Side Comments on Progress after the Royal Society Conference

[Rumraket (Mikkel R.)]

Continuing the discussion from Progress after the Royal Society conference? thread which got locked so now has to continue in this new thread lol:

@lee_merrill

@Rumraket

@lee_merrill

So how are the papers not saying they encountered a limit at about two mutations? Please show how this is not their conclusion.

By such a sentence not appearing anywhere in the result or conclusion.

How about “Waiting on Two Mutations…” (Durrett and Schmidt, Genetics), where we read “Conclusion: Achieving two coordinated mutations in humans (where the first mutation does not confer a benefit) would take > 100 million years!”

1. That is waiting for a pre-specified result, which is textbook hindsight thinking.
2. Where there are no other equivalent sets (in other words, only these two coordinated mutations, as opposed to any possible set of two coordinated mutations).
3. Humans have small effective population sizes and long generation times.
4. Life has existed on Earth for about 3.8 billion years, perhaps a bit more.

All of these factors matters quite a lot. So the authors do not in fact state that evolution in general has a two-mutation limit. Heck, even saying it would take 100 million years on average for 2 pre-specified mutations to occur and fix in the human lineage isn’t a limit either. Evolution isn’t “the time since humans split from our common ancestor with the chimpanzee.”

@lee_merrill

@Rumraket

@lee_merrill

And then show me where some biologists disagree with this limit.

Here on this very forum. In the various papers (by Michael Lynch and others) that have responded to Behe and others who have attempted to publish papers claiming such a limit. You should watch the video by evolutionary biologist @talkpopgen (Zach Hancock) I linked earlier.

Can you point to one of these papers, please? That was what I was asking for.

Can you also then acknowledge that ALL the biologists you have interacted with on this forum specifically on this question disagrees with the claim that there’s a 2-coordinated-mutation-limit to evolution?

@Mercer doesn’t agree with it.
@John_Harshman doesn’t agree with it.
@AndyWalsh doesn’t agree with it.

@lee_merrill

And I did watch one of your videos by Zach Hancock over in the Royal Science Conference thread, and found it unsatisfactory. But I don’t see a video by Hancock in this thread, could you repost it?

Amazingly it’s like three posts above yours. Here it is again:

Waiting-time? No Problem.

@lee_merrill

@Rumraket

In fact I can do better. Biochemist Michael Behe himself disagrees with it. In his first so-called waiting time paper (Behe & Snoke 2004) he writes(about the population sizes required to fix various numbers of individually deleterious mutations):

Such numbers seem prohibitive. However, we must be cautious in interpreting the calculations. On the one hand, as discussed previously, these values can actually be considered underestimates because they neglect the time it would take a duplicated gene initially to spread in a population. On the other hand, because the simulation looks for the production of a particular MR feature in a particular gene, the values will be overestimates of the time necessary to produce some MR feature in some duplicated gene. In other words, the simulation takes a prospective stance, asking for a certain feature to be produced, but we look at modern proteins retrospectively. Although we see a particular disulfide bond or binding site in a particular protein, there may have been several sites in the protein that could have evolved into disulfide bonds or binding sites, or other proteins may have fulfilled the same role. For example, Matthews’ group engineered several nonnative disulfide bonds into lysozyme that permit function (Matsumura et al. 1989). We see the modern product but not the historical possibilities.

This is Behe and Snoke admitting a colossal caveat with interpreting their results, and that the whole thing suffers from the hindsight thinking. Lynch 2005 demonstrated what happens for the waiting times if Behe and Snoke’s assumption of only a single set of mutations is violated (see figure 3).

Thank you! Comments follow.

First we read (in Lynch’s paper) “Although the authors claim to be evaluating whether Darwinian processes are capable of yielding new multi residue functions, the model that they present is non Darwinian (King and Jukes 1969). Contrary to the principles espoused by Darwin, that is, that evolution generally proceeds via functional intermediate states, Behe and Snoke consider a situation in which the intermediate steps to a new protein are neutral and involve nonfunctional products.”

Um, seriously?

Yes seriously. It’s a mistake to claim it is a “Darwinian” process if it explicitly involves the absence of selection on intermediate steps.

@lee_merrill

No one denies that evolution can reasonably be expected to proceed from step to selectable step, if the new fitness peak is nearby. And yes, Darwin expected such a progression, with lots of intermediate steps, but he admitted he didn’t know of any.

Yeah that never happened. But even if he didn’t know of any, we know of plenty of examples now.

@lee_merrill

And the gaps are still there! The expected continuous intermediates all along the path of life, have not been found.

No intermediate fossils founds since Darwin’s time? LMAO.

What’s next, you’re going to claim any new discovered fossil just creates two more gaps?

What is this Kent Hovind creationist bingo crap you’re pulling now? It’s certainly funny.

@lee_merrill

So the only real alternative is neutral steps, certainly everyone all agrees that harmful mutations are not a way to expect evolution to continue.

Even that is at least partially untrue. Deleterious mutations can fix through bottlenecks/founder events or even genetic hitchhiking, so even in cases where to get further up on a fitness peak requires dipping into lower fitness on the way there, that still wouldn’t prevent the higher fitness from being reached. There are still ways to get there from here.

@lee_merrill

“Moreover, given that the authors restricted their attention to one of the most difficult pathways to an adaptive product imaginable, it comes as no surprise that their efforts did not bear much fruit.”

Right, evolution doesn’t work very well, even with a path of mostly neutral mutations. Just Behe and Snoke’s point.

A single neutral path out of all possible paths, with no alternative paths yielding a similar result. I’m going to have to make a figure to make you understand what I am saying here, right?

@lee_merrill

Then Lynch points out:

“To stretch this statement to imply that all amino acid changes lead to nonfunctionalization is a gross mischaracterization of one of the major conclusions from studies on protein biology—most protein coding genes are tolerant of a broad spectrum of amino acid substitutions (Kimura 1983; Taverna and Goldstein 2002a,b).”

Then turning to Behe and Snoke, we read:

“… most mutations that alter the amino acid sequence of a protein effectively eliminate function (Reidhaar-Olson and Sauer 1988, 1990; Bowie and Sauer 1989; Lim and Sauer 1989; Bowie et al. 1990; Rennell et al. 1991; Axe et al. 1996; Huang et al. 1996; Sauer et al. 1996; Suckow et al. 1996).”

So it seems to be Lynch’s references against Behe and Snoke’s references! So which side is right?

Lynch. I know because I read those papers, we even had a thread on it on this forum where I pointed out that Behe misrepresented what those papers say. Read the thread from that post on.

@lee_merrill

There seems to be an ongoing discussion, and Behe and Snoke would seem to be completely within their rights to say “we pick this side.”

I actually think it’s a bit of a red herring. I wouldn’t say Behe & Snoke’s model is like physically impossible (after all mutations that eliminate function in proteins do clearly exist), but it is a very restrictive scenario and can’t be extrapolated to protein evolution in general.

@lee_merrill

Then Lynch says:

“under the assumption that all 20 amino acids are equally substitutable in the intermediate neutral state.”

Oops, I hope that didn’t mean like what it sounded he said. That seems extreme, if that is what he means, he didn’t say that earlier.

It is less extreme (there are proteins known where some sites have that degree of tolerance, or close to it such as 19 out of 20 being tolerated) than Behe & Snoke’s assumption that the mutations are individually null mutations that are compensated for by a duplicate completely free of purifying selection, and that there is only one such set possible for the gene.

The two most fundamental problems with Behe and Snoke 2004, is (1) the Texas Sharpshooter fallacy (hindsight thinking), which they themselves indirectly admit in the discussion.

Lynch also says as much in his 2005 response paper:

Second, Behe and Snoke assume that only two specific amino acid sites within a protein are capable of giving rise to a new selectable diresidue function. Given that the average protein in most organisms contains between ~300 and 600 amino acids, this assumption is also unrealistic. Increasing the number of participating amino acid sites from n =2 to just 10 can magnify the probability of neofunctionalization by more than 10- fold

That is, if there are 10 alternative positions in the protein that can participate in a new function, rather than just 2, this radically increases it’s probability of evolving.

And of course this would be further amplified by the consideration of the possibility of some function in some gene, rather than a specific function in a specific gene as Behe and Snoke affirms.

And (2) That their duplicate genes are assumed to be completely unlinked, such that purifying selection is initially absent until such a time that the 800 times more likely null mutation renders one of the copies nonfunctional. Which it would be expected to do 800 times more frequently than it would gain just one of the 2, 3, or more, novel function-creating mutations. When one such null mutation occurs in one of the copies, which is basically inevitable given their assumptions, this then instantly puts the other gene under purifying selection so that the novel function creating mutations, which are assumed to be individually deleterious, are purged away in that one. Now neither gene can evolve towards the new function. Because one has been irrecoverably destroyed by a null mutation, and the other has come under purifying selection against the novel deleterious mutations required in sets of 2 or more before they take functional effect.

The whole thing is rigged to fail. And the above is in fact an absolutely 100% accurate characterization of the model.

Yes, we’ve had all these discussions before around here. Suffice it to say, Lynch is right and Behe is wrong. Behe’s papers don’t support his strange scenario as being something typical for evolution.

@lee_merrill

All this seems to cover the principles where Lynch said Behe and Snoke were mistaken,

LMAO

@lee_merrill

on to the problem with Behe and Snokes’ caveat!

Alright, in the caveat, they admit their estimate may be an underestimate in view of some considerations, or an overestimate based on other considerations! So hence the “colossal caveat”? It could even need both of these adjustments together! So I don’t see how they abandoned or discredited their results here. Certainly not in a colossal way.

And as far as hindsight thinking, I scanned the article looking in vain for this, could you post the pertinent quote?

The caveat they give IS them admitting to having engaged in hindsight thinking. You know how you can describe the same phenomenon with different words or synonyms, so don’t go looking for the term “hindsight thinking”.

Think man! Think!