Convergence, specially at the molecular level, somewhat defeats the Texas sharpshooter argument.
Explain why you think so.
2 Likes
Larry is obviously cherry-picking and engaging in straw-man fallacies (accusing those who dare to disagree with him as irrationally believing that MOST genes are alternatively spliced), as @Art pointed out in the comments. Do you call citing Larryâs book an example of academic rigor?
Has Larry published anything in the field?
So what? Is that the only relevant paper?
Because there is no description of the methods of alignment. Iâm just noting the most obvious problem; there are many other possibilities.
How can anyone know? The paper is a jokeâthere are no methods and no supplement, but there is this disturbing passage:
The inferred graphs for the original paper were also provided as supplemental files, but these would be challenging for most readers to interpret. Further- more, the broad scope of the data evaluated along with the resulting complexity of the inferred graphs made it challenging to draw useful conclusions from the results. Instead, this work focuses on a small example, that of prestin in echolocating species, and thus, provides an eas- ier to grasp example and illustration of the model.
Would you bet that prestin is representative or that it was the closest data set to what Ewert wanted to be true?
He only shows an alignment of 19 residues of prestinâs ~750 residues. Also, prestins are not the same length.
2 Likes
The fact that, when confronted to a given challenge, organisms converge toward the same solution indicates that very few solutions really exist to solve the problem. Take the example of chloroquine resistance by malaria. Despite the enormous number of organisms under selective pressure from chloroquine, it is only through a few point mutations in the same pre-existing protein that malaria developed chloroquine resistance. Thus, contrary to what evolutionists sometimes argue, it is most probably not true that many possible but unrealized biochemical solutions exist to complex biochemical challenges.
Beware of dichotomous thinking.
See what I did there?
Consider the article I referenced above:
I bolded âitâ in the first sentence here because it is crucial. It only applies to the ATP binding motif. The same experiment yielded radically different results for GTP binding solutions.
I cited the paper intending to show that convergence is a demonstrated possibility. However, had you read the paper, you would have got this next part:
The isolation of a GTP binding aptamer (T8R16/409) (Figure 4) from an ATP selection experiments may seem fortuitous, but is less surprising considering the previous discovery of numerous GTP binding aptamers at a relatively short mutational distance from the canonical ATP aptamer motif.28 Furthermore, GTP aptamers appear to be more structurally and functionally diverse and may therefore also be more abundant in sequence space. Indeed, âCAâ repeats as short as 3 nucleotides29 or G-quadruplex motifs30 have been reported to bind GTP. Finally, while the GTP binding aptamer only binds GTP in solution (as judged by ITC), it clearly binds to the ATP-agarose resin (both as the full-length T8R16/409 aptamer as well as a truncated core motif (T8R16/409core) (Supplementary Figure S11, Supplementary Table S3).
So this is saying we would expect way more convergence for ATP binding sequences, than for GTP binding sequences, as it appears there are many more functionally equivalent GTP binding sequences that are similarly accessible from most of sequence space. However in contrast, for ATP binding sequences, there appears to be fewer equivalent solutions, with the canonical ATP binding motif apparently being much better than most others, and accessible from more of sequence space, than any particular instance of a good GTP binding sequence is.
The other papers I cited even earlier also show that convergence is generally unexpected for the specific adaptations tested in those experiments.
The whole point is to show that convergence isnât just some sort of ad-hoc idea invented to dismiss phylogenetic incongruence. That gives credence to the idea that, when we discover rare but strong convergence in two distinct clades in a phylogenetic tree, weâre not just making stuff up that nobody knows whether is even feasible in principle.
The fact that strong (the degree matters) convergence is nevertheless quite rare, and that most organisms have increasingly dissimilar âsolutionsâ to the same biochemical problems with time since their separation, actually substantiates that convergence is generally quite rare, and that many distinct but equivalent solutions exist to the majority of problems faced by living organism.
But that just doesnât follow from the observation that given some challenges convergence happens, that there are generally no other reachable solutions to many other challenges. You canât just extrapolate from a handful of examples where convergence happened, to most of evolution.
You have to look at all of the evidence to get the full picture.
2 Likes
I donât see how thatâs an example of convergence. Do you know of another organism that has a convergent mechanism for chloroquine resistance? If not, what was your point?
No one anywhere is claiming that evolution can accomplish anything and everything.
Thatâs quite the straw man, Gil!
Quotes pleaseâI donât know any biologist who argues that, so quote me five such cases.
1 Like
Well I have said something (but crucially, not exactly) like that before. The problem with @Giltilâs statement there is that it seems to describe a universal generalization. I would not say that "many possible but unrealized biochemical solutions exist to [all] complex biochemical challenges." I think his statement implied the [all] that I inserted there.
I would say that we know of enough examples of it being true that many possible but unrealized biochemical solutions exist to a large diversity of complex biochemical challenges, that we have no good reason to think the types of âwaiting timeâ problems ID/creationists concoct retrospectively for adaptations in existing lifeforms, are founded on anything real. They are just extrapolating from a handful of cases of adaptations known to be relatively constrained, to basically most of the genetic differences between species. A particularly egregious (and utterly ridiculous) example is the Sanford et al. 2015 paper that was brought up in this thread by Gilbert.
1 Like
I would.
No idea, but he certainly is capable of reading the literature. What was the purpose of that publication list?
Would you like to cite another that you think would be a better reference?
Actually there is, though it isnât a good one. They used Clustal, and they say, in a general way, where they got the sequences from. There being no methods section, and the alignment not being published, thatâs all we have.
Because thatâs what the paper says, repeatedly. They compare amino acid differences, site by site. No mention of anything about gaps.
I would bet that itâs the gene he knows has something to do with echolocation. Whether he ever looked at any other such genes is unknown.
As a result of alternative splicing or as a result of short indels?
1 Like
Each time an evolutionist accuses an ID proponent to commit the Texas sharpshooter fallacy, he precisely argues this, ie that many possible but unrealized biochemical solutions exist to complex biochemical challenges. And on this site alone, we have lost count of the number of times the accusation of committing this TSF has been levelled at the proponents of ID.
The phenomenon of convergence doesnât necessarily concern different species.
This appears to be untrue. Different strains of malaria have developed CQ resistance via different routes. From here:
No genetic mutations have been linked to resistance to CQ by P. vivax . Nomura et al. investigated mutations in the P. vivax ortholog of the crt gene of P. falciparum , which has been linked to CQ resistance. The mutations incriminated in P. falciparum crt did not occur among CRPV isolates, and no other mutations in that gene correlated with the phenotype. The genetic determinants of resistance to CQ apparently differ between P. vivax and P. falciparum.
1 Like
There could be, and should be. Iâm running way behind on this discussion (replying to comment #6?) but mean to look into that question.
I saw that, and this might be a basis for comparing graphs, but falls short of a actual comparison.
This seems correct, but possibly incomplete. I need to read more.
Without some guiding âDesignâ hypothesis (such as Common Descent for evolution) to specify what sort of dependencies should occur, Ewertâs method seems bound to hit on false positives. IIRC this is not a new criticism.
Interesting, but generating hypotheses is not comparing hypotheses. Also, the infinite number of monkeys in my basement would like everyone to know that generating possible models is generally not difficult.
On many past occasions I have praised Ewert for putting forward a testable hypothesis.- a significant first for ID. So far this discussion shows that a lot more specifics are needed before we actually get a testable idea.
Yes, and no theoretical guidance on what/when/why mutually exclusive patterns should appear. Itâs a machine made to generate false positives.
If this were a paper presenting a new statistical method in a stats journal, I would expect to see a simulation study showing relative type I and II error rates comparing methods in a variety of conditions. Notably absent is any other method for comparison to AminoGraph. For example, what are the results for the current best methods of fitting trees to the same data?
Rummy touches on this in comment #22.
Yes. I donât see any goodness-of-fit results from AminoGraph, which is at odds with the discussion of âBayesian model selectionâ. Can we get a BIC?
Even to my limited understanding of biochemistry, this must undoubtedly be the case for a strong majority of biochemical functions. There are usually minor variations that perform the same function equally well, or alternate sequences that may perform not as well. It is fair to ask; how often should ID props be called on TSF?
3 Likes
Since in most cases itâs the ID supporter who introduces the âargumentâ without showing that only 1 or a few âtargetsâ exist, the âevolutionistâ isnât making any such argument, theyâre only highlighting a flaw in the ID âargumentâ. You are burden-shifting.
In any case, the approximately 10^93 possible forms of cytochrome C, not including different proteins or ribozymes that may provide the same function, show that there can be many, many, many unrealized solutions to biochemical challenge.
If you want to claim otherwise, you need to support that claim - and your recent comment about chloroquine resistance in malaria indicates you canât be bothered.
2 Likes
False. Utterly false. Sorry, but you clearly donât understand this, which is probably why you think the TSF is not fallacious.
I donât see any convergence in that case. Why did you bring it up?
If his approach to this subject is the same in the book as it is in his blog, I would not.
Being capable of reading the literature doesnât mean that one has done so.
Reading his blog post, he doesnât seem to have done a simple search for the clear demonstrations of functional alternative splicing published before the age of genomics. Have you?
Consider that of the tiny handful of molecular biologists who post comments here, two have published such clear demonstrations.
âAnotherâ singular? Wow. Shades of Ann Gauger, who will only compare/contrast two papers at a time when challenged! I suggest searching and sampling some of the thousands of relevant papers before claiming that
So, how many is a âhandful of genesâ?
I would not call that a description.
If populations of the same species independently converge to the same solution when confronted to a given challenge, then we have convergence. This is what happened with chloroquine resistance in P. Falciparum.
I think it is an abuse of language to define P. Falciparum and P. Vivax as strains. In fact, it seems that they donât belong to the same species, not even to the same subgenus, only to the same genus. So it is not surprising that these 2 different bugs donât necessarily converge to the same solution when confronted to chloroquine for their biology are very different.
Ok, call them species of malaria rather than strains of malaria.
But whatever you call them, they are still both malaria, so your claim that âit is only through a few point mutations in the same pre-existing protein that malaria developed chloroquine resistanceâ is untrue, and your focus on the terminology while ignoring the mechanism of resistance is avoidance.
1 Like
Demonstrate that population converged, i.e. the same mutations happened multiple times independently, rather than a single population spread.
1 Like
That is a very silly point. Where the lines between clades are drawn (why something should count as a species rather than a genus or strain and so on) is rather arbitrary.
I would say it is an abuse of language to make that sort of appeal to the cladistic category, to describe the magnitude of differences between two members of some clade.
Is their biology âvery differentâ (by what measure, and what degree of difference are we talking here?) and why would that difference translate to differences in the methods of resistance to CQ? What is it CQ does in P. Vivax it does differently in P. Falciparum?
And relatedly, how does the comparison between P. Vivax and P. Falciparum differ from the differences in the biology of, say bottlenose dolphins, and orcas, which are two related but distinct species of echolocating whales that merely belong to the same family?
1 Like
In evolutionary biology, that is not convergence.
Actually, it didnât. Behe grossly misrepresented the data available when he wrote that book, and far more data have been published since to show that he was wrong.
1 Like