From Liu et al., we have evidence that many mutations are probably damaging or probably benign, but that paper does not claim to provide any direct evidence that the mutations labeled “benign” are constructive. Indeed, Richard Lenskiadmittedthat “The program [used by Liu et al.] simply cannot detect or suggest that a protein might have some improved activity or altered function.”
Here the authors imply that Lenski is “admitting” that PolyPhen can detect benign mutations and damaging ones, but not “constructive” ones, completely missing the point that what Lenski actually pointed out was that “constructive” mutations could only be classified as “damaging” by PolyPhen because they alter function.
Indeed, one paper on PolyPhen-2 explains that when the program labels a mutation as “benign” that means “most likely lacking any phenotypic effect” — i.e., neutral. So the data that Behe doesn’t list isn’t the kind that would challenge his thesis — i.e., constructive mutations. Rather it shows mutations that aren’t damaging and most likely had a neutral effect.
If “benign” means “lacking phenotypic effect”, then the only other category of prediction must mean…
Man, they are really spinning their wheels now, but I find it interesting that they keep on discussing and citing the PS forums. I was skeptical, but I guess Josh @swamidass was right: they really do watch our discussions here.
That’s the impression that I’m getting also. Do they think these polar bear seminars are helping their case? Because it seems to be getting worse to me.
It’s simply not possible that they’re so incompetent that they can’t understand this very simple (but crucial) point. I can’t think of any explanation other than that they are willfully misleading their readers.
I don’t think it’s willful. Best I can gather is that Behe pulled the quote mine, and they are post hoc trying to justify it, but don’t realize the depth of the hole they are digging.
ENV, I’m still waking up. I’ll post a full (re) explanation in a moment of how we know for a fact the authors did not conclude “damaging.”
It is the conclusion the authors of the paper drew, although I would call it a very weak conclusion. You are making claims that Behe took the conclusions from the paper, but it appears he didn’t. Behe has confused predictions made by an algorithm and the conclusions drawn by the authors.
Transport and clearance are two separate functions. The authors of the paper are suggesting that the clearance function of APOB is increased. They did not conclude that transport was decreased. Behe is allowed to disagree with their conclusions, but he isn’t allowed to misrepresent their conclusions.
In this case, the program is designed to look at the substituted amino acids and assess whether they have similar chemical properties to the residues at the same position in other homologues. If they don’t, it’s safe to assume that the function of the protein is going to begin to diverge from the original function. That means the protein’s native function is probably being degraded. Thus, in a very large proportion of cases, it’s safe to assume that a mutation diverging from the chemical properties of a suite of homologues will probably damage that protein’s function. That’s exactly what the program looks for. So Swamidass may be correct that the program detects a “change in function,” but the methods used to detect such changes make it very likely that the changes will be damaging.
It’s completely circular reasoning.
The vast majority of mutations affecting function are damaging.
PolyPhen predicts mutations that cause a change in function.
We should interpret most of those change in function mutations to be damaging (see 1).
Therefore:
These PolyPhen results support the hypothesis that the vast majority of mutations affecting function are damaging.
Bill, the problem is that the PolyPhen results are incapable of contradicting Behe’s hypothesis, regardless of whether or not it’s correct. Therefore, it’s not a valid test.
I and others have done so repeatedly, even in threads you were participating in. In brief:
Behe’s hypothesis is that adaptive evolution proceeds primarily via mutations that degrade molecular functions instead of improving them. The chief example he gives in the beginning of his book is the APOB protein (among others) in Polar bears, which appears to be under positive selection. To support this example, he uses predictions from PolyPhen, a program that examines sequence alignments to predict functional changes in proteins. If a variant is not predicted to change any amino acids, PolyPhen calls it benign. If a variant does change an amino acid, this is evidence of a potential functional change, and PolyPhen will label this change as either “possibly damaging” or “probably damaging.”
Thus, PolyPhen is incapable of producing a result that contradicts Behe’s hypothesis. Either a mutation will not change the protein and be labeled as benign (and is thus unlikely to be under positive selection) or will change the protein and be labeled as damaging. There is no possible way for PolyPhen to predict an enhancement in function, so the results cannot be used to determine how frequently adaptive evolution proceeds via degrading function versus improving function.
Basically, Behe has set up a “heads I win, tails you lose” scenario.
That isn’t how science works.
Edit: I want to add that I missed the fact that some variants that change the protein can be labeled as benign, based on the level of conservation at the site and the likelihood of structural changes. Still, there is no category of “enhanced” function or anything like it.
Second, Behe’s critics are making a big deal about his not showing the data in the table where mutations were predicted to be “benign” — but benign or neutral mutations don’t challenge his thesis . Data only challenges Behe’s thesis when a mutation is shown to be constructive. In this regard, it’s crucial to point out that the mutations that Behe didn’t list from the chart that were not said to be damaging were also NOT said to be “constructive.” They were said to be “benign.” That’s a key point that completely undermines Lents’s charge that the data Behe doesn’t list is somehow “contrary” to his position.
However, ALL of the mutations listed in Table S7 of Liu et al. are in genes that show clear indications of being under positive selection. While benign mutations in these genes may be neutral, they may also be beneficial or constructive. Indeed, to cite an example from the data Behe omits, at least one of the benign changes in the polar bear OR5D14 gene must be constructive or beneficial, since this gene that is under positive selection only carries mutations flagged by PolyPhen2 as benign. To give another example, the sole missense mutation in another gene under positive selection – EHD3 – must be beneficial or constructive (PolyPhen2 flags this change as benign).
Just how does the data Behe omits affect his thesis? From this essay:
The first gene (call it A) would be helpful if it mutated (call the mutated protein A*) at a particular residue of the protein it coded for to give a new constructive feature (perhaps a helpful new binding site). The second gene (call it B) would be helpful if it mutated (to B*) so that its activity were substantially degraded or eliminated entirely. Yet there are orders of magnitude — a hundred to a thousand — more ways to degrade B than to improve A. That means that if neither mutation were originally present in the population of a species, B* would be expected to appear in only a hundredth to a thousandth of the time needed for A* to show up.
(Bold emphasis added to point out the actual numbers Behe is thinking of.)
In other words, for every benign and beneficial mutation, there should be 99-999 damaging mutations. If one considers all of Table S7, and not just the parts Behe presented in his first response, then we see easily that Behe is quite completely wrong in his assertion. There are 23 benign missense changes, 15 predicted to be either possibly or probably damaging, and nine with conflicting predictions. As we see from the above (and from closer perusal of the Table), at least three of the benign changes MUST be constructive or beneficial, and the number likely is much greater. This is plainly not in line with the numbers Behe asserts.
The bottom line is, the parts of Table S7 that Behe omitted plainly conflict with his ideas and claims.
Please stop misrepresenting these poor authors. The evidence is so trivially against the claim that they agree with you, it is best to just apologize and stop.
Constructive neutral evolution demonstrates that Behe’s argument fails because he neglects that neutral mutations are often constructive. This is one of the important facts we can see on display in the ENCODE data.
I encourage ENV to look at some of the first responses to Behe’s work, which were spot on:
That’s… how circular reasoning works. The point is that the interpretation in question can’t support the initial hypothesis if it assumes the initial hypothesis as a premise. If there’s independent evidence supporting claim 1, that’s what should be focused on, not this.