should deliberately create ‘big-leap’ categories, there, just to evade the point. The results are not arranged hierarchically, so there will be artefactual discontinuities anyway. Nonetheless, between organisms supposed to be more closely related, we aren’t seeing massive leaps. You don’t refute gradualism by saying “what about worm-amoeba, huh?”.

    +2
  38. Flint: The important point here is, a proposed explanation is not circular. A proposed explanation, to be of any use, must make testable predictions. So people go test, and sometimes the predictions are correct, sometimes not. There is an important difference between making a prediction and presupposing.

    There’s a few aspects to testing theories:
    1. Is it compatible with reality? And is it uniquely compatible? If compatible but not unique, then it’s just “affirming the consequent”.
    2. What observation would invalidate the theory? I have yet to see any proposal here re “evolution”. Aside from Dawkins’ imbecile Precambrian rabbit or some other such cretinoid clause.
    3. Interpret correctly the observation. You already failed here:

    Flint: As a simple example, let’s say my model contends that the earth is rotating with respect to the sun. IF my model is correct in this respect, then we on the surface of the earth should see the sun appear to rise and set.

    We see sunrise because the earth is rotating around one particular axis of itself. At the pole, the sun doesn’t rise for days on end.

    Flint: The proposal here is, any significant change to certain parts of the gene would be fatal to the organism.

    I’m looking at your “theory” and see no such proposal: “there shall be immutable sections of the DNA”. This is already a major disconnect between your “theory” and the observation. Your “theory” is not being tested!

    Flint: The common argument against ID is exactly as you imply – that there IS NO evidence against ID, and can’t be. Because no matter what we find, we can always say that it was intelligently designed that way! And therefore, if anything and everything is consistent with ID, there is no test for ID that could ever fail.

    You’re wrong because…
    1. There’s no definitive proof in science but…
    2. We reject the null “random” (the alternative to ID). So you go to whatever planet and find some unknown patch. If regular you infer Design, if not, you infer Random. This is not proof. But it is science.
    3. We also reject the null “evolution” (the alternative to ID) since there is no KNOWN (observed!) path from a mass of atoms to the exquisite biologic structures. Or even from one exquisite design to another equally exquisite design.
    4. Once we reject 2 of 3 and only 3 competing theories, we just confirmed ID with the scientific method.

    In summary, the observable is compatible with ID and only with ID.

    Flint: So we either presume a designer and attribute all the various forms of life to Its efforts, or we look for some other process that could produce the observed result.

    There is no other process. I proved that repeatedly.

    And anyway, this OP is supposed to somehow prove “evolution”, not disprove ID.

    0
  39. Allan Miller: We aren’t talking of the entirety of the changes, we are looking at specific genes and how they change.

    You presume “change”. Which is the “theory” that you want to prove. This is circular reasoning. In fact, we are simply looking at DIFFERENCES. Which you have YET to confirm are due to “change”.

    Allan Miller: The big surprise is that a gene that is not involved in morphology nonetheless follows the hierarchy built solely on morphology.

    That may be a surprise for you. But not to any designer that knows a thing or two. As I explained.

    Yes, you tripped yourself with the analogy. But analogies cannot be ignored, so I would have tripped you. Either way, you’re stuck.

    Allan Miller: You’d only expect genes involved in forming the features that 18th Century naturalists could arrange on to vary in line with those features.

    WHY? That’s an irrational expectation.

    Allan Miller: In the case of methylmalonyl CoA epimerase, there is actually no difference between the human and gorilla sequence. There is a difference between them and the chimps. So this confounds your ‘expectation’ anyway.

    I don’t see why I’m supposed to have an expectation of this random fact. It doesn’t confound anything.

    Allan Miller: There is no expectation that this has to be done an increasingly different way as we ascend the hierarchy from a given point, in step with gross morphology, in every organism in my fridge.

    There is no expectation either way. Not based on Design. And not based on “evolution”.

    Allan Miller: You don’t refute gradualism by saying “what about worm-amoeba, huh?”.

    No. I wrote an OP that does that job.

    This is your OP, remember? Will you do your job or not?

    0
  40. Nonlin.org: You presume “change”. Which is the “theory” that you want to prove. This is circular reasoning. In fact, we are simply looking at DIFFERENCES. Which you have YET to confirm are due to “change”.

    You love your semantics, don’cha? If I create two organisms with a difference, I’d call that a ‘change’. Change is not a statement about descent.

    Me: The big surprise is that a gene that is not involved in morphology nonetheless follows the hierarchy built solely on morphology.
    Nonlin: That may be a surprise for you. But not to any designer that knows a thing or two. As I explained.

    It’s a surprise to me because I know a thing or two about genetics and biochemistry. It’s not a surprise to you because ‘any designer knows a thing or two’. Gosh, how very specific. What does ‘any designer’ know about performing the same specific low-level biochemical change in a variable manner with increasing morphological difference? What’s the specific design reason for this correlated variation?

    Yes, you tripped yourself with the analogy.

    No, I gave you an opportunity to misunderstand my point by over-extension

    But analogies cannot be ignored, so I would have tripped you.

    Of course they can. They are illustrative, not prescriptive.

    WHY? That’s an irrational expectation.

    Explain.

    I don’t see why I’m supposed to have an expectation of this random fact. It doesn’t confound anything.

    You just said you ‘expect’ these things to vary in line with morphology. Now you say you don’t. As I say, your expectation seems conditional on whatever turns up.

    There is no expectation either way. Not based on Design. And not based on “evolution”.

    There certainly is an expectation on (look everyone! I can’t say the word without coughing-up-a-hairball!) “evolution”.

    No. I wrote an OP that does that job.

    An irrelevancy here. Your OP was not responsive to my point about BLAST sequences.

    This is your OP, remember? Will you do your job or not?

    Have you noticed that I have been active under this OP? Comments headed ‘Allan Miller’ are from me. And I could have sworn you had responded to some of them, and I had come back … so what is it I’m not doing?

    0
  41. Nonlin.org: And anyway, this OP is supposed to somehow prove “evolution”, not disprove ID.

    Good grief. Can you not absorb the simplest points? Science does not deal in proofs. Science is about observing aspects of reality, developing hypotheses and testing them for accuracy.

    “Intelligent Design” is non-disprovable because there is no testable theory of ID.

    +1
  42. Nonlin.org,

    And anyway, this OP is supposed to somehow prove “evolution”, not disprove ID.

    This OP is examining an interesting phenomenon in which the vast majority of genes follow the morphological hierarchy (approximately), such that I can confidently allow a gene to be randomly picked as the start point and let BLAST do the rest. Whether it proves “evolution” (hairball!) or not is not the issue. Evolution is one possible explanation of the pattern, Design another. So we look to see which explanation best fits the pattern, based on the expectations each mechanism entails.

    It certainly seems, so far, that Design cannot do the job without a lot of special pleading. No particular rationale is offered for the correlation, other than handwaving and analogy, by contrast with evolution, for which it is a definite corollary.

    +1
  43. The relationship of MCEE to morphology can be investigated by looking at loss-of-function mutations. There is a rare condition methylmalonic aciduria, usually caused by mutations in methylmalonyl CoA mutase (a different gene). It’s quite serious, but a milder version associates with the complete truncation of MCEE, the gene we’re looking at here. A C->T mutation in exon 2 causes Arg->STOP at position 47, which happens to be the start of the VOC domain, which is also the less variable segment in the multiple alignments above, leading one to conclude it is the functional ‘business end’ of the protein.

    It only presents in homozygotes, so presumably the dosage of the full peptide in single copy is sufficient for normal activity. We might also suppose, since the condition is mild to moderate, that its function is weakly supported by a different gene, or spontaneous epimerisation.

    Be that as it may, both heterozygotes and homozygotes for this mutation are definitively, categorically human. So on this evidence, it is not involved in Linnaean distinctions.

    +1
  44. Nonlin.org: Climb a mountain and observe the biology change. You see transition to different organisms, not same organisms with slightly different characteristics. This disproved “gradualism”.

    Odd. Transition is defined thusly:

    the process or a period of changing from one state or condition to another.

    A transition does not include a line where on one side of that line you have X and on the other side you have Y. It’s a process. As you climb the mountain you see less of X and more of Y the higher you go.

    So you’ve actually just disproved your own claim. In your very own words you note that the change of distribution of organisms across niches can indeed be gradual. Otherwise there is no need for a a transition.

    +1

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.