The Discovery Institute doesn’t understand the protein folding problem. I mean that literally: they don’t understand the problem. Scientists don’t know the answer, but they have a clear understanding of the problem.
PNAS published a “Perspective” article, “The Nature of Protein Folding Pathways,” by S. Walter Englander and Leland Mayne. Unsurprisingly, they try to approach the problem from purely materialistic presuppositions. There is no mention of specificity, amino acid sequence, or digital information.
You see, there really is an interesting problem here: given a specific amino acid sequence in a protein, what 3-dimensional shape will the protein take? It turns out that’s really hard to calculate. We also can’t do it in reverse very well: given a certain desirable shape for a protein, what sequence will fold into that shape? There are so many interactions with the environment as the protein is assembled, that we can’t calculate them all. The paper takes a simplifying approach to estimate the shape by breaking it down into little modules called foldons. It’s still hard and only an approximation.
But here’s the thing: you can take any amino acid sequence — you can design your own, or you can assemble random amino acids — and it will fold. It will often even fold relatively reliably in certain ways, although proteins also demonstrate a misfolding rate. Why? Because it is the nature of proteins to be floppy and twist and bend until their energy is minimized. It is not surprising or unexpected that proteins fold. The challenge is to calculate in advance how it will spontaneously fold.
Let’s talk about the weather. It’s a similar problem; we know that moisture and temperature and air movements and topography will necessarily interact to produce a pattern of weather. We know that some generalities are reliable: Minnesota will be cold in January, Florida will be hot in July. That the weather is hard to predict does not imply that a god must be an imponderable variable in the equation — the complexity of the problem is sufficient to account for our difficulties.
The Discovery Institute does not understand any of this. Reading their summary is a bit like listening to Dark Age Scandinavians talk about the weather. Have you taken Thor into account?
While it is true that sophisticated techniques can see the protein fold as a series of stages, this ignores the main problem: How does DNA know in advance what sequence will build the first foldon, and the second, and the third, and so on, till a functional protein results?
But…but…how do the clouds know to make rain? How did the lightning know that Arnfrøðr had stolen Bjálfi’s goat, in order to strike him righteously?
DNA does not know anything. The protein will fold, following the rules of physics and chemistry. The DNA did not “know” in advance; protein sequences were selected after the fact for their functional effects, not built to spec.
We would like to ask these authors how DNA knew to code for chaperones that can help other proteins fold correctly. Those chaperones had to be coded in DNA such that they would fold correctly themselves first. And add another question: how were all the other molecular machines in the cell built that repair or dispatch misfolded proteins? What created the Golgi apparatus where proteins are finished, packaged, and delivered to their work sites?
How did Thor know to make Dyggvi’s shield arm strong, so he could defend Erik in the battle line?
It’s a kind of molecular retrospective coronation to decree that because X exists, X had to have been made present for a specific purpose. The whole collection of proteins and chaperones co-evolved; it’s a constraint on protein evolution that an amino acid change that leads to a non-functional folding would be as deleterious as a null mutation. And the pattern of that folding is a product of multiple components in the cell, so it’s not at all surprising that there are contingencies between multiple proteins.
I look at their questions, and I see a bunch of people who really have no grasp at all on the basic principles of evolutionary biology…who think that invoking Thor somehow simplifies the whole problem.
The information in DNA cuts through the vast conformational search space, and finds the route to the functional fold rapidly, sometimes within microseconds. ID is the only way to navigate a vast search space and arrive at a destination on time. Intelligent design can build self-organizing systems that obey the laws of thermodynamics, just as in the analogy of magnets on a string spontaneously folding into a wrench. Even if entropy increases as the protein folds, guiding the polypeptide through the funnel-shaped energy landscape to the final product, this does not negate the design that went into the system. It amplifies it.
The whole thing has this nonsensical attitude behind it, confusing cause and effect. It’s like assuming that since all the phenomena going on inside a tornado are beyond our ability to calculate precisely, somewhere there must a super-computer that is calculating exactly, moment by moment, the movement of every particle of dust, every broken bit of lumber, every flying cow in the vortex.
I hate to break the news to you, but tornadoes happened before the weather service had the capability to predict where they would appear. Knowledge isn’t causal in these cases.
Also, Yngvi is a louse!
GapGod is getting smaller and smaller, isn’t He? He’s way beyond fighting a giant spider with a pin…
Welp, they certainly know all of the words! That’s got to count for something, right?
Right?
Do these Dunning/Kruger victims even know that protein folding is nothing other than electrostatic attraction & repulsion?
That’s like asking how DNA knows how to fold. They aren’t rational thinkers, they are reactionaries.
How dare you call him a louse!
Signed
Yngvi’s mum
Funny, I just gave a talk on protein folding to a bunch of high-school kids and not a single one asked how a protein knows how to fold itself. Not a single one asked about Yngvi either…I missed my chance didn’t I?
What? Is this an Appeal to Speed argument?
Proteins fold according to intra- and intermolecular interactions (to which they will always be subject – what with being real objects). They don’t meaningfully search the entire conformational space as seems to be implied there – it is almost as though they have confused molecular dynamic simulation with real protein folding.
It is amusing to see that they think coding sequence so clearly dictates protein folding. If only that were the case; our researchers wouldn’t spend so much time having to optimise expression systems!
I think they have also failed to understand the generic nature of many chaperone systems, the function of proteosomic and other degradation pathways, or recognise that misfolding is not only a theoretical problem, but a significant factor in many disease pathologies.
When I was little my mom used to fold my DNA right after ironing it.
Does the designer have proteins, or protein-equivalents, and what folded them?
I don’t know what the Compleat Enchanter reference is for but I approve.
So, they’re proposing that proteins are somehow sentient? Because they “know” how to fold?
Egad…too much palm, not enough face.
Speaking of “on time”, I wonder how that Wedge Document timetable is coming along.
(Wedge Docuemnt Phase I continued)
I was shocked too when I saw this on the DI a day or 2 ago. I think this was posted by O’Leary and its so looney more serious IDers such as Behe would probably be quick to disavow it. Its a shame this wasnt brought up in a live debate and immortalized on Youtbube because what she’s saying here is the God folds each protein himself miraculously. Its not an exaggeration to say this is Medieval thinking applied to biochemistry.
Is there any reason to think it doesn’t?
Huh?? You increase entropy, and that means you’re “amplifying” a “design”?? Is that supposed to be a joke?
@#15:
It’s just that certain ones are getting really creative with the gobbledygook – but highly creative gobbledygook remains gobbledygook. C.f.: the pig and lipstick syndrome.
If you are familiar with the novels by Richard Kadrey, you know it was the Angra Om Ya, the original creators of the universe who sorted out the whole protein folding business.
The clepto-god (Jahweh) stole the universe from them, but never really worked out how to run it without the manual. That is why everything is such a mess.
“Thermos, hot or cold, how does it know”? — Stephen Wright
Cursed magnets!
The DI doesn’t understand the problem. But they know the solution.
Jeffrey Shallit reports on a new ID fake journal:
I See Berlinskis
It’s been twenty years since my microbiology day. Aren’t a lot of proteins folded with the assistance of cell membranes, not completely self folding?
Define “specificity” and “information” for us, along with some example values for some actual object, and then we’ll talk. Until then, its just gibberish pretending to be words.
There can be no greater indication of ignorance than asking how an inanimate object “knows” something.
Sounds like something they came up with while watching a turd go down the toilet, which is as apt a metaphor as I can imagine.
Gosh, it all sounds so silly when analyzed by someone who actually knows what he’s talking about.
Big deal. If self-organizing systems were disobeying the laws of thermodynamics, then maybe I’d think something was up.
The DI’s idiocy is like holding a shoelace vertically, dropping it, and claiming the resulting pattern is only possible due to the work of Satan.
Or in my case, the cat. :)
I’m reminded of a ramble by Gamzee Makara (Let me tell you about Homestuck!) wondering how his bottle of Faygo knows when to hiss upon opening, declaring it a miracle, and getting irate whenever someone proposes explanations for his “miracles.”
Naturally, he’s a reference to Insane Clown Posse, though IIRC, he declared the “f’ing magnets, how do they work?” video to be blasphemous, possibly because even he felt insulted by them.
Protein folding is an interesting problem, but I wonder if the modelers have taken into account that proteins are made beginning from one end, so the protein folds as it is being made. That should greatly limit the number of possible shapes that it can make.
How did Thor know to make Dyggvi’s shield arm strong, so he could defend Erik in the battle line?
He didn’t, you silly poopyhead: that’s Tyr (aka Tiw)’s job!
@marcoli
It is really is much more complicated than that, although of course the primary structure (amino acid sequence) constrains the structure to some extent. Proteins don’t simply fold during translation, and that is that.
– Structures may be dependent on the formation of homo- or hetromers made up of multiple peptide subunits, and intra or intramolecular interactions (disulphide bond formation)
– Mature proteins may undergo post-translational modification (phosphorylation, sumoylation, methylation, ubiquitination, proteolytic cleavage, glycoslylation, intramolecular rearrangment), enzymes may form internal cofactors (e.g. MIO) or incorporate cofactors and prosthetic groups (haem, metals, phosphopantetheine)
– Proteins change structure when bound to different ligands (this is often the mechanism for activation of enzymes)
– Protein structures may be more or less ordered – different regions of the protein may be more or less rigid, and function is often related to rigidity (e.g. lipases, amino acid lyases/mutases).
Even predicting the fold of a single, relatively lightly modified peptide is difficult – hence the protein folding problem.
In reality, protein structures are highly dynamic, and highly dependent on the local environment – pH, ionic potential, temperature, buffer, etc. The fold may change radically as the subcellular localisation of the protein changes. In some respects, the often necessary simplification of thinking about a protein structure is very misleading, and often at odds with the reality of the functional protein.
We often produce beautiful model structures based on x-ray diffraction data, and the reality may be that the very definitive structure depicted in the model is actually rather poorly resolved electron density in the original data – like a blurry photo. Some parts of a protein may never be solved in a physiological context, because they may be inherently dynamic.
They wouldn’t dare! They’d find themselves up before the High Court of Entropy before you could say “Sadi Carnot”!
One of the major problems that these IDiots suffer from is the idea that natural phenomena are rational. Nothing could be further from the truth, of course, but they must have that explanation. Otherwise their crackpot theories about “Intelligent Design” are proven to be false. But then, their ultimate answer is “It must be true, because the Bible says so!” I am endlessly amused by those who mistake Faith for Reason. Not the same thing at all.
Randall Munroe weighed in on protein folding recently, in a way that really cleasrly explains the magnitude of the problem.
My computer has been folding virtual proteins for several years now, as part of a Stanford University distributed computing project. You should check it out.
I have three projects active, Seti@home, Einstein@home, and Rosetta@home. Looking for ET, neutron stars via gravity waves, and a protein folding project.
Wow.
I don’t know if most people appreciate just how off the mark the ID paper is. It’s not that the ID paper is wrong. It’s not even wrong.
http://rationalwiki.org/wiki/Not_even_wrong
When the non-ID paper uses the term “search space”, it is being used in the computer science sense. It’s talking about the difficulty of writing a computer program on available hardware to model the folding of proteins. “Search space” here refers to splitting the problem up into smaller chunks, which allows for parallelization of the computer program. The ID paper thinks that the non-ID paper is using the term “search space” to talk about evolution, random mutation, and natural selection searching the space of all amino acid sequences to find functional proteins!
When the non-ID paper uses the term “thermodynamic equilibrium”, it refers to the processes at work during the folding of the protein. The protein “wants” to achieve the minimal energy state, just like a ball rolling down a hill. It’s a description of basic particle physics involved in protein folding, something that a high school student of physics should understand. The ID paper thinks that the non-ID paper is using the term “thermodynamic equilibrium” to refer to entropy and information, and to the information content of the genome or protein in terms of the information content of the amino acid sequence, and how that information could not possibly have come around by chance [thermodynamics]!
See:
This is like if the ID people responded to a computer science paper on P = NP, and wrote that the computer scientists are overly complicating a very simple problem, and that the obvious answer is N = 1. (A particularly astute version would note that P = 0 is another solution. /sarcasm) The ID paper is not right. It’s not even wrong. It fails the criteria of being right or wrong by being a complete non-sequitir by completely failing to understand the terms of the discussion and using the terms in a completely different meaning.
Also see:
I’d be willing to bet that in the non-ID paper, when they say “folding in moderately small […] may be necessary for proteins to fold at all”, they meant to say “folding in moderately small […] units will be a necessary design decision in any plausible computer program to fold virtual proteins on a computer”. Similarly, when the non-ID paper says “viable choice”, it refers to viable choices of possible computer programs they could write to model protein folding – not referring to the “choice” of which model is correct to explain the diversity of life!
I really don’t think that the IDers are saying that god is personally involved in every single instance of protein folding. I saw that suggested in several comments here. I don’t think that’s right at all. I really do think that the IDers are so incompetent that they completely failed to understand a whit of the non-ID paper.
lightning sorts tghrough a vast computational space to choose which path trhough the air it will take! It does this in microseconds!
Snark aside, they are seemingly forgetting that DNA is already coded. It doesn’t have to think on the spot for what protein to make. The design is already coded. Even if we assume an intelligent designer, it could have taken millions of years for that intelligent designer to figure out how to do it and we wouldn’t know the difference, because they get to code it to happen every time after that. They don’t need to intelligently figure it out in microseconds every time the protein is made. Sheesh these people.
@35, Enlightenment Liberal
hmmm, I dunno:
They seem to say that the intelligent designer would need to be in a similar hurry, which only makes sense if the designer has to do it each time. At least I can’t make any other sense of that phrase…what other reason would the designer have to be in a microseconds-long hurry?
@brianpansky
Again, I believe the ID paper is based on a complete misunderstanding. I believe that quote to be the IDers saying that ID is the only way for evolution, mutation, and natural selection to navigate the vast search space of strings of amino acid to find functional proteins. The IDers are imaging an intelligent designer looking at strings of amino acids, and determining which way the proteins fold, and choosing which protein to use. The IDers (wrongly) understand the non-ID paper to be talking about a naturalistic alternative to explain how the amino acid strings we have properly fold into functional proteins. I think the ID paper is really that far off the mark. The ID paper makes a lot more sense IMHO in that reading.
@38
ah, they mistake the paper as saying “how can nature search quickly” instead of what it actually says, which is “how can we make a computer quickly predict what nature will do?”
…Though I’m not sure that actually rules out the possibility that they are proposing microsecond interventions anyways.
I love the Harold Shea books – thanks for the reference- quite made my morning!
The I.D.ers don’t think the protein knows what shape is required- because that would be silly. it’s the DNA that knows!
DNA is actually an incredibly thin but very clever worm.
Thread won.
Day saved.
@brianpansky
Indeed. Hard to tell.