What you don’t think is still not an answer to my question. What is the new function, and why do you think it’s 1000 substitutions away? Whether you think it got there by design or by accumulated mutations is irrelevant here. I want to know how you determined it has a new function and how you know that all those 1000 differences are required for it.
1 Like
Simply comparing the sequences is evidence of the number of substitutions for the new function. A Rum “how do you know” argument that avoids an objective view of the evidence is not interesting here as the differences themselves are problematic to the duplication and divergence claim based on the risk of null mutations and the number of differences between the sequences.
The different function is identified in uniprot. One is generic cellular function the other is muscle function.
Why? The fact that there are 1000 differences observed doesn’t tell you which ones, or how many, are required for any new function it might have.
Again, merely observing the number of changes there currently are in the protein doesn’t tell you how many of them are required for any purported new function it might or might not have.
1 Like
It does when like MYH proteins in different animals are well preserved.
If all you observe is the number of changes this is true but the what is also important is how many AA changes on average a protein can withstand without becoming non functional.
Why? The fact that there are 1000 differences observed doesn’t tell you which ones, or how many, are required for any new function it might have.
That also doesn’t tell you how many of the differences between proteins are strictly required for any putative new function.
The data on variants is building in uniprot if you look under disease and variants for these human proteins.
It tells you traveling to new AA structures by mutation/fixation is restricted. This is all you need to know when you are observing differences over a couple of dozen AAs. The preservation data tells you the same thing.
Do you believe that all MYH proteins share a common ancestor?
Bill, as long as you just toss off minimal one-liners, nobody is going to know what you’re talking about. You need to assemble a reasoned argument, step by step, skipping no steps. Try that.
2 Likes
But you’re ignoring how changing residues (they aren’t amino acids when incorporated into proteins) CHANGES function. You’re desperately trying to pretend that this is a binary thing. None of it is.
Hi John
Here is the most important point when looking at Mike and David’s simulation. The issue is not just finding a MR feature it is avoiding a null mutation. The odds of a null mutation goes up with every substitution.
This is the problem with the common ancestry claim of proteins separated by more then a few dozen substitutions.
The model is most sensitive to the value of λ—the number of loci that must mutate before a new MR function occurs—which appears as an exponent in equation 4. If in the case just mentioned, because of the particular initial sequence of the parent gene, either three or nine nucleotide changes were necessary instead of six, then the population sizes required to fix the feature in 10^8 generations would vary from 10^11 to 10^31 organisms.
.
OMG Bill, do you think that anyone is suggesting that a duplicate stops being under purifying selection and then it just drifts into a new function after dozens, or hundreds of mutations?
We all agree if purifying selection stops, then the gene likely drifts into nonfunctionality before it gains a new function.
The idea is that either:
-
Pre-duplication, the ancestral gene gains a new function (so it now has 1 more than however more it used to), then gets duplicated allowing the individual genes to specialize to their requisite functions. This is how many of the divergence trajectories elucidated by ancestral sequence reconstruction have happened.
-
Or alternatively, following 1 or more duplications, all the genes still evolve under purifying selection (meaning the deleterious and null mutations are removed, allowing neutral and beneficial ones to accumulate) because they still carry out a fitness-enhancing function (the increased copy-numbers are maintained by a beneficial dosage effect), and then 1 or more of the additional copies gains a novel function through accumulation of neutral or beneficial mutations.
The model you have in your head that the gene loses purifying selection, then drifts into a new function a thousand mutations away before being rendered nonfunctional, is fatuous. If that was really the only way a duplicate gene could gain a novel function, then we’d all think that would be extremely improbable. It’s just that that really is a fatuous model. Who in their right mind would propose such a model? Only a creationist lunatic would think biologists are proposing something like that is what happens.
2 Likes
If that was the only way to get a novel functional gene through duplication and divergence, that would be mad.
It’s just that, it isn’t. So there’s no problem with common ancestry. There’s a problem with the assumptions going into the model.
2 Likes
1 Like
This is not the model. Purifying selection does not go away. It eliminates null alleles and at some point (after a few mutations) starts becoming a strong impediment to further variation…
How do you think that purifying selection helps with the variation needed to evolve a new function in realistic evolutionary time scales?
In diploids and other polyploids, it typically doesn’t. That’s basic pop gen and basic math, maybe even basic enough for you.
How so? Show us how, mathematically.
In Behe’s model it does. It is nonexistant in the duplicates. Explicitly so. In his model the duplicate just accumulates mutations until a null mutation hits, in which case the gene is rendered nonfunctional. This immediately puts the other gene under selection to reject new null mutations.
Remember, in Behe’s model the null mutations are required to accumulate in sets of 2 or more to create the new function.
Not in Behe’s model. One of the copied genes is free from selection. When this gains a null mutation, the other gene then comes under selection against the new-function-creating mutations(which are null mutations.)
In Behe’s model, a specific set of new-function-creating null mutations are supposed to occur in sets of two or more. Either this has to happen by first inactivating one of the copies and then waiting for the other specific null-mutation that creates the new function in conjunction with the previous specific null mutation, before other null mutations happen elsewhere in the inactivated gene. The other gene copy comes under selection against each of the new function-creating mutations when they occur individually.
Or alternatively, the novel function evolves when in one of the two genes, the two or more specific null mutations that create the new function when together, occur in the same individual by luck.
Quoting Behe & Snoke 2004:
The basic “task” that the model asks a duplicate gene to perform is to accumulate λ mutations at the correct nucleotide positions to code for a new selectable feature before suffering a null mutation.
The process we envision for the production of a multiresidue (MR) feature is illustrated in Figure 1, where a duplicate gene coding for a protein is represented as an array of squares that stand for nucleotide positions.
However, if several point mutations (indicated by a “+” in the figure) accumulate at specific nucleotide positions (indicated by the three squares outlined in blue in the figure) in the gene coding for the protein before a null mutation occurs elsewhere in the gene (indicated by a red “X”), then several amino acid residues will have been altered and the new selectable MR feature will have been successfully built in the protein (indicated by the green-shaded area). By hypothesis, the gene is not selectable for the new feature when an intermediate number of mutations has occurred, but only when all sites are in the correct state.
The pertinent feature of the model is that multiple changes are required in the gene before the new, selectable feature appears. Changes in these nucleotide positions are assumed to be individually disruptive of the original function of the protein but are assumed either to enhance the original function or to confer a new function once all are in the compatible state. Thus, the mutations would be strongly selected against in an unduplicated gene, because its function would be disrupted and no duplicate would be available to back up the function.
That’s Behe’s model. That’s why it’s insane. The model is insane.
Also wrong. Not Behe’s model, and also not how real protein divergence works. Behe correctly allows neutral mutations to accumulate indefinitely.
In real biology, purifying selection removes deleterious mutations. Neutral and beneficial ones still accumulate. Roughly half of amino acid substitutions in a typical protein (there is of course large variation here) are conditionally neutral. An occasional rare one is beneficial. These types of mutations are allowed to accumulate basically indefinitely. So the protein accumulates mutations over time as generations pass, and sooner or later it wanders into a new function that selection acts on because it has a positive fitness effect. That’s it.
Typically, for real proteins, it just takes a few (meaning on the order of one to five in most cases) neutral or beneficial mutations to wander into a new function. They really do overlap very closely in sequence space. I have cited endless amounts of literature on this to you before.
Meanwhile, it’s time for you to realize how absolutely nuts Behe’s model is.
2 Likes
So there’s two genes A and B.
- Mutations occur.
- Most mutations are not the ones we are supposed to be “waiting for”, there’s only a few of those, and way more mutations we are not “waiting for”.
- So chances are a null mutation occurs in either A or B that isn’t one of the ones we are waiting for.
- One of these happens in A. A is now dead due to the null mutation we weren’t waiting for. B is now the only gene carrying out the original function.
- One of the mutations assumed to be individually disruptive of the original function occurs in one of the genes (say A).
- But that’s irrelevant, because A is already dead to the previous null-mutation, and having the second specific null mutation occur now would be useless because A already suffered a deactivating null mutation elsewhere.
- The mutation in A is not removed by purifying selection because the other gene B is there to carry on the function.
- If the mutation occurs in B, it is removed by purifying selection, because B is now the only gene carrying out the function. Gene A died, remember?
- So B cannot evolve the new function except by a lucky double-mutation.
- A continues to accumulate mutations, but it’s dead. So we just wait and wait and wait until we get lucky that reversals of null-mutations elsewhere—which occur approximately 1000 times more often than the mutations we’re “waiting for” (fraction represented by the ρ symbol) —occur by chance. This basically never happens.
Same scenario happens the vast majority of the time the simulation is run. Once in a blue moon we get a lucky double mutant. Or we get to the state where A gains the first mutation we are waiting fore, it then deactivates, and then by sheer luck we get the 2nd mutation in A we are waiting for before the 800 times more likely null-mutation elsewhere renders a resurrection of A hopeless. When the results of the simulations are taken together, the mean time to the new function is in billions of generations in populations of trillions of individuals.
We are all deeply impressed by this utterly psychotic model scenario.
2 Likes
Continuing:
The other nucleotide positions in the gene, corresponding to the black squares in Figure 1, which if they were changed would yield a null allele, are represented only implicitly in our computer model by the constant ρ, which is the ratio of the number of mutations of the original duplicated gene that would produce a null allele to the number of mutations of the original duplicated gene that would yield a compatible residue. (Definitions of terms are given in Table 1.) As an example, consider a gene of a thousand nucleotides. If a total of 2400 point mutations of those positions would yield a null allele, whereas three positions must be changed to build a new MR feature such as a disulfide bond, then ρ would be 2400/3, or 800.
So that’s where the null-mutations-elsewhere-are-800-times-more-frequent number comes from. It’s actually worse because we see in their figure 6 legend that they use a ratio for ρ of 1000, not 800.
1 Like
I think your number of roughly half maybe a little aggressive but I will yield here for arguments sake.
Can you describe the process of how you think purifying selection removes deleterious mutations in vertebrates especially deer?
In the duplicated gene deleterious mutations will accumulate since there is no selective pressure on the duplicated gene.
This may be true for a few bacterial enzymes but may not be relevant to the problem of novel proteins in deer. Certainly this is not true for proteins like MYH and WNT.
I notice you’ve suddenly stopped defending Behe’s model.
Do you truly find you have trouble comprehending how deleterious mutations are removed from a population by selection? Do you understand what the word deleterious means?
That’s the model assumption we are saying is wrong by observational reality. Many duplicate genes are in reality maintained by selection because they contribute a dosage effect when the genes are expressed. Many genes are actually present in our genomes in many copies, and each of the copies are all maintained by purifying selection because their functions are beneficial.
That’s why we have many copies of the genes encoding the ribosome, for example, as many ribosomes need to be constantly constructed in parallel to maintain their high numbers in the cell.
Bill now you’re just making stuff up. Zero evidence that eukaryotic genes are different from bacterial ones in terms of how many mutation it takes for new functions to emerge in them.
2 Likes
Then you should have no trouble describing this beyond the conceptual level.
Are they maintained after a null mutation?
There is plenty of evidence against your claim including how rarely these sequence are close to each other. How does your model explain genes of the same family being over 100 substitutions apart?
There are also many different functions in multicellular eukaryotic cells that don’t exist in prokaryotic cells.