I’m not sure what you mean, here…
You didn’t mention switching between structures.
Most? No, the norm is much more subtle–slight alterations in function. You might want to look at some data, like ours:
Note that none of these cardiomyopathy-causing TPM1 mutations caused any loss of structure. In fact, all made the coiled-coil structure tighter, despite having opposite effects on contractility. All that’s really changing is the equilibrium between the metastable structures. That’s enough, in the wrong genetic background, to be lethal.
I did, very clearly. Why did you not quote my explanation?
That’s a testable hypothesis, as is my hypothesis that those folds are common in sequence space. Can you see how?
Again, that’s testable. Have you done so?
It is, as are secondary and quaternary structures. Do you have a point that you are trying to make?
I am temporarily closing the thread for reasons. It will reopen when after a cooling-off break…
Me not mentioning switching between structures doesn’t mean what I did say, contradicts what you said.
I am surprised that you think this somehow undermines the general observation(which is a real observation) that in the absence of purifying selection it takes relatively few mutations to make a protein lose it’s structure, and through that the function performed by that structure.
I also don’t see how it follows that because you characterized those 7 single substitutions out of the 5396 possible ones in α-tropomyosin, that don’t make it lose it’s structure, this should imply that it is not the case that most proteins that lose their structures also lose their functions.
I didn’t quote it because it’s clear that it wasn’t an explanation at all.
You making this statement:
“It’s a structural, not a functional, classification”
-doesn’t explain how me making this statement:
“And it is probably also accurate to say any specific protein fold is rare in sequence space, within a very large range of “rare”.”
-means I’m adopting Axe’s use of the term fold.
It just doesn’t. Neither does you linking me the SCOP database explain how I am adopting Axe’s use of the term fold. You’re going to have to pick out what I actually say and then do the explaining part.
Yes, and that evidence shows that what I say is absolutely correct. Most proteins are multifunctional as I’m sure you’re aware, and a very low frequency of protein sequences out of the total space adopt a particular structure:
You can take a look at the numbers in table 1. None of which can be said to constitute “incredibly common” by any reasonable understanding of that term. The most frequent number found is 10-24. One in ten to the twentyfourth power is more accurately described as unfathomably rare.
As they also write (my bold):
On the other hand, the relative sequence capacity, i.e., normalized by the total number of possible sequences, is an extremely tiny number and is strongly anti-correlated with the protein length. Thus, although there may be more foldable sequences for larger proteins, it will be much harder to find them.
We can then keep that in mind when looking at figure 5:
This has not escaped the notice of the authors, who write in the conclusion:
Our SC estimates for the CATH database enable us to estimate the total SC of the known universe of protein structures, and to correlate the SC of a fold with its evolutionary age. We find that more recently evolved proteins have higher SC∗, which may be an advantage for initial discovery of a folded structure, but that more ancient proteins have a higher absolute SC, suggesting that evolution guides proteins toward more designable structures.
Believe it or not, but I’m actually quite familiar with the state of knowledge in this field.
There you go again. Functions are usually mediated by transitions between structures. Structural proteins are an exception, but their stability makes them prime disease/aging candidates, as in glycation of collagen.
Because they represent real-world variations, detected only because they rarely cause disease, in a system that is still rapidly evolving. That evolution appears to have been a huge factor in differentiating us from the other apes.
It does, because Axe is selling the falsehood that a stable structure is what evolution is trying to find, and that folds correspond to functions. Look at what you quoted below from the simulation paper, and:
“function performed by that structure”
“most proteins that lose their structures also lose their functions”
A simulation is not evidence, and the authors are making the same false assumption that you and Axe are: that more stability corresponds to more function. They are taking still photographs, while the real world is a frenetic movie.
The numbers in Table 1 are from a simulation, not evidence. It’s also easily fathomable, given the numbers of sequences that are easily surveyed by variation in nature.
So what? There’s nothing about function there. Nothing.
I can’t believe that I have to point out that what people write about the data is not the data. It’s not even the simulated data in this case. Evolution isn’t trying to find anything.
It doesn’t appear that way to me, since you’re repeatedly and vehemently denying the importance of structural instability in both normal function and preventing disease. I’m quite familiar with the limitations of in silico structure/function predictions, which are particularly acute in the muscle field; while they can be useful, I would never describe them as evidence.
You’re not saying anything that contradicts what I say. It’s still absolutely true that it takes relatively few mutations (in the absence of selection) to make a protein lose it’s structure and the function performed by that structure.
So what? How does it follow from them being real-world variations detected because they rarely cause disease that this should imply that it is not the case that most proteins that lose their structures also lose their functions?
It doesn’t. It doesn’t follow at all. It’s a totally irrelevant fact.
How is that relevant to what I say? Again, look at what I write. Not what you think I’m trying to imply, but what the words I write actually mean.
Me writing “And it is probably also accurate to say any specific protein fold is rare in sequence space, within a very large range of “rare” doesn’t mean I think evolution is trying to find stable structures, or that functions have to be carried out by some specific fold. All I’m saying is particular folds(meaning a specific one, pick one from the SCOP) are very rare in sequence space, which is to say it is only some extremely tiny minority of randomly picked protein sequences that adopt that structure. And to the extend that structure performs some function because of that structure, the loss of that structure (which takes relatively few mutations in the absence of selection) entails the loss of that function too.
That doesn’t mean the function itself is necessarily rare in sequence space too(other structures might be able to perform it, which there’s definitely reason to think), but the structure is.
But I’m not saying that folds correspond to functions. What I am saying is that there are structures that perform functions, and if you take a protein that has that structure, and if that structure is critical to a function that protein performs, then mutations that destroy that structure likely also destroys the function it performs.
I’m looking at it. It’s completely correct. It’s you who is going further than what I actually write and taking me to imply something I’m not actually stating. Not me. I’m not writing what you seem to take me to imply and it doesn’t follow from what I write.
Of course it is. What an utterly ridiculous claim to make. To the extend a simulation includes realistic parameters and/or good approximations of them, and/or it can produce results that agree with real data, it is definitely a form of evidence. It can even teach you something about the phenomenon in question that might be difficult to observe experimentally and help guide the establishment of new models, future research, and experiments.
No they absolutely don’t. They note that there is a correlation between stability of folding proteins and protein fitness, but nowhere do they assert or assume more stability implies more function. That’s just blatantly false.
Still evidence, and the numbers are not in any way inconsistent with numbers derived from experimental studies on the relationship between protein structure, stability, and sequence space.
The number of sequences that are easily surveyed by nature doesn’t determine what numbers are or aren’t easily fathomable. Dude come on.
That’s because the crucial point we are discussing in that exchange is your flatly false statement that:
There are fewer than 1600 folds identified to date in all of biology:
That would make specific folds incredibly common in sequence space.
That’s why I’m citing this paper because it really does show that the statement you made is false. Specific folds are incredibly rare in sequence space.
Unless you mean to tell me a frequency of 10-24 counts as “incredibly common” in your world. Everything is relative of course but that would still be silly.
I can’t believe I have to point out that you are the one who spurred this exchange with a mere verbal statement (not data), which prompted that rebuttal of mine. So I might aswell ask you to give me a frequency. Find a fold in the SCOP and then tell me what fraction of protein sequence space adopts that fold, and then tell me with a straight face that the fraction is “incredibly common”.
You’re not keeping track of the discussion. It seems we are simultaneously debating two questions:
Do structures adopted by some proteins perform functions which, if those structures were lost, the function would stop too? I say yes. You seem to be either saying no, amd/or to be taking me to imply that there are no other ways to perform those functions than by that particular structure.
Are particular folds (a particular one of those folds from the SCOP data base, for example) “incredibly common in sequence space”? I say no, that they’re actually incredibly rare. But that doesn’t prevent the evolution of those structures, nor the functions performed by them.
You seem to be saying they ARE incredibly common. But there’s just no actual evidence of that, and only evidence against it. The small number of distinct folds in the SCOP database isn’t in any conceivable way evidence for what fraction of sequences in the space of all protein sequences is able to adopt any one of a particular fold.
Gee really? Are you not able to stipulate that they are simply taking about the probability of de novo discovery of a large and folding protein sequence, not that they literally think evolution is somehow consciously trying to accomplish something. I don’t believe they’re actually that stupid and misinformed. But hey, you’re welcome to email them and tell them evolution isn’t really doing a “search” in a literal sense. I’m sure they’ll be shocked.
That’s funny because you won’t be able to find and quote me making even a single statement that remotely implies such a thing. I think you should try to work harder to understand what I actually write rather than these completely misguided and fictional extrapolations you come up with.
In light of your complete failure to understand what I’m saying, or trying to imply, I find that I’ve lost a lot of concern for your opinion on this matter in general.
Yes, I am. I am saying that structure (particularly stability) is not a proxy for function, particularly because function typically involves the same protein switching between different structures. That was illustrated by the TPM1 mutants.
Because they aren’t losing any structure. The mutants have better coiled-coil structure. Structure is not a proxy for function, even for a protein that most structural biologists would choose as the prototypical coiled-coil and is the most functionally simple part of the sarcomere. It’s a helical rod that sits in a groove of a helical filament and is moved azimuthally by a protein complex (troponin) for which we don’t have a structure. More function, less structure.
If you think that structure is a proxy for function, you can’t explain it. Instead, you can only make ex cathedra pronouncements.
Because you cited a paper about stability. Stability is not a proxy for function; in fact, it usually is in opposition to function.
That’s not true, as function rarely involves single structures. It typically involves changes in structure, often massive ones.
Nature doesn’t respect our need to classify things. That’s the case for both species and structures.
My issue is with your repeated insistence on the singular “the structure.”
Good, but you are saying that single structures correspond to functions, and you cited a paper on stability, which is often in opposition to function and even pathogenic–prions being the most famous example.
I would suggest a bit more tentativeness, as it’s much more scientific.
It does, but you’re (the authors not so much) touting a simulation of stability as a proxy for function. That is an approximation not supported by evidence.
You’re touting this as a proxy for function, remember?
I say that the vast majority of protein function involves multiple structures and requires sufficient instability to change between them, so structural stability is more often antithetical to function, simple structural proteins being an exception. What evolution is doing is changing the equilibrium between those different structures.
So no, I would never say that protein function is performed by structures. It’s far more often performed by proteins changing between different ones. By claiming that single structures perform functions, you are reinforcing Axe’s false framing.
See all of your statements quoted above. You keep asserting that single structures perform functions, and you even cite a paper about stability to support that.
This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.