One of the densest Creationist tropes has to be ‘Common Design’. It is proposed as a direct competitor to Common Descent – template mediated copying of DNA – as an explanation for the high sequence similarity of two DNA segments. But what is actually held in common? If we look at a particular transposon sequence, and find it is in A and B but not C, and another that is in A but not B, etc, we can generally organise a set of such markers into a ‘tree’ structure, much as would be predicted by Common Descent. But no, we are assured that these apparent markers are in fact part of the ‘design’. If A is a whale, B a pig and C a deer, there is something that is vital for the function of both whale and pig but is definitely not required in deer. Instead, a sequence which, in whale and pig, sits either side of the insertion, runs uninterrupted in the deer. That, too, is functional, supposedly, even though the insert would give a product which was the A/B one with a gap and possibly a frameshift, if it were transcribed.
But this is held to be the case even if the sequence, with and without transposon, is never transcribed. A sequence that does nothing, and organises hierarchically exactly as would be expected of common descent, is nonetheless functional … because?
On observation, there must be some genome pairs that are highly similar because they are commonly descended. We can see it happening. But there are, on this notion, supposed to be identically-patterened runs of similarity that are not due to common descent, but instead result from a completely different cause – some entity bolting together genomes, or parts thereof, from scratch, I guess, and choosing to repeat a known pattern – up to a point – in a manner that fools our most adept molecular taxonomists into seeing descent.
There must be a line in a taxonomy where the one shades into the other – on one side, sequence commonality is all Common Descent; on the other, Common Design. Where does this discontinuity reside? Species, genus, family, order? Is it a gradual transition, gene by gene, or all at once? How could you tell? Why does it not show up in computer analyses of blind datasets?
If I were to provide 3 genomes shorn of all differences, it would be impossible to tell which were commonly descended and which commonly designed from the data. But there must logically be a transition of causes, were I to take the genomes from sufficiently distant species and this idea were true. What persuades us to adopt this causal explanation in preference to that which explains the pattern better: Common Descent?
Transcription isn’t everything, as shown with the LINE-1. But also there are scaffolding and docking sites for molecular machines. Histone tails can serve as docking sites, so until we actually see the histone profiles we won’t know for sure will we?
Robert Tjian points out every cell type has different regulatory mechanisms and binding sites for it’s regulatory machines. It’s not too much of a stretch to say it is different on a species basis. That would break the supposed C-value paradox, and it would also destroy common descent if so.
I’m hopeful that will be the case because over the last decades, it’s been a bad bet to accept evolutionary hypotheses as much of an explanation for the molecular and biochemical activity — molecular biology and biochemistry are much better qualified disciplines than evolutionary biology to answer questions about biological function.
“If ENCODE is right, evolution is wrong.” My bet is ENCODE is right, and beyond that, not just about what is transcribed, but also what isn’t. DNA acts as a scaffold, and the extra DNA is important for the 3D higher order chromatin structures. I pointed out one example of this with untranscribed CTFC binding sites. Who knows what else we’ll find as we learn more.
See:
One thing about evolutionary theory. It’s quick to make assertions about molecular biology as if they are facts only to get embarrassed after appropriate experiments and measurements are carried out.
The list of failures continues to grow…
That doesn’t follow.
The bottom line is that evolutionary mechanisms explore the sequence space around functional sequences. It is observed that those functional sequences are connected to other functional sequences. The size of the overall sequence space is therefore irrelevant. There is no rational basis for colewd’s incredulity based as it is on the size of the sequence space.
Agreed, the ways you continue to fail with your silly anti-science rhetorical games does keep growing. 🙂
Is it? How do you know>
stcordova,
It really doesn’t matter Sal. I know you’d love to make this about something else, but the Common Descent/Common Design argument does not turn on whether a sequence is functional or not (though do you seriously think that there is not a single marker in any genome anywhere that is non-functional?)
But yeah, let’s talk about function. What possible function could cause pigs, hippos and whales to need a SINE copy in position x of their genomes, but deer to have the flanking sequence alone? And another to need a SINE at position y in whales and hippos but not in deer or pigs? And another … I mean, I know there’s always ‘you never know’, but this is far too vague to be taken seriously as an objection to common descent as the preferred explanation. Whether functional or not.
Frankie,
Their movements are constrained to the portions of the genome they can get at. That does not make them guided, any more than a river is. Note that they also cause a substantial number of diseases. If that is the designer at work, I wish he’d fuck off. Him and his viruses and flagellated pathogens.
Mung,
This is what continually puzzles me. I don’t know why one would expect commonly descended features to be identical anyway, but if they are not, it is hardly as if Common Design pops into the head as the obvious reason they are different.
So taking arguments to extremes, the acme of Common Design is either those ‘identical’ twins, or two organisms that could not be more different.
Neither makes sense.
And as shown by every regulatory sequence. Are you confusing non-coding DNA with junk DNA again?
Won’t know what? What will the histone profiles show us?
Please make smaller intuitive leaps if you want to be comprehensible. Cite your sources; “Robert Tjian” is not a sufficient citation. Yes, it is way too much of a stretch. How would it break the C-value paradox, and how would it destroy common descent? Show your work.
This setting up of evolutionary and molecular biology as adversaries is all in your head.
As has been pointed out several times, your quote mine here is mistaken. If ENCODE is right, some of the things we know about population genetics are wrong. Common descent is untouched.
Your triumphalism is both misplaced and unseemly.
Mung,
It would still be the most robust conclusion. Clones, possibly, but more likely twins, and the point against Common Design remains unchanged by this gotcha.
If I’d said “If you were given two genomes differing at only 200 sites out of 6.5 billion (99.99997% identity, being generous about the expected amount of difference), I think the most robust conclusion would be that they were identical twins, not that they had been separately created.”, you might be able to figure out for yourself whether the argument is changed by that difference between 99.99997% and 100%. [eta – I’m not even sure 200 sites is above the margin of error for sequencing].
You really need to move beyond the gnomic and tell me which link(s) you’ve had a problem with. I just click ‘Reply’, which I am aware does not work over the page.
“Common design,” besides being an obvious rip-off of “common descent” (and it’s an old creationist term, not beginning with Sarfati), gives away the game by the modifier “common.”
I mean, why isn’t it just “design”? Because, of course, design isn’t so heavily derivative of other apparently related organisms, nor is it so blind to what happens in separate lineages, and naturally, hardly exhibiting the branching hierarchies that organisms do.
If design explained it, they’d simply invoke design. “Common design” just attempts to take over commonality for design by making a similar-sounding term (never mind that many creationists believe it–I did when a child). But if design were responsible, “common” simply wouldn’t have to modify “design” at all.
Glen Davidson
Mung,
Oh, and another thing …
That was not the claim made. The claim was that, if you were presented with two identical genome sequences, you would be safer concluding ‘twins’ than ‘common design’.
One dodge I notice, which Sarfati does too, is to say, when talking of DNA evidence ‘oh, it’s expected that commonly designed organisms would have similar biochemistry’. This misses the mark by a mile – as well as being a curious retreat by a bunch that spend post after post in other arenas telling us how digital life is.
In phylogeny, it’s not that DNA is a ‘biochemical’ that matters. There are scratches and scuffs that make no difference upon it, even in ‘functional’ sequence. And these scratches and scuffs arrange hierarchically, just as functional sequence does. One would have to insist that there is no such thing, not just as non-functional DNA, but no such thing as a ‘silent’ deletion, insertion, inversion, SNP, transposition (both TE and non-TE associated), duplication, copy number variant, karyotype change …
This would clearly be hogwash.
I’ve cited him several times at TSZ. Guys like you ignore it, you don’t bother to learn and reconsider. What’s the point?
Is this the Robert Tjian you’r referring to? I’ve not heard of him before but he looks a perfectly respectable research scientist. What could he have possibly published that supports YEC?
That’s a good one Sal. I can see you trying to publish a scientific paper with the references “I’ve talked about them before at other times so I don’t have to cite them properly now”.
John Harshman,
John,
I had a chance to briefly look at the Theobald paper.
If Doug was able to make a probably calculation of the different scenarios do you agree that a probabilistic mechanism (RMNS and neutral mutations) was assumed for either UCD or separate origins comparative analysis?
The way the regulatory machinery parks and works differs by cell type as Robert Tjian demonstrated.
But here is a paper anyway (that will probably be ignored by anti-complexity crowd at TSZ):
http://europepmc.org/articles/PMC3660745
[There is that hated term again “epigenetic”.]
Below is the D4Z4 repetitive element serving as a docking/binding/parking site for Poly Comb Repression groups (PcG). It’s not too much of a stretch to think other kinds of molecular machines could park onto transposons.
How does the PcG know specifically where to park? DNA patterns, histone tail patterns, DNA methylations? Who knows. Too early to tell.
http://www.cell.com/cell/abstract/S0092-8674(12)00463-1
Regulation isn’t as cut and dry, nor as well understood as evolutionist represent it to be. They wouldn’t want to accept the possibility suggested by the actual data that different cell types and species regulate the way they do business because they have differing repetitive elements (aka repetitive junk DNA).
The below diagram shows regulatory machinery that is only proximal in a 3-dimensional sense, not in a linear sense on the DNA strand.
Bottom line, with respect to the OP:
As demonstrated several times at TSZ, what was once thought to be a do-nothing piece of DNA, is functional, and it won’t be uncovered easily by simple knockout experiments many times because of redundancy of function in the genome.
I’ve already listed 27% of the human genome that are transposons that are now believed to have function (11% Alu, 17% LINE-1). So it’s not too much of a stretch to say that is probably the case for other transposons.
Typical evolutionary simplistic, jump-to-conclusions-in-favor-forgone-conclusions thinking. Not actual skeptical, critical thinking and molecular biological investigation.
If you aren’t interested in making yourself comprehensible, why bother to post at all?
But then you link to a paper by someone else. It smacks of a Gish gallop. If Tjian has published something you think supports a view you are promoting, why not link to it and explain the connection? If not, why mention Tjian at all?
Not quite. He assumed a particular model of evolution on branching trees, but the model wasn’t used for calculations when there were no branches, or when trees were not connected. If you want to claim that a different model would be more reasonable, you should propose one. What do you mean by RMNS?
John Harshman,
Random mutation and natural selection.
Do you know what model he assumed?
As I read it, they find that HOX protein binding is influenced both by the nearby binding of other transcription factors specific to the cell type and by epigenetic marks that make chromosomal segments more or less exposed to binding. This is not at all surprising and is seemingly unrelated to whatever your point was.
Nobody hates the term. It’s a standard term in biology. Cell differentiation is an epigenetic phenomenon. What else could it be?
Look in the Methods section under Likelihood Phylogenetics and Bayesian Phylogenetics. He tested a vast number of models, each consisting of a particular amino acid transition matrix and presence or absence of parameters for number of invariant sites, gamma-distributed rate classes, and empirically determined amino acid frequencies. A full explanation of what all that means would take quite a while (you might wish to consult Joe Felsenstein’s book Inferring Phylogenies), but I think the main point is that he bracketed a great number of possibilities, none of which made much difference to the results.
Frankie,
newton: “For ID it could be one step from knuckle draggers to humans.Given the right designer ,anything is possible.”
FrankenJoe: “Is it? How do you know”
We see this kind of leap all the time in human design. And since ID is nothing more than a lame extrapolation from human designed structures to biological structures, it must be true.
John Harshman,
The above says that if archaea, eukaryotic, and bacteria share a common ancestor then the closest multi origin path which is archaea and eukaryotic sharing a common ancestor and bacteria evolving on its own.
Does it make sense that this would be 10^2860 less likely.
Does this say that the probability of evolving bacteria is at least 10^2860 since it should be easier to evolve archaea and eukaryotic than all 3.
Well, no Allan, it wouldn’t necessarily clearly be hogwash to suggest that we really don’t know what all of the minor variation of DNA means, if we still don’t fully understand how the whole process came to be anyway. You call them minor scrapes and scuffs that are meaningless, yet every honest biologist will tell you that what we totally understand about the human (or other organisms) genome is just a fraction of all that is taking place (I recently read one biologist state that we only probably know about 10% of what is really going on.)
We still need an explanation for a whole host of processes in the development of life. To say that these scrapes and scuffs have no role in that is a completely premature and unsupported assertion. Their roles might not be obvious now, but it may have been important in their development.
For example, where does our innate knowledge exist in our body? We have ZERO accounting for this. Where does our ability to know how to move our body, our ability to know that hunger means use your mouth to satisfy it, our knowledge to move away from pain, for animals to know ANYTHING. We are hardware that comes pre-programmed with software. WHERE does that software reside. You have no idea. No one has any idea. Why don’t humans attempt to flap their wings and fly? Why don’t they attempt to use their nose to borough a hole in the ground? Or sting you with a poison tail (Ok fine, Richard probably attempts that, but still…)?
phoodoo,
This is why we should never censor Phoodoo. Phoodoo is an ID proponent.
Also, how come cats eyes are exactly where their eyeholes are? Notice your legs are the exact distance between you and the floor?
colewd,
Bill,
Sorry, but that was word salad. I can’t respond to it.
Bill: “Does this say that the probability of evolving bacteria is at least 10^2860 since it should be easier to evolve archaea and eukaryotic than all 3.”
You say something as stupid as this and expect to be taken seriously?
The entire probability argument is a non starter. If you don’t think so, try calculating the probability of someone with your exact genome existing given a starting point of ten generations ago. Do the math. It is astronomically improbable.
This makes absolutely no sense whatsoever. Every cell has the same DNA, they just don’t all have the same expression patterns, meaning they don’t all express the same genes or in equal amounts or at identical timings (that is, after all, why they are different cell-types). Nevertheless, the principle of regulation is the same across cell-types.
The same is true for species. Gene regulation happens by the same principles of binding certain DNA sequences and initiating or inhibiting additional activity. The binding-sites (and the regulatory elements that bind to them) diverge over time through accumulation of mutations too. None of this has any effect on the C-value paradox.
You keep trying to make up this revisionist history that evolutionary biology has mindlessly dismissed pretty much everything as junk and that as a result of this, researchers aren’t bothering trying to find out what some putative functional element does. I’m sorry but this isn’t something that takes place.
No, that’s not what that says.
To say that X is 10^2860 times more likely than Y, is not to say that X has a probability of occurring of 10^2860. It is a relative, not absolute measure of probability.
To say that X is more likely than Y, does not tell you how likely X is. All it says is, however unlikely X might be, Y is even worse.
So however unlikely you think common descent is, independent origins is even worse. Notice that independent origins is the set of hypotheses that contains design. So if you think evolving bacteria, archaea and eukaryotes from a common ancestor is very unlikely, the prior probability of designing them with their particular properties is even less likely.
The problem with probability arguments is they never come up with a probability to compare it to. They say silly shit like “evolving the E coli bacterium with it’s particular genome and properties has a probability of 1 in 10 to the power of fifty million”. Let’s just pretend that’s true, now we need to compare it to the probability of alternative hypotheses. What is the prior probability of designing the E coli bacterium with it’s particuar genome and properties?
You can’t just state the prior probability of a single hypothesis and pretend your work is done. You need at least two prior probabilities to compare to each other before you can choose among them. One thing having a low probability does not mean another thing you don’t even know the probabilit of, is automatically true. It simply doesn’t follow. These lunatics wouldn’t recognize logic if it bored itself into their retina.
Or perhaps you are not fully understanding. However I’m sure that won’t make any difference to your confident dismissals.
stcordova,
Come on, anti-complexity crowd, make youselves known! Who are you, and what has complexity ever done to you? Me, I love a bit of complexity. Although in Design, simple is often more elegant.
phoodoo,
This whole ‘you never know’ schtick is beyond dumb. We will never ‘know’ with absolute 100% certainty, but this is no reason to throw out common descent because common descent does not depend on non-functionality anyway.
But even so, it would be ridiculous to suppose that every bit in DNA has to be that way. Do you know of a single for-sure designed system which cannot be subject to any change whatsoever without fundamentally changing function? Of course not. So why is biology supposed different, when it is actually much more tolerant of perturbation than anything we design?
Hogwash, phoodoo. Pure, unadulterated hogwash.
By the way, the probability of something can’t be greater than 1. 1 means it is 100% certain that it will occur. It can’t get higher than that.
I think what colewd is hoping is that, since Theobald used a probabilistic analysis of models, that this is related to a probability attached to the origination of the domains. They aren’t related, and the latter is definitely beyond computation. Theobald is examining change after one or more LCAs to see which model best fits the data. This is not related to the prior probability of a LUCA from a standing start (which I think colewd still thinks of as the first ever cell).
QFT
Except that common design came before common descent. Whoopsie
No, it turns on the fact that Common Descent cannot account for the anatomical and physiological differences observed. And without that you don’t have any science to support the claim of Common Descent.
And how does blind and mindless process account for HOX genes and transcription factors? How can we test the claim that said processes produced them?
That explains why I had nothing to add on the topic of there being no theory of evolution? What I am supposed to add, another NOT theory?
You seem to have nothing to add about where knowledge is stored in the genome, and about how you know that non-coding DNA has no function.
That is a completely uneducated and unfounded position. No wonder you don’t defend it.
Yeah only if it can be achieved and still get the result the designer wanted.
Correcting a mistake:
Just because two different teams came up with two similar yet different methodologies to measure the same thing doesn’t mean they aren’t measuring the same thing nor does it mean that CSI (biology) is NOT the same thing as FSC.
Exactly, what colewd wants is prior probabilities. He thinks the prior probability of the data we have (the organisms that exist) is very low, so therefore it must be design.
He’s never even tried to so much as get at the prior probability of the data we have on design.
Frankie,
Of course it doesn’t. Molecular analysis of Common Descent requires no input from anatomy or physiology. One has already moved some distance away from impossible simply by noting the commonality of sequence, and the ‘tree-like’ pattern of datasets with more than 2 members.
If it is indeed impossible, still ‘Design’ does not address the commonality. If we exclude Descent on your rather spurious grounds – it is only accepted for transitions we can imagine – we are left with a puzzle and no conceivable explanation, having discarded a perfectly valid one.
Nor, for that matter, has anyone even gone even two inches towards showing that badgers or bears – say – can physically be designed.
You’re absolutely right – though that is somewhat conditioned by that to which I am responding. I try to vary the words a bit though, unlike yourself.
I’ve mentioned him several times at TSZ. I assumed if you guys really were paying attention, it would be common knowledge by now. Goes to show you guys ignore some of the most important points by some of the renowned experts in the field.
Tjian is president of Howard Hughes Medical Institute and a 30-year professor of biochemistry at UC Berkeley.
I posted this at TSZ before. It was probably one of the most important points. Did it get any serious discussion? No. Just ignored. And then I get accused of not supporting my points. If I repeat it, as I did several times, I get accused of being repetitious and spamming. Far be it for any of the regulars to admit I actually introduced data points from an every changing field they may not have been aware of.
A lot of evolutionary ideas are being argued by obsolete viewpoints of biochemistry and molecular biology. Allan’s OP shows a bias to such obsolete viewpoints:
That is a presumptuous viewpoint. No one knows for sure, and lots of sequences thought not to do anything, like repetitive elements, actual do something. We find out after decades of experiments. The D4Z4 repeat is one example of DNA repetitive elements acting as a scaffold. This has relevance to Allan’s point.
Below is picture from one of Tjian’s lectures. Students at the NIH who watched his video think its awesome stuff. Did anyone here bother to even watch the video I linked to by Tjian?
Do I have any reasonable expectation people are going to spend 45 minutes trying to learn from the jedi master?
I provided links to the video several times to challenge the problem of homology (or lack thereof) in the regulatory networks of various cell types. It has relevance to the potential lack of homology in regulatory networks between species. Does anyone here really care to go into these details?
Anyway here is the link to the video in question:
The question is what point are you making by mentioning his name. Link to a specific paper. Explain how something in the paper supports some argument you are making. We are not mind-readers.
Moved a couple of comments to guano.
Fair enough old friend. I will try. 🙂
Allan’s OP is premised on this claim:
I’ve argued the assumption of the “sequence does nothing” is premature. I’ve said until we know the profiles of the histones around which these supposedly-do-nothing sequences are wrapped, we really don’t know whether they indeed do nothing!
Allan said this:
There is not one way of doing things. A whale, a pig, and a deer can regulate gene expression differently! The actual variations in DNA (variations which Allan argues “do nothing”) may actually do something!
I specifically mentioned histones in a comment above. John Harshman said I made no sense. Is his problem that he doesn’t understand the relevance of histones and binding of molecular machines or complexes to histones? The problem could be his lack of familiarity with the topic, not my lack of clarity.
Here is a paper that supports some of the claims I made in this thread. It is technical, it is not easy reading. It was published in 2013.
http://www.cell.com/molecular-cell/abstract/S1097-2765(12)00944-6
It shows again that the view that cell-types and species regulate genes the same way is antiquated. This also specifically addresses some of the many roles of transposons which evolutionary biologists have argued are junk. Now that molecular biologists are destroying that viewpoint, arguments like that in the OP are shown to be premature at best, maybe dead wrong. Why the rush to judgement? Isn’t science about be cautious rather than bold about claims of the unknown? Where is the skepticism and critical thinking?
There are lots of similarities in gene regulation between species and the cell-types within species, but there are substantial differences, enough differences that one can’t make a sweeping claim the differences have no consequence or in Allan Miller’s words:
I’ve shown sequences don’t have to be transcribed to be important. The act as regulatory scaffolds, parking lots, road signs for molecular machines. In the case of other transposons they generate A-to-I editable RNAs or change the somatic genome. The list is endless of how these things are known and will be discovered how to work.
NOTE:
Larry Moran teaches at University of Toronto. He is in the minority of professors with his negative view of functionality of the genome. One of the co-authors of the above article is from University of Toronto.
“Michael D. Wilson — SickKids Research Institute and Department of Molecular Genetics, University of Toronto, Toronto.”