Origin of Cytochrome C

[Rumraket (Mikkel R.)]

@scd

indeed. so?

So all the rest I wrote in my post.

remember that we only need few of the rare examples to challenge unguided process

You can challenge things all you like, what matters here is whether those challenges make sense. In your case, they don’t.

. and there are probably much more than just few rare examples in nature.

Is that so? How do you know? You don’t know it for cytochrome C. Even if you could show cytochrome c as we know it is relatively rare in sequence space, that wouldn’t tell you that it couldn’t evolve from some other more abundant function.

The problem is when you are claiming that X could not have evolved, you are taking on an enormous burden of proof. You now have the burden of demonstrating that no conceivable scenario for X’s evolution could work.

@scd

problematic assumption.

I know, I explained why. But that reason also undermines your meaningless calculation, since you don’t actually know the relationship between protein length and frequency of function in sequences space.

It could be that most of those 10^93 functions cluster densely in the L=80 range, just to pick an example, and then they grown increasingly rare as you go shorter or longer. But that could also imply, in turn, that there are selectable paths to shorter or longer sequences from sequences of L=80.

We simply don’t know. But for your claim that X could not evolve to work, you would need to actually show what this relationship is like. To make matters even worse for your claim, you would also need to show that there are no other functions nearby that overlap anywhere in sequence space from which cytochrome c could have evolved.

I’m sorry for you that you have taken on this massive burden, but that is really what you have done.

remember that about third of the cytochrome c protein is crucial for its function. so only about 30 aa are needed theoretically. thus the chance is 20^30

A chance can’t be 20^30. It can be between 0 and 1. You must mean 1 in 20^30, but that would imply only one sequence in L=30 space is a functional cytochrome c.

very close to 10^37

No that doesn’t make sense you are getting your numbers mixed up. 10^37 isn’t a size of the space, it’s a probability, or a frequency. The frequency is actually ~10^-38. Notice the minus.
That is:
\frac{\displaystyle10^{93}}{\displaystyle20^{100}} \simeq7.9\times10^{-38}.

You are saying that, if hypothetically there was a 30 amino acid long but functional cytochrome c, then the total sequence space for 30 amino acid proteins is approximately 10^39. So far so good.

But then you say that is “very close to 10^37”. But that number isn’t the size of sequence space.

That number is supposed to be ~10^-38, which is the the ratio of functional cytochromes c to all possible sequences 100 amino acids in length, if all 10^93 cytochromes c are 100 amino acids long. But they aren’t. There are many shorter, and many longer. And we don’t know how the lenght correlates to frequency functional ones in that space.

If that was the same density of function in L=30 sequence space in absolute terms, that would basically mean that out of all sequences in that space of proteins 30 amino acids in length, only one of them would be a functional cytochrome c. We just have no reason to believe any such thing.

[Roy (30-year veteran)]

@scd

problematic assumption. remember that about third of the cytochrome c protein is crucial for its function. so only about 30 aa are needed theoretically. thus the chance is 20^33, very close to 10^37 (i guess that Yockey calculation might base on that). it means that the chance to get even a shorter cytochrome might be similar. alternative explanation is that a third of the shorter version of cytochrome c is needed. in that case we are talking about 23 aa, or 20^23. i think that this number is more realistic.

If one third of a 100aa protein is constrained, that suggests that the binding and/or folding site(s) are about 33aa long in total. A shorter variant (say 70aa) would still need the same 33aa binding/folding site(s), and so a higher proportion of constraint - nearly half.

The idea that one third of a long protein being constrained means that only one third of a shorter variant is constrained is a basic misconception that can only lead to arguments that are total gibberish.

(I currently feel very constrained. Also restrained)

[scd]

@Rumraket

You don’t know it for cytochrome C. Even if you could show cytochrome c as we know it is relatively rare in sequence space, that wouldn’t tell you that it couldn’t evolve from some other more abundant function.

so your countner argument is that the cytochrome might evolved stepwise from other simpler function. is that right? if so i need to show that a protein cant evolve stepwise. one evidence is the fact that many proteins need at least several domains for their minimal function. so if we will remove some amino acids it will not give us a different function but a non functional protein. for instance take the aminoacyl trna synthetase protein:

(image from Protein Synthesis and Folding | Clinical Gate)

this protein needed at least 2 binding sites: one for the trna and one for amino acid. if we will remove one of these sites the protein will be non functional since there is no use for just binding trna or amino acid. thus we cant realy evolve such protein stepwise by changing function each step.

[Rumraket (Mikkel R.)]

@scd

so your countner argument is that the cytochrome might evolved stepwise from other simpler function. is that right? if so i need to show that a protein cant evolve stepwise. one evidence is the fact that many proteins need at least several domains for their minimal function.

No, that is a misunderstanding. There is no requirement that the ancestral protein had fewer binding spots or fewer domains or anything like that.

It could have had, we don’t know, but that is not a strict requirement for the evolution of one protein function from another. The same protein, relatively unaltered, either exchanged one function for another due to one or a few mutations, or a latent, extremely low and unused capability of the protein got enhanced by selection of beneficial mutations once the conditions favored those mutations.

Examples are known where scenarios such as these occurred. For example, the the lactate dehydrogenase(LDH) family of enzymes in the Apicomplexa clade, evolved from malate dehydrogenase (MDH)enzymes. The ancestral enzymes had a latent, extremely low-level activity towards the interconversion of pyruvate and lactate that wasn’t used(likely because it was so low it had no effect on fitness), as it’s function was towards interconversion of oxaloacetate and malate. Originally this enzyme was duplicated, and a 6 amino acid insertion in the active site of one of the duplicate ancestral MDH enzymes radically increased it’s LDH activity. Subsequent mutations further enhanced this now beneficial function, so much that some of it’s extant descendants are no longer capable of catalyzing the MDH function.

So basically one function was exchanged for another. There were no new binding spots, no new domains, or anything of that sort. Rather, the protein really just gradually changed from one function to another. You can read all about that in this paper: https://cdn.elifesciences.org/articles/02304/elife-02304-v1.pdf

Here I have drawn some arrows on figure 5 from the paper, in order to highlight the complete exchange of function that occurred from the ancestral AncMDH1 to the descendant PfLDH enzymes

Figure 5. Evolution of novel LDHs in Apicomplexa. The y-axis of the bar graphs is log(kcat/KM), with oxaloacetate in blue and pyruvate in vermilion. Blue vertical bars (left) quantify activity of the given enzyme towards oxaloacetate; red vertical bars (right) quantify activity towards pyruvate. RbMDH is a representative α-proteobacterial MDH from Rickettsiabellii. T. gondii has two LDH proteins (TgLDH1 and TgLDH2), each expressed at different stages of the life cycle (25).

That’s one possible way the cytochrome c protein might have evolved from another protein with a completely different function, incrementally by mutations, without those mutations having to create new domains or folds or binding spots. Those mutations can simply change the shapes and intermolecular affinities of existing binding spots and existing domains, incrementally, mutation by mutation, enchancing one function while decreasing another.

If you’re going to say the cytochrome c protein could not have evolved, you will need to explain how you know something like this could not have occurred.

[Rumraket (Mikkel R.)]

@scd

this protein needed at least 2 binding sites: one for the trna and one for amino acid. if we will remove one of these sites the protein will be non functional since there is no use for just binding trna or amino acid. thus we cant realy evolve such protein stepwise by changing function each step.

Ancestral aminoactyl-tRNA-synthetase enzymes have been reconstructed in the laboratory and tested for function. They lack the tRNA anticodon recognition domain, but they still work. They promiscously aminoacylate tRNA, but even then show some level of discrimination.

[colewd (Bill Cole)]

@scd

so your countner argument is that the cytochrome might evolved stepwise from other simpler function. is that right? if so i need to show that a protein cant evolve stepwise.

This is setting yourself up to have to prove a negative. You can just say it is very difficult for evolutionary mechanisms to demonstrate. Watch very carefully how Mike Behe argues. He will not set himself up to try and prove a negative. At the end of the day science is not about proof. Its about empirical evidence and confidence in your hypothesis.

[Giltil (Gilbert Thill)]

@Rumraket

No Bill, stop saying demonstrable falsehoods please. He is taking on that burden himself when he asserts these proteins could not have evolved. Anyone advancing such a claim has the burden of demonstrating it’s truth. It not not our job to prove it false when he ’s the one claiming it’s a problem.

Let’s see that.
Imagine John says that aliens don’t exist but Tom affirms that they exist. Who bears the burden of proof? It is Tom of course, for John would have to prove a negative (that aliens don’t exist), which is notoriously difficult to do for any topic. Indeed, in this case, it would even be impossible for John to prove that aliens don’t exist.
In your dispute with Bill, you have Tom’s role and Bill has John’s one. Bill is saying that such protein could not have evolved and you are saying that it could. Here again, the burden of proof rests on you who affirm a positive, not on Bill who affirms a negative.

[Rumraket (Mikkel R.)]

@Giltil

Let’s see that.
Imagine John says that aliens don’t exist but Tom affirms that they exist. Who bears the burden of proof?

Both. Both of them are making knowledge claims, so they each have the burden of showing their respective claims to be correct.

Here’s a solution to Tom and John’s conundrum: Stop making claims you can’t support.

In answer to the question ‘do aliens exist?’ both Tom and John should answer “I don’t know”. They shouldn’t claim to know something that they don’t.

Does Tom know that aliens exist? No. Then Tom should stop claiming it.
Does John know that they don’t exist? No. Then John should stop claiming it.

This really isn’t a difficult concept.

It is Tom of course, for John would have to prove a negative (that aliens don’t exist), which is notoriously difficult to do for any topic. Indeed, in this case, it would even be impossible for John to prove that aliens don’t exist.

Then he shouldn’t claim it. When he does, he takes on that burden. If there’s a burden of proof that is very high, or impossible to meet, then don’t make a claim that would entail you have the knowledge the claim requires when you actually don’t.

In your dispute with Bill, you have Tom’s role and Bill has John’s one.

In what dispute exactly? I have numerous simultaneous disputes with Bill. One of them is if some protein could have evolved, another is if life could have been “simple” whatever that means. Depending on the exact dispute, my own claims would either be “I don’t know”, or “we have an in-principle mechanism by which something like that could happen because we know of other examples where it did, so how do you know such a thing didn’t happen in this case too?”.

I’m not claiming to know that it really happened(I don’t know what happened), I’m asking Bill and scd to explain how they know it didn’t. Which they must know when they’re claiming it didn’t, so they should support that claim. But they can’t support it, so they should stop claiming it.

Scd and Bill are making sweeping claims that they can’t support. Scd for example is saying that CYC could not have evolved. I then respond that it is conceivable CYC evolved by a mechanism like this other protein evolved (the MDH evolving into LDH example), so when scd says it couldn’t, he must have some way of ruling a scenario like that out. And I’m asking how he rules that out?

Notice I’m not claiming to know that CYC did evolve through such a mechanism, I don’t know how CYC evolved. I am merely giving examples of mechanisms that undermine any claim that it couldn’t, and pointing out that none of Bill or scd’s rationalizations accomplish ruling this out.

If scd wants to avoid showing that it couldn’t evolve through such a mechanism, he should just stop saying it couldn’t. It really isn’t difficult to not claim to know things you don’t know. The simple fact is neither of us know how it originated.

I know of ways in which proteins have evolved into other proteins, so when someone says a particular protein could not have evolved, I’d be very interested to hear how they rule such mechanisms out. They must be able to do that when they claim to know it.

Bill is saying that such protein could not have evolved and you are saying that it could.

No I’m actually asking how Bill and scd know that a scenario like the MDH->LDH example could not take place?

Here again, the burden of proof rests on you who affirm a positive, not on Bill who affirms a negative.

I’m not affirming anything, I am questioning how Bill and SCD know what they are claiming to know. When I give an example of a conceivable possibility, I am not claiming to know that actually occurred, I am mentioning it simply to remind them that when they make such claims they must show how they rule such things out.

[Rumraket (Mikkel R.)]

NOT knowing something is okay. We don’t know everything, and we don’t know enough to make universal claims about possibility. Stop making such claims.

Instead of saying “X could not happen”, just say “I dont’ know how X ocurred”. It’s not a painful statement or change to make. There are things we don’t know, there are claims we shouldn’t be making.

Don’t.

How did CYC evolve? I don’t know. Does that mean we can claim it couldn’t? No.

That’s it. That’s all there is to say on that subject. Move on with your lives.

[colewd (Bill Cole)]

@Giltil

Imagine John says that aliens don’t exist but Tom affirms that they exist. Who bears the burden of proof? It is Tom of course, for John would have to prove a negative (that aliens don’t exist), which is notoriously difficult to do for any topic. Indeed, in this case, it would even be impossible for John to prove that aliens don’t exist.

I would say John should not say something can’t evolve. If he does then he has set himself up to have to prove a negative. Difficult to explain its origin with known evolutionary mechanisms is a better claim.

[scd]

@T_aquaticus

How did you determine that for the entire history of that protein?

@Rumraket

If you’re going to say the cytochrome c protein could not have evolved, you will need to explain how you know something like this could not have occurred.

ok. several things. first: in science we go by the evidence we have and not by the evidence we dont have. so by the evidence we do have there is no stepwise to evolve that cytochrome. only speculation. so the burdon of proof isnt on my side anyway.

now, you are basically suggesting that the cytochrome might evolved by changing function each step (say even each amino acid) till the whole cytochrome evolved?. but it doesnt make sense for several reasons. the main reason is that if it was true then we need to believe that all the functional proteins are near each other in the sequence space. so it should be easy to move from one protein into another. this is clearly not true for many proteins (although not for all of them) since many proteins are very different from each other and have many different functions:

(image from The Amazing Proteins |)

it means that we cant realy move from one into another (again- not in any case) since it should be extremely unlikely.

[Roy (30-year veteran)]

@scd

@T_aquaticus

How did you determine that for the entire history of that protein?

@Rumraket

If you’re going to say the cytochrome c protein could not have evolved, you will need to explain how you know something like this could not have occurred.

ok. several things. first: in science we …

“we”? Are you a scientist?

…go by the evidence we have and not by the evidence we dont have. so by the evidence we do have there is no stepwise to evolve that cytochrome. only speculation. so the burdon of proof isnt on my side anyway.

If you say cytochrome C could not have evolved, you have the burden of supporting it.

Now for the real problem:

now, you are basically suggesting that the cytochrome might evolved by changing function each step (say even each amino acid) till the whole cytochrome evolved?. but it doesnt make sense for several reasons. the main reason is that if it was true then we need to believe that all the functional proteins are near each other in the sequence space

No. Not all of them. Only the ones which were precursors of cytochrome C.

Going from cytochrome C’s ancestral proteins to cytochrome C doesn’t require either of them being similar to collagen any more than travelling from e.g. London to Paris doesn’t require going anywhere near Sydney,

[Rumraket (Mikkel R.)]

@scd

ok. several things. first: in science we go by the evidence we have and not by the evidence we dont have. so by the evidence we do have there is no stepwise to evolve that cytochrome.

No, that’s not the evidence we have. Rather, we don’t have evidence about what other possible functions there are in the vicinity of CYC in sequence space. This is not the same as saying we have evidence that there are no such functions.
The nearby sequences have not been tested for other functions, so we do not have evidence that there are none. So to say that there are none, you would need to do experiments and test for it. You can’t just declare that there are none and demand others prove you wrong.

We can use @Giltil’s aliens analogy here. We don’t know that there are no aliens, and we don’t know that there are aliens. We simply don’t have the kind of evidence that would be needed to confidently assert either of these options. It is the same with CYC and other putative functions nearby in protein sequence space. We don’t have the kind of evidence that would be needed to assert that there are no other functions, or that there are other functions.

As such, we cannot claim that CYC could not have evolved, and we can’t claim that CYC did evolve from another function. Because we don’t have the evidence that is needed to support such a claim.

now, you are basically suggesting that the cytochrome might evolved by changing function each step (say even each amino acid) till the whole cytochrome evolved?

What I am saying is we know of examples where one protein with one function (MDH), gradually evolved into another protein with another function (LDH). And somewhere along the way, as the LDH function came to exist by mutation, it was further increased in activity by mutations, and then eventually the MDH function was lost, becoming completely replaced by the LDH function.

Look again at the graph I posted here:

Figure 5. Evolution of novel LDHs in Apicomplexa. The y-axis of the bar graphs is log(kcat/KM), with oxaloacetate in blue and pyruvate in vermilion. Blue vertical bars (left) quantify activity of the given enzyme towards oxaloacetate; red vertical bars (right) quantify activity towards pyruvate. Rb MDH is a representative α-proteobacterial MDH from Rickettsiabellii. T. gondii has two LDH proteins (TgLDH1 and TgLDH2), each expressed at different stages of the life cycle (25).

So these two functions are quite near each other in sequence space. They are so near each other, that in fact they overlap. Several intermediate proteins could do both functions(MDH+LDH), though at a significantly lower level compared to the fully specialized proteins AncMDH1, and PfLDH.

It is entirely conceivable that CYC is a protein that overlaps other functions in protein sequence space. Functions we don’t currently know what are, because we haven’t tested for them. Now we don’t know that there are other such functions, I’m not saying we know this. What I am saying is that, if you’re going to claim that CYC could not have evolved (that there are no aliens out there), you are going to have to explain how you know there are no other overlapping functions in sequence space (how you know what exists on planets in distant galaxies you’ve never been to).

but it doesnt make sense for several reasons. the main reason is that if it was true then we need to believe that all the functional proteins are near each other in the sequence space.

No, we don’t need to believe that ALL functional proteins are near each other in sequence space. We just need to know that SOME functions are near each other in sequence space, and that we do not have any a priori reasons for thinking this is not the case for CYC. And we (to my knowledge) do not have any such a priori reasons for thinking that for CYC.

so it should be easy to move from one protein into another.

And it some times is. And it could be, at least in principle, the case for CYC also. We don’t know. But to claim that it is NOT the case, we need evidence.

this is clearly not true for many proteins (although not for all of them) since many proteins are very different from each other and have many different functions:

Yes, many proteins are very different from each other, and for many of them scientists don’t think they evolved from each other or share common ancestor proteins(In fact, they think that they definitely didn’t). Though, for others still, they do think that, because they have evidence that they did.

[scd]

@Rumraket

We don’t know that there are no aliens, and we don’t know that there are aliens. We simply don’t have the kind of evidence that would be needed to confidently assert either of these options

fair enough.

as for the multifunctional protein (MDH+LDH). if that protein already had these two functions in the first place then its just about improving existing function an not about new function. but lets say that this is indeed a new function.

@Rumraket

No, we don’t need to believe that ALL functional proteins are near each other in sequence space.

actually we do. why? because even if few of these proteins are insulated in the sequence space it means that they cant evolve by small steps from other proteins . remember again my point about even few rare proteins as evidence against undirected process. now, why i think that its also true for the CYC case? because according to your scenario that protein suppose to evolve from a different protein with a different function. so even if we start with a protein that is 10 aa long, its clear that it was not similar to the CYC, since such a small protein cant be similar to a 100 aa long protein. thus they should be insulated.

@Roy

Going from cytochrome C’s ancestral proteins to cytochrome C doesn’t require either of them being similar to collagen

so how these two proteins evolved? if they are so different they should be insulated.

[Rumraket (Mikkel R.)]

@scd

actually we do. why? because even if few of these proteins are insulated in the sequence space it means that they cant evolve by small steps from other proteins

No, we still don’t need to believe that ALL functional proteins are near each other in sequence space.

. remember again my point about even few rare proteins as evidence against undirected process.

Yeah I remember it, you just haven’t been able to find any examples of it.

now, why i think that its also true for the CYC case? because according to your scenario that protein suppose to evolve from a different protein with a different function. so even if we start with a protein that is 10 aa long, its clear that it was not similar to the CYC,

None of this blather makes any sense. We don’t have to think the 100 aa protein had to evolve from a 10 aa protein. The MDH and LDH enzymes are almost exactly the same length. The LDH function was created by a 6 aa insertion into the MDH enzyme, so it grew longer by 6 amino acids, and this created the capacity for the LDH function to become selectable in the MDH enzyme. Over time, the MDH function was completely lost to the LDH function.

So these two function were in fact relatively close to each other in sequence space, separated by a 6 aa insertion in the MDH enzyme and possibly even less.

since such a small protein cant be similar to a 100 aa long protein. thus they should be insulated.

A 10 aa protein might be “insulated” from a 100 aa protein, sure. But is this 100 aa protein “insulated” from other functional ~100 aa proteins and how do you know? You don’t know, you’re just making up scenarios you think can’t work. This has be come ridiculous.

How hard is it for you to just admit you don’t know whether CYC could have evolved from another protein? It’s really not hard at all. I can do it, it works like this:

I don’t know whether cytochrome c could have evolved from another protein.

See, that wasn’t so difficult.

[Giltil (Gilbert Thill)]

@Rumraket

I don’t know whether cytochrome c could have evolved from another protein.

So, according to you, how did cytochrome c come into existence ?

[Rumraket (Mikkel R.)]

@Giltil

So, according to you, how did cytochrome c come into existence ?

I don’t know. There is no “according to me” for cytochrome c. Some times we have good evidence for how a protein evolved, some times we don’t.

[pevaquark (Matthew Pevarnik)]

@Giltil

So, according to you, how did cytochrome c come into existence ?

Is the argument by @scd and yourself here that since you scientists can’t explain the origin of cytochrome C a God must have supernaturally made it? And then throw some probabilities in that we don’t actually know but since the number is small (how small = God?) it couldn’t have come about via the physical laws that most believers believe that God made in the first place?

[scd]

@Rumraket

So these two function were in fact relatively close to each other in sequence space, separated by a 6 aa insertion in the MDH enzyme and possibly even less.

you know what? let me ask you that question: say that we have a sequence space of 10^100 possibilities. and say that we know that there are only about 2 functional sequences among this space that are coded for a completely different function and structure. do you think that in this case the chance to get a new function is high or very low?

@Rumraket

How hard is it for you to just admit you don’t know whether CYC could have evolved from another protein?

i actually never said the opposite as far as i remember. i just think that there is at least one serious problem with that notion. especially when we are talking about completely different proteins.

[Rumraket (Mikkel R.)]

@scd

you know what?

By a direct experimental demonstration. Meaning scientists showed it with an experiment. Read the paper.

Boucher JI, Jacobowitz JR, Beckett BC, Classen S, Theobald DL. An atomic-resolution view of neofunctionalization in the evolution of apicomplexan lactate dehydrogenases. Elife. 2014 Jun 25;3. doi: 10.7554/eLife.02304.