[quote=“Cornelius_Hunter, post:193, topic:9418”]Well you are misinterpreting the phrase. The “all or none” phrase is not in reference to binding affinity, it refers to the function conferred by whatever binding you have.
I don’t see any misinterpretation. Here’s what you wrote:
Hi Cornelius - I am quite familiar with “search” issues in evolution. The evidence we have shows that searches are not a problem - whether making new protein-protein binding events, making antibodies (a subset of the former), finding new functions (such as cobbling together a citrate transporter that can work when oxygen is present) or making new species (such as humans and chimpanzees arising from a shared ancestral population.
In each case, the “search” is hugely constrained by the starting point - VDJ segments, the structure of protein J prior to mutations, the genome of the human/chimp common ancestral population, and so on.
If the “search problem” was a problem for evolution the way ID folks claim it to be, then antibodies with high affinity would never form, protein J would not be able to switch to a new receptor, and new species would not arise over time. This may well be what the ID movement would like to establish, but the evidence we have strongly supports the idea that all of these events are possible, and even probable.
At the end of the day, the assertion that protein functions are exceedingly rare in sequence space is just that - an assertion - which is usually backed up by only one paper (Doug Axe’s 2004 paper). Doug’s estimate is wildly off. The main problems are that he used a protein for which there was barely any function left (despite your claim that function is all or none, there’s an example, right in the literature by an ID proponent that shows otherwise (!)) and he made multiple simultaneous mutations (sound familiar?) to it. Guess what? The vast majority of the time this removed protein function. It simply does not follow, however, that this is a reasonable way to estimate the density of functions in sequence space.
Put another way: do you really think in an environment where protein functions are only present at one in ten to the 77th power that protein J would just happen, by chance, to have another function sitting only four amino acids away? What are the chances? That’s like picking two protons in the universe at random and finding out they are only one nanometer apart.
@Cornelius_Hunter, your objections on these issues are pointless if you refuse to explain where millions of terrestrial species have come from since the days of Noah.
But on the other side of the coin, there are plenty of cases that work perfectly in Evolutionary scenarios from common descent! Genetic drift lays the ground work for all sorts of later mutations. The hollow bones of one particular group of dinosaurs wasn’t to allow for flight … but millions of years later… the low density of these dino’s make flight possible.
Well your arguments and examples aren’t doing much to demonstrate your expertise, save for your human-chimp example, but that one suffers from being circular. Sorry, but you can’t assume the truth of evolution when attempting to argue that it works just fine.
Oh, gosh no. Unless you know of some ID folks that I don’t know. ID arguments do not preclude the immune system antibodies. For these vertebrate immune system antibodies, a relatively few residues are substituted, differently across a large number of antibodies, and essentially immediately tested against targets, in a continuous procedure. This is light years away from the hypothetical, glacial, evolutionary process. Your continued appeal to the antibody search problem is not helping your case.
Well you are really fighting the science here. You are heroically fighting the tide. Axe’s paper is far from the only work showing these results. This is by no means an ID conspiracy (I wasn’t even aware of Axe’s work for quite sometime). If Axe’s estimates are “wildly off,” then so are several others. Also, the “all or none” phrase comes from the paper, and is specifically directed at the protein-protein binding problem. My point was simply that the paper’s finding of such an “all or none” is no surprise.
Well now you’ve gone beyond fighting the tide. The notion that protein sequence space is that densely packed with function is, frankly, [pick some outrageous word to go here]. The problem here is that you are zooming in on the tactical, narrow, view, and then extrapolating that to a strategic, wide view. Your result is “wildly off” to use your phraseology.
Protein functionality is not limited to a single canonical sequence. There generally is a large number of tightly clustered sequences that will provide the functionality. And most proteins have multiple functions. What you are focusing in on here is even more tactical: you are looking at the few residues that influence the binding interface. With a few mutations, you can bind to a similar membrane protein (which a similar phage is known to bind to). So you have switched from one membrane protein to another, both sharing the same fold and tertiary structure. And the protein J is providing the same overall functionality, just with a different membrane protein. This in no way extrapolates to the different question of the density of protein function across the entire, astronomical, sequence space.
So back to our topic, your article incorrectly casts [Meyer, et. al., 2012, Science] as rebuking Behe, and I would still like to hear ideas on why Meyer, et. al. found zero synonymous mutations in all their protein J alleles.
Cornelius, I don’t “assume the truth of evolution”. The evidence supporting the hypothesis that humans and chimpanzees share a common ancestral population is, to put it mildly, quite strong, despite your claims against it. As such, it is quite reasonable to use that evidence to infer what evolution might be capable of, say, given 4-6 million years to work.
There are also many references in the literature that provide rather different estimates from Axe on the rarity of function in sequence space. Axe is near the extreme end of the spectrum of estimates - and his estimate is only for one beta lactamase for that matter.
You also haven’t addressed the issue that the sequence space that antibodies search is very small relative to the entire protein space - and yet antibodies form readily. How is it that all of those functions (binding to antigens) are packed into that tiny search space if Axe’s estimate is correct? As we have repeatedly pointed out, antibodies and proteins in general search the same type of space - protein sequence space. If Axe is right, how is it that antibodies find their function in that tiny space? You haven’t explained how this is possible. If the space is really that sparsely populated, no amount of “streamlining” the system should be able to overcome the vastness of that space to be searched.
But I suspect we will have to leave it here for now - life calls, and other needs are more pressing. Perhaps @Swamidass, @Argon, or @benkirk would care to continue the conversation.
This “debate” can go on indefinitely. I really do not see its purpose.
@DennisVenema takes a position largely accepted by mainstream science. @Cornelius_Hunter does not. There are irreconcilable differences in how these two think about data, science and evidence. This conversation mirrors larger schisms between the ID movement (that has its own way of interpreting data) and those of us in mainstream science.
Of course, I have my opinions about who is “right”, but this disagreement cannot be adjudicated in this forum. This is not how scientific arguments are litigated. @Cornelius_Hunter, even if you are to “win” this argument, you will have no impact on how science sees this topic. Literally zero impact. This is a totally futile way of impacting how science is done.
Rather than pretending that this is a useful debate, I think it would be better to end this line. For those interested in the science here, rather than a debate, I think there is value in explaining how mainstream science works and why it is so certain about evolution. Disagree if you must, but there is value in understanding how mainstream science works. It is not nearly as ideological or unreasonable as some might assume.
So, I have no interest in continue a debate here of any kind. I have no interest in participating in an entirely futile effort.
On the other hand, those curious about how mainstream science works and how it comes to the conclusion of evolution, well I am happy to explain that =). Despite the fears, evolution is not an assumption of science, it is the conclusion arrived at based on a vast amount of scientific evidence. Of course, using rules other than mainstream science (e.g. assuming YEC) one will come to different conclusions.
It is clear that many people disagree with that conclusion. Great. We all know that. Our effort here, however, is to explain why mainstream science has come to this conclusion. I do not have the desire to debate it.
You can’t argue science with people who aren’t doing science, and anyone who can’t give an age for the earth or explain the diversity of species even after Noah (let alone the entire fossil record), isn’t doing science.
For the record, my reason for engaging critics - even if they are well outside the mainstream - is not because I expect those critics to change their views, but that I hope to help others watching the exchange understand why their views are outside the mainstream. One danger of this approach, of course, is that it can appear to inflate the value of the critique. It’s hard to find the right balance.
I also do want critics like Cornelius to feel that they can come to BioLogos and converse with folks here. One of the things I appreciate about BioLogos is that we do welcome critical voices (sometimes, perhaps, to a fault). One reason for this is, as I see it, is that we feel the truth need not hide behind a wall that shuts out dissenting voices.
Well, time to get back to other things. Goodnight all.
@Cornelius_Hunter is not YEC. He appears to be an old earth creationist. I can’t explain why he doesn’t just come out and say it when asked, but it seems he thinks God specially created each species. The implication that he is YEC probably isn’t that helpful.
I understand he isn’t YEC, and didn’t think you were implying he is. But my points stand regardless. They’re two points typically avoided by IDers, regardless of how loudly they proclaim themselves to be not YEC, and totally onboard with science. If they’re onboard with science, then where is the actual science?
This is one part of the solution. There is more too. It turns out that function is rather dense in sequence space, so it is pretty normal (though not always predictable, and therefore serendipitous) that a sequence for one function is very close to a sequence for another function. It is common to start with one function, and then switch to another. This is so common that we have a name for it: “exaptation” Many proteins even have mutliple functions for just this reason: it is called “moonlighting.”
Moreover, even if Doug is off by 20 orders of magnitude, cancer would be impossible. Viruses would not be able to adapt. Evolving nylonase would have never been possible. And several direct tests to determine of the proportion of functional sequences (e.g. phage display) in sequence space would be wrong.
Disagree if you must, but that is why we think that Doug is unambiguously wrong.
Of course mainstream science can be wrong. But this isn’t either/or, and you don’t get points for being "right’ in science. You have to make a coherent, evidence-based case, following the rules of science. That hasn’t happened yet in the ID movement. Maybe it can. We just haven’t seen it yet.
[quote=“Billcole, post:212, topic:9418”]
I agree with Cornelius here and don’t think it has anything to do with the bible. Proteins are organized by a sequence and long sequences have essentially an infinite variety of possible outcomes. [/quote]
No, it’s not infinite. Anyone can do the math.
Interesting. You said 21 hours ago that you didn’t understand Dennis’s argument:
Which is it? You do realize that it’s not a mere argument or analogy, correct?
Did you read the paper? How can you know it to be contradictory if you don’t understand Dennis’s position?
Doug is off 50-60 orders of magnitude. Cornelius is wrong.
It STUDIES only one, very rare, temperature-sensitive mutant. Axe tries to generalize this to all nonmutant proteins. Can you justify his choice of a mutant specifically chosen to be at the very edge of thermal stability?
What about them? Do you know that one of them (Abl) completely falsifies your claim that “specific binding requires specific sequences”?
[quote]There is, however, a real possibility that mainstream science is wrong here. What a fascinating story
[/quote]You don’t appear to be fascinated enough to look at the evidence and understand Dennis’s (and my) point…
For these vertebrate immune system antibodies, a relatively few residues are substituted,[/quote]
Correct, which greatly constrains them to a tiny slice of nonrandom sequence space…
Large relative to what, exactly? It’s virtually nothing relative to Axe’s extrapolation.
Here’s where you are missing the point by a mile.
The primary B-cell library is only 10^7 - 10^11 different antibody sequences. From those, we get not one, but multiple cases of binding to a single antigen. Moreover (this is the important point you are eliding), that initial, unselected binding is functional, because it is sufficient to trigger proliferation of the reactive B-cell.
That’s all empirical, no inference or extrapolation.
So Axe’s extrapolation to wild-type proteins from a carefully-chosen single mutant protein of 10^77 is off by a factor of far more than 10^50.
[quote]This is light years away from the hypothetical, glacial, evolutionary process. Your continued appeal to the antibody search problem is not helping your case.
[/quote]You’re missing the point that we observe multiple cases of functional binding to ANY antigen, before ANY selection from a relatively tiny library.
I agree with Cornelius here and don’t think it has anything to do with the bible. Proteins are organized by a sequence and long sequences have essentially an infinite variety of possible outcomes.
The solution Dennis and Joshua have to this is lots of ways that you can arrange the amino acids and still get function.
Doug Axe’s paper is contradictory to Dennis position. If Doug is even off 10 orders of magnitude in his calculations then Cornelius is right.
Axe’s paper talks about bacterial proteins. But what about human nuclear proteins that are very mutation sensitive (cancer)
Joshua is right that Cornelius does not represent main stream science. There is, however, a real possibility that mainstream science is wrong here. What a fascinating story