Annoying creationists

[Paul C. Anagnostopoulos]

You forgot your strawman argument as well. Only two of the three selection conditions are set to zero. There are no moving goalposts here Paul. I started this discussion stating that your computer model shows that evolution is mathematically impossible. The only thing that has been added is the explanation why your computer model shows it is impossible. It is the competing selection processes that slow down and ultimately stop ev from converging.

But you have not proved that it stops, so the entire argument has no basis.

I am running the following experiment:

population 64
genome size 16,384
binding sites 16
weight/site widths 5/6
4 mutations per generation

These are identical to your parameters, except for 4 mutations per generation instead of 1. This should make no difference, since your argument has nothing to do with mutation rate. After 800,000 generations the mistake count is 4. It hasn't "stopped" yet. I'll let you know if it does.

Paul, it is the multiple selection conditions which slows down the convergence of ev. I’ll start posting series of cases which make it more apparent to you.

Alan, really, you're wasting your time and ours. I agree that less stringent selection pressures will allow faster evolution. That is not what I am arguing with. I am arguing with the word STOP.

Why don’t you put a two binding site set model on your to do list? Then you can evolve perfect creatures that satisfy 6 selection conditions. This would make kjkent1 happy except I doubt this model would ever converge no matter what the genome length.

So your current thesis is what? That the maximum number of selection pressures that can converge is three? Four? Five selection pressure and suddenly no convergence?

You do realize that you are saying that no evolution has ever taken place in the real world. That is a pretty extraordinary claim.

~~ Paul

 

[Paul C. Anagnostopoulos]

Finally, he applies the third stroke, and HIV's evolution comes to a practical halt. Remarkable -- the first 10 billion odd environmental stresses had no perceivable effect, and in fact, HIV evolved OUT of this huge number of stresses.

The only conclusion I can reach from this remarkable logic is that no evolution has ever occurred at all.

~~ Paul

 

[Taffer]

Alan, really, you're wasting your time and ours. I agree that less stringent selection pressures will allow faster evolution. That is not what I am arguing with. I am arguing with the word STOP.

In reality, stronger selective pressures increase the rate of evolution.

This is modeled by the time to fixation of a new mutant allele at a specific loci. The stronger the selection against that allele, the faster the ratios of that allele change (either to fixation, equilibrium or loss, depending on the type of selection).

 

[Paul C. Anagnostopoulos]

In reality, stronger selective pressures increase the rate of evolution.

This is modeled by the time to fixation of a new mutant allele at a specific loci. The stronger the selection against that allele, the faster the ratios of that allele change (either to fixation, equilibrium or loss, depending on the type of selection).

Yes, what I should say is that fewer selection pressures in Ev mean that the mistake count can reach zero faster. After all, the mistake count includes fewer sorts of mistakes.

~~ Paul

 

[delphi_ote]

In reality, stronger selective pressures increase the rate of evolution.

This is modeled by the time to fixation of a new mutant allele at a specific loci. The stronger the selection against that allele, the faster the ratios of that allele change (either to fixation, equilibrium or loss, depending on the type of selection).

Unless we're talking about a real population. High enough selection pressure will slow the rate of evolution because the population will die off.

Also, if you think about it, it's possible the earlier mutations in a series of smooth mutations to adapt to one selection pressure might be maladaptive to a different selection pressure. This is possible, but I can't think of a natural case where this is likely.

In the end, the number of selection pressures don't matter. It's the shape of the resulting fitness landscape that matters. We can only really talk about fitness landscape in a concrete way in terms of simulations like ev.

 

[Paul C. Anagnostopoulos]

I am running the following experiment:

population 64
genome size 16,384
binding sites 16
weight/site widths 5/6
4 mutations per generation

These are identical to your parameters, except for 4 mutations per generation instead of 1. This should make no difference, since your argument has nothing to do with mutation rate. After 800,000 generations the mistake count is 4. It hasn't "stopped" yet. I'll let you know if it does.

Well, it doesn't. After 1,083,137 generations, a creature with zero mistakes has emerged from the primordial depths. It has an Rsequence of 11.13, higher than the Rfrequency of 10 and lower than the Rcapacity of 12. The sequence logo is ACGACA.

Let the goalpost moving continue ...

~~ Paul

 

[Taffer]

Unless we're talking about a real population. High enough selection pressure will slow the rate of evolution because the population will die off.

Evolution is defined as a change in allele frequences. Strong selection pressures cause a fast change in allele frequences, leading ultimately to fixation or loss (generally the latter). Thus, strong selection pressures cause evolution to occur more quickly.

Also, if you think about it, it's possible the earlier mutations in a series of smooth mutations to adapt to one selection pressure might be maladaptive to a different selection pressure. This is possible, but I can't think of a natural case where this is likely.

It is possible. But, as noted above, strong selective pressures will cause beneficial alleles to quickly fix, thereby speeding up evolution. (I know what you're saying, btw, but I'm making a point with regard to kleinman.)

In the end, the number of selection pressures don't matter. It's the shape of the resulting fitness landscape that matters. We can only really talk about fitness landscape in a concrete way in terms of simulations like ev.

Very true. Although there are plenty of pop gene simulations which model fitness landscapes.

I wonder what a 50 dimensional fitness landscape would look like?

 

[kjkent1]

The only conclusion I can reach from this remarkable logic is that no evolution has ever occurred at all.

~~ Paul

Well, that's Kleinman's base conclusion -- there is no evolution -- every significant change in an organism is caused by a miracle of God.

I wonder how a miracle looks, up close, when it produces a macro-evolutionary change? Would we see a little index finger appear out of nowhere, flipping alleles? Or, would it just look like a random accidental occurrence (better known in the legal profession as an "Act of God")?

The irony of the theistic view is, that even were God actually involved in macro-evolutionary change, the most devout believer would observe nothing more than evolution.

 

[kleinman]

Annoying Creationists

So what? The ev model won’t converge with all three selection conditions imposed. Turn off two of the three selection conditions and you can evolve that condition in the genome.

No, you can't, because the other condition is physically present and its hosts are never selected against. You can produce a "perfect creature," but it's not evolved in any meaningful sense, according to the mathematics upon which ev is postulated.

Physically present? Do you think a creature with a 256 bases is physically present? Kjkent1, this is a mathematical simulation. Reducing the number of selection conditions in ev still produces meaningful base sequences that satisfy the remaining selection conditions. All ev does is a random point mutation and selection condition(s) and shows that sequences of bases can evolve that satisfy those selection condition(s) (under very restricted circumstances).

It is the competing selection conditions which slow and ultimately stop evolution in ev.

Laughably incorrect conclusion.

Not as laughable as your string cheese theory of evolution but let’s see if ev shows this. The following series is based on Dr Schneider’s published case. Starting with G=256 and a mutation rate of 1 mutation per 256 bases per generation (used for all cases in this series) and initially all weight factors for binding site mistakes, spurious binding in gene mistakes and spurious binding outside of gene set to 1. All other parameters in the model use default values.
G/ Generations to Perfect Creature
256/ 662
512/ 1251
1024/ 2991
2048/ 5382
4096/ 7715
8192/ 20237
16384/ 49963
32768/ 140958
65536/ Yours Paul
Well, what happens when we set two of the three selection conditions to 0? Let’s see what ev has to say.
G/Gens PC/bind mist/spur bind gene/spur outside gene
65536/ 1/ 1/ 0/ 0/
65536/ 5/ 0/ 1/ 0/
65536/ 18/ 0/ 0/ 1/
131072/1/ 1/ 0/ 0/
131072/17/ 0/ 1/ 0/
131072/27/ 0/ 0/ 1/
262144/3/ 1/ 0/ 0/ pop 32
262144/1/ 0/ 1/ 0/
262144/15/ 0/ 0/ 1/
In order to get the G=262144 case to run, I had to reduce the population to 32 but this case still rapidly converges for single selection conditions.

So what does this data say? It says that with three selection conditions the number of generations for convergence to a perfect creature (all selection conditions satisfied) rapidly increases as genome length increases. When you reduce the number of selection conditions to 1, even with genome lengths that are much larger than what Paul’s Rcapacity value says should converge, rapidly converge to a perfect creature for that selection condition. Multiple selection conditions slow and ultimately stop evolution in ev.

This effect is seen in reality.

BZZZZ! Incorrect!

Ask the infectious disease experts why triple antiviral agents are used to treat HIV.

This principle is used to slow if not stop the evolution of resistant strains of HIV viruses.

Is it? Let's test this spectacular logic. In the real world there are countless environmental stresses on every creature, and evolution is generally slow but inexorable. Along comes the mighty Alan Kleinman and he subjects one of these creatures, the HIV virus, to one more selective pressure, and the creature starts to evolve more slowly. Then he does it with one more stress, and the creature's evolution really starts to slow. Finally, he applies the third stroke, and HIV's evolution comes to a practical halt. Remarkable -- the first 10 billion odd environmental stresses had no perceivable effect, and in fact, HIV evolved OUT of this huge number of stresses. Then suddenly, just by coincidence, the application of three more stresses brings evolution to a dead stop -- just like Mr. Logic says that it should.

Hey, that is what ev shows and that is what the use of triple antiviral agents shows when treating HIV.

Makes sense huh? Of course a reasonable person might just conjecture that the reason why HIV was slowing down in the lab is because the anti-virals were killing most of the virus off entirely, or that the drugs were targeted in a manner which would be more likely to destroy the virus -- not necessarily because of the coincidence of environmental stresses.

It does make sense. What the results from ev show is that you must have multiple random mutations occurring at the right loci at the same time in order to satisfy the multiple selection conditions for the evolutionary process to proceed. This is very unlikely to happen as the selection conditions become more complex.

I'm sure you'll have a startling explanation for why my logical analogy is wrong and your ridiculous conclusion is correct -- and I can't wait to read it in the next episode of: "When the wine cork turns."

The explanation is not startling, it is very simple.

You forgot your strawman argument as well. Only two of the three selection conditions are set to zero. There are no moving goalposts here Paul. I started this discussion stating that your computer model shows that evolution is mathematically impossible. The only thing that has been added is the explanation why your computer model shows it is impossible. It is the competing selection processes that slow down and ultimately stop ev from converging.

But you have not proved that it stops, so the entire argument has no basis.

I am running the following experiment:

population 64
genome size 16,384
binding sites 16
weight/site widths 5/6
4 mutations per generation

These are identical to your parameters, except for 4 mutations per generation instead of 1. This should make no difference, since your argument has nothing to do with mutation rate. After 800,000 generations the mistake count is 4. It hasn't "stopped" yet. I'll let you know if it does.

Paul, there are many ways to stop the evolutionary process. You’ve just demonstrated that an extremely high mutation rate stops evolution. Another reason that evolution stops is multiple selection conditions. Try your case and see whether a using a single selection condition allows evolution to proceed more quickly.

Paul, it is the multiple selection conditions which slows down the convergence of ev. I’ll start posting series of cases which make it more apparent to you.

Alan, really, you're wasting your time and ours. I agree that less stringent selection pressures will allow faster evolution. That is not what I am arguing with. I am arguing with the word

STOP.

Paul, why should I STOP when it is your own program that shows that evolution is mathematically impossible? And the reason why the theory of evolution is mathematically impossible is that multiple selection conditions STOP evolution.

Why don’t you put a two binding site set model on your to do list? Then you can evolve perfect creatures that satisfy 6 selection conditions. This would make kjkent1 happy except I doubt this model would ever converge no matter what the genome length.

So your current thesis is what? That the maximum number of selection pressures that can converge is three? Four? Five selection pressure and suddenly no convergence?

Paul, your own program shows this. When you lengthen the genome beyond what your Rcapacity value says will converge, still converges if you eliminate any two of the three selection conditions but only for that one nonzero selection condition.

You do realize that you are saying that no evolution has ever taken place in the real world. That is a pretty extraordinary claim.

No macroevolution has ever taken place in the real world.

Finally, he applies the third stroke, and HIV's evolution comes to a practical halt. Remarkable -- the first 10 billion odd environmental stresses had no perceivable effect, and in fact, HIV evolved OUT of this huge number of stresses.

The only conclusion I can reach from this remarkable logic is that no evolution has ever occurred at all.

No macroevolution has ever occurred, that is what your computer model shows and that is what triple antiviral medications for the treatment of HIV shows. Care to describe the selection process that evolved the original HIV virus. If you say simian virus, I’ll ask how the simian virus evolved.

Alan, really, you're wasting your time and ours. I agree that less stringent selection pressures will allow faster evolution. That is not what I am arguing with. I am arguing with the word

STOP.

In reality, stronger selective pressures increase the rate of evolution.

That is correct Taffer, so what is the selective pressure that evolved the original HIV virus?

This is modeled by the time to fixation of a new mutant allele at a specific loci. The stronger the selection against that allele, the faster the ratios of that allele change (either to fixation, equilibrium or loss, depending on the type of selection).

So you think that three new mutant alleles can evolve simultaneously as quickly as three new mutant alleles evolving one after another?

In reality, stronger selective pressures increase the rate of evolution.

This is modeled by the time to fixation of a new mutant allele at a specific loci. The stronger the selection against that allele, the faster the ratios of that allele change (either to fixation, equilibrium or loss, depending on the type of selection).

Yes, what I should say is that fewer selection pressures in Ev mean that the mistake count can reach zero faster. After all, the mistake count includes fewer sorts of mistakes.

It is only a mistake if there is a selection pressure. For example, if the selection pressure is due to the lack of an essential nutrient in the diet, addition of that nutrient removes that selective pressure.

Unless we're talking about a real population. High enough selection pressure will slow the rate of evolution because the population will die off.

Also, if you think about it, it's possible the earlier mutations in a series of smooth mutations to adapt to one selection pressure might be maladaptive to a different selection pressure. This is possible, but I can't think of a natural case where this is likely.

In the end, the number of selection pressures don't matter. It's the shape of the resulting fitness landscape that matters. We can only really talk about fitness landscape in a concrete way in terms of simulations like ev.

Well said.

I am running the following experiment:

population 64
genome size 16,384
binding sites 16
weight/site widths 5/6
4 mutations per generation

These are identical to your parameters, except for 4 mutations per generation instead of 1. This should make no difference, since your argument has nothing to do with mutation rate. After 800,000 generations the mistake count is 4. It hasn't "stopped" yet. I'll let you know if it does.

Well, it doesn't. After 1,083,137 generations, a creature with zero mistakes has emerged from the primordial depths. It has an Rsequence of 11.13, higher than the Rfrequency of 10 and lower than the Rcapacity of 12. The sequence logo is ACGACA.

Let the goalpost moving continue ...

Don’t think of it as moving goalposts; just think of this as an evolving discussion. Try your example with two of the three selection conditions set to 0 and see what happens.

Unless we're talking about a real population. High enough selection pressure will slow the rate of evolution because the population will die off.

Evolution is defined as a change in allele frequences. Strong selection pressures cause a fast change in allele frequences, leading ultimately to fixation or loss (generally the latter). Thus, strong selection pressures cause evolution to occur more quickly.

Taffer, you are confusing recombination and natural selection with mutation and natural selection.

Also, if you think about it, it's possible the earlier mutations in a series of smooth mutations to adapt to one selection pressure might be maladaptive to a different selection pressure. This is possible, but I can't think of a natural case where this is likely.

It is possible. But, as noted above, strong selective pressures will cause beneficial alleles to quickly fix, thereby speeding up evolution. (I know what you're saying, btw, but I'm making a point with regard to kleinman.)

The mutations that occur with HIV are usually single base substitutions. In order to achieve resistance to three antiviral agents simultaneously, you must have all three random mutations appear simultaneously. If a particular virus manages to become resistant to one agent, it still must content with the selective pressure of the other two agents which is slowing down the reproduction of this strain resistant to the one drug. The net result either way is that multiple selective pressures slow down the evolutionary process.

In the end, the number of selection pressures don't matter. It's the shape of the resulting fitness landscape that matters. We can only really talk about fitness landscape in a concrete way in terms of simulations like ev.

Very true. Although there are plenty of pop gene simulations which model fitness landscapes.

I wonder what a 50 dimensional fitness landscape would look like?

Dr Schneider’s 3 dimensional fitness landscape already shows bad news for the theory of evolution. I suggested to Paul to modify ev to evolve two sets of binding sites with 6 selection conditions. I doubt he will do this because it is now quite apparent what ev does with 3 selection conditions. He knows I will co-opt this work.

 

[kjkent1]

Physically present? Do you think a creature with a 256 bases is physically present? Kjkent1, this is a mathematical simulation. Reducing the number of selection conditions in ev still produces meaningful base sequences that satisfy the remaining selection conditions. All ev does is a random point mutation and selection condition(s) and shows that sequences of bases can evolve that satisfy those selection condition(s) (under very restricted circumstances).

You're still missing the important point. Schneider's model is predicated on showing how Rseq approaches Rfreq over many generations of RMNS, because Rseq is found to approach Rfreq in living organisms. When you shut off one of the mistake weights, Rseq never approaches Rfreq. Instead, you get a perfect creature, nearly immediately. But the genome of this creature is not "evolved" according to Schneider's math.

In short, what you are doing by zeroing out a mistake weight is introducing design into the algorithm, i.e., you have created a perfect creature via a MIRACLE! If you could find a living organism with huge numbers of spurious and missing bindings in the real world, now THAT might convince someone of the possibility of divine intervention. What you're doing by shutting off the mistake weights is roughly equivalent to taking all of the parts of the proverbial Boeing 747, laying them next to each other on the tarmac, and then shouting "FLY!"

It's a ridiculous proposition -- what is the matter with you?

Not as laughable as your string cheese theory of evolution but let’s see if ev shows this. The following series is based on Dr Schneider’s published case. Starting with G=256 and a mutation rate of 1 mutation per 256 bases per generation (used for all cases in this series) and initially all weight factors for binding site mistakes, spurious binding in gene mistakes and spurious binding outside of gene set to 1. All other parameters in the model use default values.

G/ Generations to Perfect Creature
256/ 662
512/ 1251
1024/ 2991
2048/ 5382
4096/ 7715
8192/ 20237
16384/ 49963
32768/ 140958
65536/ Yours Paul
Well, what happens when we set two of the three selection conditions to 0? Let’s see what ev has to say.
G/Gens PC/bind mist/spur bind gene/spur outside gene
65536/ 1/ 1/ 0/ 0/
65536/ 5/ 0/ 1/ 0/
65536/ 18/ 0/ 0/ 1/
131072/1/ 1/ 0/ 0/
131072/17/ 0/ 1/ 0/
131072/27/ 0/ 0/ 1/
262144/3/ 1/ 0/ 0/ pop 32
262144/1/ 0/ 1/ 0/
262144/15/ 0/ 0/ 1/
In order to get the G=262144 case to run, I had to reduce the population to 32 but this case still rapidly converges for single selection conditions.

So what does this data say? It says that with three selection conditions the number of generations for convergence to a perfect creature (all selection conditions satisfied) rapidly increases as genome length increases. When you reduce the number of selection conditions to 1, even with genome lengths that are much larger than what Paul’s Rcapacity value says should converge, rapidly converge to a perfect creature for that selection condition. Multiple selection conditions slow and ultimately stop evolution in ev.

What it says, is, that you can't admit that you're conclusion is based on a false premise, so you're gonna keep on telling everyone you're right.

Earlier in this thread, I thought that a zero mistake weight was useful. But, after a quick post from Paul, I realized that zero mistake weights cause ev to malfunction. Then I created an experiment to test the various mistake behaviors without introducing the zero mistake. I had no idea what that experiment would show. But, the result is that missed bindings have more selective power than do spurious bindings, and spurious bindings have less selective power than missed bindings and spurious bindings in combination (see results). Also, when all the possible combinations are considered, we find that the combinations fit reasonably within a normalized distribution. Which makes a great deal of sense, because most behaviors of living things do in fact observe normal distributions, so why should selective effects be any different.

This also aligns itself with your view that adding selective effects slow down ev. This is true. What is not true, is that the effect of aggregate seletive forces is cumulative without end. Reality is that selective effects center around a mean average dependent upon their respective selective powers. The problem for the geneticist is to be able to quantify any particular selective effect in the real world. In ev, it's easy, because there are only three available selective effects, and my experiment shows exactly what happens for every one of the possible combinations.

You just don't want to recognize that you are only looking at one tail of the curve -- you're ignoring the other tail, because that tail falsifies your conclusion.

 

[kleinman]

Annoying Creationists

Physically present? Do you think a creature with a 256 bases is physically present? Kjkent1, this is a mathematical simulation. Reducing the number of selection conditions in ev still produces meaningful base sequences that satisfy the remaining selection conditions. All ev does is a random point mutation and selection condition(s) and shows that sequences of bases can evolve that satisfy those selection condition(s) (under very restricted circumstances).

You're still missing the important point. Schneider's model is predicated on showing how Rseq approaches Rfreq over many generations of RMNS, because Rseq is found to approach Rfreq in living organisms. When you shut off one of the mistake weights, Rseq never approaches Rfreq. Instead, you get a perfect creature, nearly immediately. But the genome of this creature is not "evolved" according to Schneider's math.

Selection in this model is based on mistakes, whether they be failed identification of binding sites or spurious identification of binding sites. Dr Schneider’s selection process shows a coincidental finding where Rsequence approaches Rfrequency when using all three of his selection conditions and the genome length is short enough. Selection is not based on this condition. I don’t know whether this coincidental finding has any real physical significance. Bottom line is that Dr Schneider’s model shows more than just this mathematical peculiarity when you do the parametric studies that he suggested be done. The most significant finding is that multiple selection processes slow and ultimately stop evolution in his model. This finding does have real physical significance.

In short, what you are doing by zeroing out a mistake weight is introducing design into the algorithm, i.e., you have created a perfect creature via a MIRACLE! If you could find a living organism with huge numbers of spurious and missing bindings in the real world, now THAT might convince someone of the possibility of divine intervention. What you're doing by shutting off the mistake weights is roughly equivalent to taking all of the parts of the proverbial Boeing 747, laying them next to each other on the tarmac, and then shouting "FLY!"

I remove selection conditions from Dr Schneider’s computer model and I have introduced design into the model? Your analogy of Dr Schneider’s computer model and a Boeing 747 is strange considering how highly designed a 747 is.

It's a ridiculous proposition -- what is the matter with you?

Hey, it is your proposition comparing ev with a Boeing 747.

Not as laughable as your string cheese theory of evolution but let’s see if ev shows this. The following series is based on Dr Schneider’s published case. Starting with G=256 and a mutation rate of 1 mutation per 256 bases per generation (used for all cases in this series) and initially all weight factors for binding site mistakes, spurious binding in gene mistakes and spurious binding outside of gene set to 1. All other parameters in the model use default values.

What it says, is, that you can't admit that you're conclusion is based on a false premise, so you're gonna keep on telling everyone you're right.

Kjkent1, don’t confuse the fact that you don’t understand my premise as it being a false premise. Ev is a stylized model of random point mutation and natural selection and under certain circumstances you can evolve sequences of bases that satisfy its three selection conditions. Under more realistic circumstances (realistic genome lengths) ev fails to converge with its three selection conditions but still can evolve sequences of bases that satisfy one selection condition. The conclusion is that multiple selection conditions slow and ultimately stop evolution. This what ev shows and this effect is seen in reality.

Earlier in this thread, I thought that a zero mistake weight was useful. But, after a quick post from Paul, I realized that zero mistake weights cause ev to malfunction. Then I created an experiment to test the various mistake behaviors without introducing the zero mistake. I had no idea what that experiment would show. But, the result is that missed bindings have more selective power than do spurious bindings, and spurious bindings have less selective power than missed bindings and spurious bindings in combination (see results). Also, when all the possible combinations are considered, we find that the combinations fit reasonably within a normalized distribution. Which makes a great deal of sense, because most behaviors of living things do in fact observe normal distributions, so why should selective effects be any different.

The only thing that ev shows malfunctioning is the theory of evolution by mutation and natural selection because multiple selection conditions slow and ultimately stop evolution. Since you have trouble understanding this concept, why don’t you ask Paul to modify ev to evolve two sets of binding sites with 6 selection conditions, two conditions for missed binding sites, two conditions for spurious binding the gene and two conditions for spurious binding outside the gene. What do you think will happen in this simulation?

This also aligns itself with your view that adding selective effects slow down ev. This is true. What is not true, is that the effect of aggregate seletive forces is cumulative without end. Reality is that selective effects center around a mean average dependent upon their respective selective powers. The problem for the geneticist is to be able to quantify any particular selective effect in the real world. In ev, it's easy, because there are only three available selective effects, and my experiment shows exactly what happens for every one of the possible combinations.

Explain this to Paul so he can put this in ev.

You just don't want to recognize that you are only looking at one tail of the curve -- you're ignoring the other tail, because that tail falsifies your conclusion.

Which alternative universe are you telling your tale from?

 

[Paul C. Anagnostopoulos]

Don’t think of it as moving goalposts; just think of this as an evolving discussion. Try your example with two of the three selection conditions set to 0 and see what happens.

Are you ever going to acknowledge that you are a ****ing liar? You said, multiple times, that those particular parameters cause evolution to stop, halt, hang up, never happen. I just showed that you are wrong. In fact, you never ran that particular case long enough to find out whether a perfect creature evolved. And then you wasted our time blathering on about multiple selection pressures causing evolution to halt.

You’ve just demonstrated that an extremely high mutation rate stops evolution.

Are you seriously going to argue that dropping the mutation rate by a factor of 4 would stop the perfect creature from evolving? You recall that we ran a series of experiments that showed that generations to perfection varies linearly with mutation rate.

Notice how you just moved the goalpost to include mutation rate.

~~ Paul

 

[kjkent1]

Selection in this model is based on mistakes, whether they be failed identification of binding sites or spurious identification of binding sites. Dr Schneider’s selection process shows a coincidental finding where Rsequence approaches Rfrequency when using all three of his selection conditions and the genome length is short enough. Selection is not based on this condition. I don’t know whether this coincidental finding has any real physical significance.

According to the ev paper, "The average distance betwen accetor sites is the average size of introns plus exons, or ~812 bases, so the information needed to find the acceptors is Rfrequency = log₂812 = 9.7. By comparison, Rseqence = 9.4 bits, so in this and other genetic systems, Rsequence is close to Rfrequency."

Schneider's ev paper is intended to "demonstrate that Rseqence can indeed evolve to match Rfrequency." You're the person who throughout this thread has advocated that ev is an accurate model of evolution. If true, then you cannot now argue that it is mere coincidence that Rseq approaches Rfreq, because to do so, is to reject the accuracy of the ev model. This is incredibly disingenuous, and it frankly suggests that I've got you up against the ropes, because you're so obviously wrong right now, and you know you are.

Bottom line is that Dr Schneider’s model shows more than just this mathematical peculiarity when you do the parametric studies that he suggested be done. The most significant finding is that multiple selection processes slow and ultimately stop evolution in his model. This finding does have real physical significance.

If this were actually true, then based on your findings of evolution producing perfect creatures long before there is any reasonable convergence from Rseq to Rfreq, we would find no creatures in nature with DNA sequences wherein Rseq is found to approach Rfreq. And, if this is true, then Schneider would never have written ev or the paper because he would not have found this measurement apparent in living organisms.

Checkmate.

Your experiments which produce perfect creatures wherein there is no approach of Rseq to Rfreq do not model real life. Therefore, your experiments cannot be used to extrapolate to real-world evolution. The genomes of your proposed perfect creatures are not viable, according to the principles of evolution stated in the ev paper.

I remove selection conditions from Dr Schneider’s computer model and I have introduced design into the model? Your analogy of Dr Schneider’s computer model and a Boeing 747 is strange considering how highly designed a 747 is.

No, you set a selective condition to zero and thereby create a "perfect creature" whose genome is completely inconsistent with any known living organism.

Your creature is magical, because it did not evolve -- it was designed -- by you.

 

[kleinman]

Annoying Creationists

Schneider's ev paper is intended to "demonstrate that Rseqence can indeed evolve to match Rfrequency." You're the person who throughout this thread has advocated that ev is an accurate model of evolution. If true, then you cannot now argue that it is mere coincidence that Rseq approaches Rfreq, because to do so, is to reject the accuracy of the ev model. This is incredibly disingenuous, and it frankly suggests that I've got you up against the ropes, because you're so obviously wrong right now, and you know you are.

You may have me up against the string cheese but not the ropes. I love it when you evolutionists want to argue if ev is an accurate model of evolution. I have always called this a plausible model which captures some elements of reality. But let’s let Dr Schneider speak for himself.
The following quotes were taken from Dr Schneider’s blog web page: http://www.lecb.ncifcrf.gov/~toms/paper/ev/blog-ev.html

The following are Dr Schneider’s responses to a critique of his paper Evolution of biological information by Dr Stephen E Jones.

"Schneider's paper is misleadingly titled: "Evolution of biological information". But it is just a *computer* simulation. No actual *biological* materials (e.g. genomes of nucleic acids, proteins, etc) were used, nor does Schneider propose that his simulation be tested with *real* genomes or proteins

Actual biological materials were used to determine the original hypothesis. Read the literature: Schneider1986

It only becomes *real* biological information and random mutation and natural selection, when the simulation is tested in the *real* world, using *real* DNA, proteins, with *real* mutations and a *real* environment does the selecting. It is significant that Schneider does not propose this, presumably because he knows it wouldn't work.

You are very bad at reading my mind, I have considered doing this experiment. Given the right conditions, it WILL WORK. Do you have th gumption to do the experiment yourself? That's the way real science works! FURTHERMORE, if you read the literature, you will recognize that related experiments have been repeatedly done for 20 years. Look up SELEX.

In the rest of the paper he uses the single word "selection". I take this as a tacit admission that his model is not a simulation of *real* biological natural selection.

No. A rose is a rose by any other name. Selection is selection whether it be natural (generally meaning the environment of earth), breeding (by humans usually, though perhaps some ants select their fungi), SELEX or in a computer simulation. Of COURSE it is a simulation of natural selection! The paper would not be relevant to biology and would not have been published in a major scientific journal if it were not!

Schneider lets slip that there is another unrealistic element in his (and indeed all) computer simulations in that it (they) "does not correlate with time":

So? Run the program slower if you want. Make one generation per 20 minutes to match rapid bacterial growth. THIS WILL NOT CHANGE THE FINIAL RESULT!

Well, when Schneider's simulation is actually tested with *real* "life" (e.g. a bacterium), and under *real* mutation and natural selection it gains information, then, and only then, would "creationists" be favourably impressed. But if they are like me, they would already be impressed (but unfavourably) that Schneider does not mention in his paper that his simulation should now be so tested in the *real* "biological" world.

1. The simulation was of phenomena in the "real" world.
2. Dr. Jones is invited yet again to do an experiment.

The following is a response Dr Schneider made to a statement made by David Berlinski.

Where attempts to replicate Darwinian evolution on the computer have been successful, they have not used classical Darwinian principles, and where they have used such principles, they have not been successful.

The ev program disproves this statement since it uses classical Darwinian principles and was successful.

I don’t go quite as far as Dr Schneider has gone but his model does properly simulate the effects of multiple selection conditions.

Bottom line is that Dr Schneider’s model shows more than just this mathematical peculiarity when you do the parametric studies that he suggested be done. The most significant finding is that multiple selection processes slow and ultimately stop evolution in his model. This finding does have real physical significance.

If this were actually true, then based on your findings of evolution producing perfect creatures long before there is any reasonable convergence from Rseq to Rfreq, we would find no creatures in nature with DNA sequences wherein Rseq is found to approach Rfreq. And, if this is true, then Schneider would never have written ev or the paper because he would not have found this measurement apparent in living organisms.

We all know that evolutionists never misinterpret their measurements.

Checkmate.

The only problem here is you are playing checkers.

Your experiments which produce perfect creatures wherein there is no approach of Rseq to Rfreq do not model real life. Therefore, your experiments cannot be used to extrapolate to real-world evolution. The genomes of your proposed perfect creatures are not viable, according to the principles of evolution stated in the ev paper.

If you want to model real life, tell us what the selection process is that evolved the original HIV virus.

I remove selection conditions from Dr Schneider’s computer model and I have introduced design into the model? Your analogy of Dr Schneider’s computer model and a Boeing 747 is strange considering how highly designed a 747 is.

No, you set a selective condition to zero and thereby create a "perfect creature" whose genome is completely inconsistent with any known living organism.

You’ve just proved that ev also models abiogenesis.

Your creature is magical, because it did not evolve -- it was designed -- by you.

Are you sure I didn’t create it, just to annoy you.

 

[kjkent1]

You may have me up against the string cheese but not the ropes.

(kleinman follows with a cacophony of irrelevancy designed to obfuscate from the fact that his argument is completely destroyed -- redacted to protect the sanity of the audience.)

Alan, your arguments are "a tale full of sound and fury, signifying nothing." Shakespeare, Macbeth, Act 5, Scene 5.

If you decide to come back to reality, let me know. Your current argument is frivolous.

 

[Mercutio]

Alan, your arguments are "a tale full of sound and fury, signifying nothing." Shakespeare, Macbeth, Act 5, Scene 5.

Um... you left out 4 words. Whether you were being polite or ironic is left for the reader to discover.

 

[kleinman]

Annoying Creationists

Don’t think of it as moving goalposts; just think of this as an evolving discussion. Try your example with two of the three selection conditions set to 0 and see what happens.

Are you ever going to acknowledge that you are a ****ing liar? You said, multiple times, that those particular parameters cause evolution to stop, halt, hang up, never happen. I just showed that you are wrong. In fact, you never ran that particular case long enough to find out whether a perfect creature evolved. And then you wasted our time blathering on about multiple selection pressures causing evolution to halt.

Paul, it is your theory of evolution which is the lie. Your own computer model shows this. Paul you can list lots of other ways of stopping ev from converging but it is the multiple selection pressures which explains why ev stops evolving as you lengthen the genome not your silly Rcapacity. This issue is not going away because it is the explanation why the theory of evolution is mathematically impossible. It is easily shown by eliminating two of the three selection conditions that ev will then evolve the third condition. The theory of evolution is mathematically impossible because competing selection conditions slow and ultimately stop evolution. There are no mutation mechanisms that will change this mathematical fact. You need to learn to live with this mathematical fact, your own computer model shows this. By the way, did I tell you how good a job you did on programming ev?

You’ve just demonstrated that an extremely high mutation rate stops evolution.

Are you seriously going to argue that dropping the mutation rate by a factor of 4 would stop the perfect creature from evolving? You recall that we ran a series of experiments that showed that generations to perfection varies linearly with mutation rate.

Your experiments examining the effect of mutation rate are incomplete. If you examine a wider range of mutation rates you will find the mutation rate/generations for convergence curve is actually paraboloid, not linear. If you want, I’ll repost the data that shows this.

Notice how you just moved the goalpost to include mutation rate.

Stop whining Paul; it is unbecoming for a moderator on the James Randi forum. You should be proud that your own computer program proves the theory of evolution to be mathematically impossible and the reason it is impossible is that conflicting selection conditions slow and then ultimately stop evolution.

 

[Paul C. Anagnostopoulos]

Paul, it is your theory of evolution which is the lie. Your own computer model shows this. Paul you can list lots of other ways of stopping ev from converging but it is the multiple selection pressures which explains why ev stops evolving as you lengthen the genome not your silly Rcapacity.

Alan, you must have been quite good at dodgeball. Let's cease talking about the reasons why evolution stops, and ponder the more fundamental question of whether it stops at all. Could you present a complete list of the experiments you ran that you think stopped? In addition to the parameters, please tell us how many generations you ran before you concluded that evolution had stopped.

Your experiments examining the effect of mutation rate are incomplete. If you examine a wider range of mutation rates you will find the mutation rate/generations for convergence curve is actually paraboloid, not linear. If you want, I’ll repost the data that shows this.

Yes, please do. I ran a series of experiments with a genome size of 1000 and mutation rates from 1/1,000,000 bases down to 1/200 bases. The results fit g = 8.7 m^1.0 with r = .996.

~~ Paul

 

[kleinman]

Annoying Creationists

Paul, it is your theory of evolution which is the lie. Your own computer model shows this. Paul you can list lots of other ways of stopping ev from converging but it is the multiple selection pressures which explains why ev stops evolving as you lengthen the genome not your silly Rcapacity.

Alan, you must have been quite good at dodgeball. Let's cease talking about the reasons why evolution stops, and ponder the more fundamental question of whether it stops at all. Could you present a complete list of the experiments you ran that you think stopped? In addition to the parameters, please tell us how many generations you ran before you concluded that evolution had stopped.

Ok, let’s talk about whether evolution really stops. Clearly extinction stops the evolution of that genetic line. If a genetic line does not go extinct, microevolutionary process don’t stop. These are the type of events that are seen for example with the evolution of drug resistance in the HIV virus. What ev shows is that macroevolutionary processes can not succeed because of the competing selection conditions. So how do you get microevolutionary processes to accumulate to make a macroevolutionary change? The answer to that question is you don’t. There are no selection processes that lead to macroevolutionary changes and if there were, they would be slowed profoundly by competing selection conditions as shown by your computer model. Evolutionists have mistakenly extrapolated microevolutionary processes to macroevolution. There are no selection processes to do this.

Your experiments examining the effect of mutation rate are incomplete. If you examine a wider range of mutation rates you will find the mutation rate/generations for convergence curve is actually paraboloid, not linear. If you want, I’ll repost the data that shows this.

Yes, please do. I ran a series of experiments with a genome size of 1000 and mutation rates from 1/1,000,000 bases down to 1/200 bases. The results fit g = 8.7 m^1.0 with r = .996.

Ok, here is a typical series, G=1024, population=64, binding site width=6
Muts/G Generations
1 / 10108
2 / 6669
3 / 3432
4 / 2546
5 / 1268
6 / 1874
7 / 2147
8 / 3626
9 / 15351
10 / 81112

Try 1 mutation per 100 bases in your series and I think you will see the paraboloid behavior of the curve.

Here’s another series, G=2048
Mut/G Generations
1 / 35486
2 / 21666
3 / 8697
4 / 6001
5 / 7501
6 / 6143
7 / 3425
8 / 4998
9 / 4154
10 / 4043
11 / 8250
12 / 3570
13 / 4278
14 / 3650
15 / 5546
16 / 9424
17 / 38333
18 / 11250
19 / did not converge
You may get a linear fit in the range that you examined but if you increase the mutation rate further than what you examined you will see the paraboloid nature of the curve.

 

[Taffer]

No macroevolution has ever taken place in the real world.

Macroevolution is not a meaningful distinction in evolutionary theory.

No macroevolution has ever occurred, that is what your computer model shows and that is what triple antiviral medications for the treatment of HIV shows. Care to describe the selection process that evolved the original HIV virus. If you say simian virus, I’ll ask how the simian virus evolved.

Triple antiretrovirals have nothing to do with 'macroevolution', nor do they have anything to do with the subject at hand. They are used because three antiretrovirals kill faster then one. They also decrease the risk of any one resistance evolving at a particular time, but it does not decrease the probability of triple resistance mutants arising

That is correct Taffer, so what is the selective pressure that evolved the original HIV?

I believe there is lots of evidence for the evolution of the HIV virus. See here for a very brief run down on the origins of HIV.

So you think that three new mutant alleles can evolve simultaneously as quickly as three new mutant alleles evolving one after another?

Yes, of course. The chance of a mutant allele arrising at any one loci is fixed and independant on the number of other mutations arrising on the genome.

Taffer, you are confusing recombination and natural selection with mutation and natural selection.

No, I am not. Given that I am, in essence, a population geneticist (ok, a phylogeneticist, but same difference ), I trust my own studies over what you say. Secondly, evolution is defined as the change in alleles over time. There is no difference between "recombination and natural selection" from "mutation and natural selection". Natural selection works on any existing variation in a population, no matter how that variation arose.

The mutations that occur with HIV are usually single base substitutions. In order to achieve resistance to three antiviral agents simultaneously, you must have all three random mutations appear simultaneously. If a particular virus manages to become resistant to one agent, it still must content with the selective pressure of the other two agents which is slowing down the reproduction of this strain resistant to the one drug. The net result either way is that multiple selective pressures slow down the evolutionary process.

No it isn't, and you obviously do not understand how antivirals work. Most, if not all, antivirals do not lead to a complete death of the virus. The slow the rate of viral infection, but not 100% of viruses are killed. Thus, a virus which evolves resistance to a single antiretroviral agent will have an advantage over its peers, as it is only being selected for by two selection pressures, not three. Stop thinking in black-and-whites. Biology does not work that way.

Dr Schneider’s 3 dimensional fitness landscape already shows bad news for the theory of evolution. I suggested to Paul to modify ev to evolve two sets of binding sites with 6 selection conditions. I doubt he will do this because it is now quite apparent what ev does with 3 selection conditions. He knows I will co-opt this work.

Whatever, kleinman. Since you have shown to have little to know understanding of even basic microbiology, evolutionary theory or population theory, I will trust population and evolutionary genetics over what you claim anyday. Since I have done the maths myself, I know 3 dimensional fitness landscapes work.