Register | Sign In


Understanding through Discussion


EvC Forum active members: 65 (9162 total)
4 online now:
Newest Member: popoi
Post Volume: Total: 915,818 Year: 3,075/9,624 Month: 920/1,588 Week: 103/223 Day: 1/13 Hour: 0/1


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Explaining the pro-Evolution position
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 343 of 393 (792865)
10-14-2016 4:41 PM
Reply to: Message 335 by Dr Adequate
10-14-2016 3:38 PM


Re: Lenski
quote:
Lenski's experiment works because he doesn't starve his populations to death.
Yes. Extinction does stop evolution, it's probably the only thing that can.
rmns is suppressed long before extinction occurs. Even if the population can survive multiple different simultaneous selection pressures, you won't get the amplification necessary for rmns to work efficiently. That's what all the real, measurable and repeatable examples of rmns show.
quote:
So if you put starvation pressure and thermal stress on the population at the same time, well you figure it out.
I figured it out, it turns out that the results of the experiment depend on how hard the selection is.
So let's say you put an itsy bitsy little bit of thermal stress on the Lenski bacteria, do you think the evolutionary process will speed up (more beneficial mutations per generation) of slow down? Do you think the evolutionary trajectories and the mutations for adaptation depend on the intensity of selection?
quote:
Why doesn't anything evolve resistance to Iodine? The reason is that Iodine is a very reactive molecule binding to all kinds of biological molecules, denaturing the molecules, far too many targets for rmns to have any chance for a replicator to evolve resistance to this chemical.
Yes, it's the ultimate in hard selection --- it kills everything.
Actually, you might find a target gene, you might get a spore that can survive.
quote:
That's not an answer to my point. Let me ask again --- how would make a difference in the math whether you have one selection pressure acting on many loci, or a different selection pressure for each of the loci?
There's no real difference in the math. You are still dealing with nested binomial probability problems. We can get a sense what goes with a selection pressure that target many genes with the Lenski experiment. The improvement in fitness is so small with each beneficial mutation that amplification is very difficult for that slightly more fit variant due to competition with other variants. Lenski is running his experiment in 10ml tubes. If he ran the experiment in 1 liter containers, you might see amplification occur more quickly because you are allowing for larger populations than his e7-e8.
On the other hand, you can see from the video of bacteria evolving resistance on the large petri dish, you don't have the environmental resource limitation, only the targeted antibiotic selection pressure so that rmns process works very quickly.

This message is a reply to:
 Message 335 by Dr Adequate, posted 10-14-2016 3:38 PM Dr Adequate has not replied

Replies to this message:
 Message 346 by Taq, posted 10-14-2016 4:50 PM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 349 of 393 (792871)
10-14-2016 5:02 PM
Reply to: Message 336 by bluegenes
10-14-2016 3:40 PM


Re: Lenski: Cit+ isn't slow.
quote:
Kleinman writes:
Don't get me wrong, there's more than one way replicators can adapt to selection pressures other than rmns. Recombination is a much faster way replicators can adapt and they can do it to multiple selection pressures simultaneously. But they have to have the correct alleles already in the gene pool.
On the other hand, rmns is the creation of new alleles in order to adapt. And if the adaptation requires the creation of multiple different new alleles at different genetic loci due to multiple different selection pressures simultaneously, the chances of adaptation are extremely low and the process is extremely slow if it going to happen (see the Lenski experiment for an empirical example).
Very rapid adaptations involving new alleles occurred in all cultures in the Lenski experiment. But perhaps you are thinking of the complicated sequence of mutations that produced the Cit+ strain in just one of the twelve cultures after ~30,000 generations? Is that what you are thinking of as slow?
If so, here's something to consider. The Lenski cultures are all in the same environment and undergo the same processes. These are not actually very conducive to the emergence of Cit+ as a dominant strain. Change the circumstances, and Cit+ organisms can emerge in less than 100 generations. The potentiating, actualizing and refining mutations, a considerable sequence, can all take place in one culture in a few weeks.
Did your probability calculations tell you that?
E. coli long-term evolution experiment - Wikipedia
quote:
Although the bacteria in each population are thought to have generated hundreds of millions of mutations over the first 20,000 generations, Lenski has estimated that within this time frame, only 10 to 20 beneficial mutations achieved fixation in each population, with fewer than 100 total point mutations (including neutral mutations) reaching fixation in each population.[6]
That's about 1 beneficial mutation per thousand generations. If you think that's fast, would you argue about Haldane's dilemma?
As for the Citrate metabolizer (e coli has alway been able to metabolize Citrate because it has a Krebs cycle), appeared at about generation 31500, not at 100 generations. The problem is not in the citric acid cycle for e coli, it is in the ability to transport citrate in the presence of oxygen. If I recall, those variants were spherical forms and so it may be that a mutation causing a defective cell wall allows the citrate to enter.
And yes, my equations include the possibility for this type of more rare mutation.

This message is a reply to:
 Message 336 by bluegenes, posted 10-14-2016 3:40 PM bluegenes has replied

Replies to this message:
 Message 352 by Dr Adequate, posted 10-14-2016 5:13 PM Kleinman has replied
 Message 358 by bluegenes, posted 10-14-2016 6:00 PM Kleinman has replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 351 of 393 (792873)
10-14-2016 5:12 PM
Reply to: Message 337 by Dr Adequate
10-14-2016 3:40 PM


Re: Is it summation time?
quote:
How does evolution of drug resistance differ than evolution by rmns to any other kind of selection pressure?
Evolution to drug resistance occurs in the face of the hardest hard selection that our brightest and best scientists can devise. As I have proved, the hardness of the selection is a crucial variable in the math, you can't ignore it.
You think starvation is a soft selection pressure?
All adaption by rmns requires lineages to address nested binomial probability problems. It doesn't matter whether they are antimicrobial agents, starvation, dehydration, thermal stress, predation, diseases, you name it. And why do you keep ignoring my question whether the intensity of selection alters the evolutionary trajectory to adaptation? If the same selection pressure is applied at low intensity, does it require different mutations for adaptation than if the selection pressure occurs at high intensity?

This message is a reply to:
 Message 337 by Dr Adequate, posted 10-14-2016 3:40 PM Dr Adequate has replied

Replies to this message:
 Message 353 by Dr Adequate, posted 10-14-2016 5:18 PM Kleinman has replied
 Message 356 by Taq, posted 10-14-2016 5:37 PM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 355 of 393 (792878)
10-14-2016 5:33 PM
Reply to: Message 339 by Modulous
10-14-2016 3:51 PM


Re: Lenski
quote:
Lenski's starvation stress slows the doubling time for his population to about every 7 hours.
I was referring to the number of generations, not hours. Thus everything else you said is irrelevant. Unless you think that dinosaur generation time was changed from 2 years to 20 years because they weren't birds?
So do you think the amplification time (starvation alone=>about 1000 generations per beneficial mutation) would speed up if a small thermal stress was added to his populations or would the number of generations per beneficial mutation increase?
quote:
In fact, if there is too many selection pressures on the population yet the population is not driven to extinction, what you will see is the population will just drift. This is what happens to HIV when subjected to the three drug therapies.
The number of pressures is considerably less important than their magnitude. This is the central problem with your whole thesis. You need to address this.
Large numbers of low intensity selection pressures do not cause rmns to work, you get drift under these conditions.

This message is a reply to:
 Message 339 by Modulous, posted 10-14-2016 3:51 PM Modulous has replied

Replies to this message:
 Message 363 by Modulous, posted 10-14-2016 6:31 PM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 357 of 393 (792880)
10-14-2016 5:47 PM
Reply to: Message 344 by Taq
10-14-2016 4:45 PM


Re: Lenski
quote:
What you are not seeing is that the thermal stress will be impairing the replication of the energy fit variants when compared to running the experiment at the ideal temperature. The ability to amplify any beneficial mutation for a given selection pressure can and is impaired by other selection pressures.
Thermal stress is imparing the energy fit and the less energy fit by the same amount in the hotter environment. What differentiates the energy fit and the less energy fit in the same environment is their energy efficiency. This causes an amplification of the energy fit in that environment.
Of course, that's what selection pressures do. Selection pressures drive down populations unless the populations can adapt to those pressures.
quote:
So if the probability of a beneficial mutation occurring for a population size N is let's say 0.6 and you double the population size to 2N, the probability becomes 1.2?
The probability of drawing the Ace of Spades is 1 in 52 attempts. The odds of drawing the Ace of spaces in 104 draws is 2. Not that hard to figure out.
Taq, there is no such thing as a probability value over 1. Why don't you watch some youtube videos on probability theory, eg Khan Academy
https://www.youtube.com/watch?v=uzkc-qNVoOk&list=PL06A16C...

This message is a reply to:
 Message 344 by Taq, posted 10-14-2016 4:45 PM Taq has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 359 of 393 (792884)
10-14-2016 6:10 PM
Reply to: Message 350 by Modulous
10-14-2016 5:11 PM


Re: the equality of pressure?
quote:
They are not competing for food initially because the petri dish is so large
But they don't evolve antibiotic resistance initially. So how does this demonstrate that rmns is working 'better'?
But the populations are amplifying improving the probability that the first beneficial mutation will land on some member of a population
quote:
At what point do you say that the probabilities are so low that the particular outcome is impossible.
The point is that you are treating the selection pressures affecting HIV subject to combination therapy (a distinctly unnatural event) as the same as those affecting the dinosaurs and trying to conclude this means the dinosaurs could not have evolved into birds. You are trying to suggest the probabilities are the same, or worse, without providing any evidence of this. You just use the obfuscatory 'the maths is the same' line I have previously criticised as insufficient.
I disagree with you. I think combination selection pressures are the rule in nature. Drought, starvation, thermal stress, disease, predation,...
quote:
You might say that winning a single lottery is possible and that winning two lotteries is also still possible but not very likely and winning three lotteries, that's getting a little iffy, how about you winning 10 lotteries or a 100 lotteries?
If the odds of winning a lottery is 1 in 10,000,000 and I buy a quadrillion tickets, I expect to win a billion times in one lottery or once in a billion lotteries or any combination between.
However, you have not provided the numbers to answer the question in the case of dinosaurs, you have assumed the lottery or lotteries the dinosaurs entered was the same as the lottery or lotteries HIV enters during therapy. This is an empirically false assumption, hence your incorrect conclusion.
Despite your attempt to distract from this point, it remains the fatal flaw in your argument.
How large is that population size for dinosaurs? And remember, or learn, that rmns occurs on lineages, you know, common descent. And that quadrillion population is now reduced to a new lineage of 1 with that first beneficial mutation. And until that variant amplifies, the probabilities are very low that the 2nd beneficial mutation will occur on one of its descendants.
quote:
You can have amplification without any change in the relative frequency of the variants in a population by having all variants amplifying simultaneously.
It's possible, but stupendously unlikely, for this to occur in nature. It cannot happen for long.
Well then how did Weinreich measure so many different variants from his one targeted antibiotic selection pressure?
quote:
That's what you are seeing in the video of the bacteria evolving resistance.
No, it isn't.
What do you think you are seeing in the video. Do you think that all the colonies are giving the same variants?
quote:
You see multiple different colonies forming and then you get mutant variants in several of the colonies which can then start growing on the increased concentration antibiotic bands.
But the variants that have not evolved the antibiotic resistance don't amplify. At least there is no evidence that if they do, they do so at exactly the same rate as the bacteria with the new food source. Indeed we can use the work of Lenski already presented to demonstrate adequately that having fewer food sources inhibits amplification compared with having more.
The non-antibiotic resistant variants amplify until the resources of the plate are exhausted. They were not able to do enough replication trials to get that beneficial mutation.

This message is a reply to:
 Message 350 by Modulous, posted 10-14-2016 5:11 PM Modulous has replied

Replies to this message:
 Message 365 by Dr Adequate, posted 10-14-2016 6:32 PM Kleinman has replied
 Message 370 by Modulous, posted 10-14-2016 7:27 PM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 360 of 393 (792885)
10-14-2016 6:17 PM
Reply to: Message 352 by Dr Adequate
10-14-2016 5:13 PM


Re: Lenski: Cit+ isn't slow.
quote:
That's about 1 beneficial mutation per thousand generations. If you think that's fast, would you argue about Haldane's dilemma?
Well, it's fast enough: see post #275.
Fast enough for what? This is an experiment with only a single selection pressure. Somehow you have gotten in your mind that adding selection pressures will speed up this process. rmns is nothing more than nested binomial probability problems, that's the mathematical fact of life. And adding more selection pressures simply adds more binomial probability problems for the lineage to solve.

This message is a reply to:
 Message 352 by Dr Adequate, posted 10-14-2016 5:13 PM Dr Adequate has replied

Replies to this message:
 Message 361 by Dr Adequate, posted 10-14-2016 6:28 PM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 362 of 393 (792887)
10-14-2016 6:28 PM
Reply to: Message 353 by Dr Adequate
10-14-2016 5:18 PM


Re: Is it summation time?
quote:
You think starvation is a soft selection pressure?
That depends crucially on whether it's hard or soft.
Do you think the intensity of selection in the Lenski experiment is hard or soft?
quote:
All adaption by rmns requires lineages to address nested binomial probability problems.
Well, perhaps you could show us some math.
Again? That would make Percy unhappy.
quote:
And why do you keep ignoring my question whether the intensity of selection alters the evolutionary trajectory to adaptation?
Where did you ask me that?
The answer is obviously yes.
Now I would like to see some empirical evidence of that, you obviously have that. If that were true, why would sequencing HIV be of any value to identify drug-resistant variants?
quote:
If the same selection pressure is applied at low intensity, does it require different mutations for adaptation than if the selection pressure occurs at high intensity?
It may.
Why might that happen?

This message is a reply to:
 Message 353 by Dr Adequate, posted 10-14-2016 5:18 PM Dr Adequate has replied

Replies to this message:
 Message 367 by Dr Adequate, posted 10-14-2016 6:40 PM Kleinman has not replied
 Message 376 by Admin, posted 10-15-2016 9:24 AM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 364 of 393 (792889)
10-14-2016 6:31 PM
Reply to: Message 354 by Dr Adequate
10-14-2016 5:19 PM


Re: Simulation
quote:
So, here are the results of some simulations differing only in the number of selection pressures operating.
Click to enlarge.
As you can see, the rate of adaptation is faster the more selection pressures are operating, in line with (a) my math (b) common sense.
That's not your fixation calculation again? Stop being sneaky.

This message is a reply to:
 Message 354 by Dr Adequate, posted 10-14-2016 5:19 PM Dr Adequate has replied

Replies to this message:
 Message 366 by Dr Adequate, posted 10-14-2016 6:33 PM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 368 of 393 (792893)
10-14-2016 6:46 PM
Reply to: Message 358 by bluegenes
10-14-2016 6:00 PM


Re: Lenski: Cit+ isn't slow.
quote:
Kleinman writes:
As for the Citrate metabolizer (e coli has alway been able to metabolize Citrate because it has a Krebs cycle), appeared at about generation 31500, not at 100 generations.
I said "change the circumstances". I know the Lenski experiment took over 30,000 generations for Cit+. The multi-mutation adaptation can happen in less than 100 generations in other circumstances.
Just a moment...
That, ironically, is a paper by creationists who didn't like the idea that the occurrence of Cit+ in Lenski is special, and set out to show that it could happen easily. It can.
Interesting paper. So do you agree with them when they said, "We conclude that the rarity of the LTEE mutant was an artifact of the experimental conditions and not a unique evolutionary event. No new genetic information (novel gene function) evolved."
quote:
So, where are your calculations? Surely they're not based on the view that an adaptation involving 5 or more mutations in sequence would take ~31,000 generations in a culture of bacteria in any or all environments? How are environments factored in?
You still seem to be assuming that organisms have to be threatened with extinction in order to evolve, as suggested before. They don't. Flying squirrels evolve alongside the non-flying versions they descend from, and both thrive. It's just better to be able to glide in certain environments. Same with dinosaurs. The fliers existed alongside the non-fliers for tens of millions of years. They just fill different niches
The basic science and mathematics of random mutation and natural selection - PubMed
The mathematics of random mutation and natural selection for multiple simultaneous selection pressures and the evolution of antimicrobial drug resistance - PubMed
Random recombination and evolution of drug resistance - PubMed
I did the first fundamental steps of the calculation earlier in the thread but you can see the full mathematics in the publications. And this mathematics is not based on near extinction. These calculations simply describe what populations have to do to accumulate beneficial mutations in order to adapt to selection pressures.

This message is a reply to:
 Message 358 by bluegenes, posted 10-14-2016 6:00 PM bluegenes has replied

Replies to this message:
 Message 371 by bluegenes, posted 10-14-2016 7:30 PM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 369 of 393 (792895)
10-14-2016 7:07 PM
Reply to: Message 365 by Dr Adequate
10-14-2016 6:32 PM


Re: the equality of pressure?
This message is a response to post 365, 366, 367
quote:
I disagree with you. I think combination selection pressures are the rule in nature. Drought, starvation, thermal stress, disease, predation,...
You admitted that conservative selection pressures don't count. Just thought I should remind you.
You think that drought, starvation, thermal stress, disease, predation,... are conservative selection pressures? NOT
quote:
How large is that population size for dinosaurs?
That's one of the many things you don't know that has prevented you from actually proving anything about dinosaurs.
Yes, one of the many things you do not know. I thought the theory of evolution was settled science? Well, at least you know that fixation and amplification aren't the same thing now.
quote:
That's not your fixation calculation again?
No, that's the result of simulations, as I said in the post to which you are ostensibly replying.
It is your fixation calculation again.
quote:
Do you think the intensity of selection in the Lenski experiment is hard or soft?
Lenski takes a random sample of 1% of the population every day to start a new colony and discards the rest. Insofar as it's meaningful to ask under such circustances whether the environmental pressures are hard or soft, they're soft.
Are you sure about that, it has been a while since I read his papers but I think he uses 10%. So if you think he increases his intensity of selection the evolutionary process will go faster or slower?
quote:
Again?
Well, so far all you've shown us could have been knocked off in half an hour by a bright middle schooler and comes nowhere close to proving your point, so yeah, some more math would be nice.
Actually elementary. So I left off in the mathematics after computing the probability of mutation A occurring in the population. I was just about to compute the probability of mutation B occurring on a member with mutation A. I think I'll start this mathematics tomorrow.
quote:
Now I would like to see some empirical evidence of that, you obviously have that.
Try considering the case where the intensity is 0.
Reducing selection pressure only increases the diversity of populations. No directional selection there.
quote:
Why might that happen?
Adaptive evolution is often a tradeoff. If there is more of an upside associated with making a change, then it's more likely to outweigh the potential downside.
Adaptive evolution, meet nested binomial probability problems and the multiplication rule of probabilities.

This message is a reply to:
 Message 365 by Dr Adequate, posted 10-14-2016 6:32 PM Dr Adequate has replied

Replies to this message:
 Message 372 by Dr Adequate, posted 10-14-2016 7:43 PM Kleinman has not replied
 Message 373 by Modulous, posted 10-14-2016 8:00 PM Kleinman has replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 377 of 393 (792913)
10-15-2016 9:29 AM
Reply to: Message 375 by Admin
10-15-2016 8:48 AM


Re: Lenski
quote:
This is the only equation you've presented:
Kleinman writes:
P(−∞ < X < +∞) = P(Ad) + P(Cy) + P(Gu) + P(Th) + P(iAd) + P(iCy) + P(iGu) + P(iTh) + P(del)+ = 1
What you must do is substitute actual values into this equation demonstrating evolution impossible. If I don't see this presentation very soon I will drop this thread into summation mode.
A review: We are presenting the mathematics of the simplest case of rmns, that is a single selection pressure targeting a single gene. The derivation of the equations is done in the context of an empirical example of rmns, that of a bacteria evolving resistance to an antibiotic. This empirical example can be found at icommons.harvard.edu
In the Weinreich paper he measured the mutations which evolved resistance to a particular antibiotic. There were a wide variety of variants which evolved resistance but all had in common in that it took 5 mutations to evolve high resistance to the antibiotic. The first beneficial mutation determined the evolutionary trajectory for that variant and the next 4 beneficial mutations required for that evolutionary trajectory.
The derivation for the mathematics of rmns for this problem will be done in general terms. That is that a sequence of mutations A,B,C,D and E occur where each ensuing beneficial mutation gives increasing resistance (and therefore improved fitness to reproduce) to the antibiotic selection pressure.
The first step in doing the mathematics of this stochastic process is to recognize that there are two random trials occurring in rmns. The two random trials are the replication (where the two possible outcomes are that a mutation occurs or does not occur at the particular site) and the mutation itself is a random trial (where only a particular mutation gives benefit). The possible outcomes for a mutation are written as follows:
P(-∞ < X < +∞) = P(Ad) + P(Cy) + P(Gu) + P(Th) + P(iAd) + P(iCy) + P(iGu) + P(iTh) + P(del) + = 1
Where Ad, Cy, Gu and Th represent substitutions of the particular bases, if the base is preceded by an "i", it means the insertion of that base, "del" means deletion of the base and the "..." term represents all other mutations possible. Included in that "..." term would be the mutation that caused the Citrate metabolizer in the Lenski experiment.
We now define a term P(BeneficialA) where the value of P(BeneficialA) is determined by the particular mutation which gives benefit. If the beneficial mutation is a substitution of Ad, P(BeneficialA)=P(Ad), if the beneficial mutation is an insertion of Th, P(BeneficialA)=P(iTh) and so on.
We then define — the probability (frequency) that an error in replication will occur at a particular site in a single member in one replication. Then the probability that mutation A occurs in a single member in a single generation is:
P(A) = P(BeneficialA)
We then use the complementary rule of probabilities to compute the probability that mutation A will not occur in a single member in a single replication.
P(Ac) = 1 - P(A) = 1 - P(beneficialA) where P(Ac) is the probability that mutation A will not occur.
We can then define "n", the population size and using the multiplication rule of probabilities, compute the probability that mutation A will not occur in "n" replications.
P(Ac) = (1 - P(BeneficialA))^n
We can then define "nGA" the number of generations the population "n" replicates for the probability of mutation A to occur. And again using the multiplication rule gives:
P(Ac) = ((1 - P(BeneficialA))^n)^nGA = (1 - P(BeneficialA))^(n*nGA)
And to obtain the probability of mutation A occurring in a population size "n" in "nGA replications we again use the complementary rule which gives:
P(A) = 1 - (1 - P(BeneficialA))^(n*nGA)
Note that n*nGA is simply the total number of replication trials for the mutation A to occur. With sufficient numbers of trials, we finally get a reasonable probability for mutation A to occur on some member of the population. That member with mutation A is the progenitor for a new lineage who are candidates for mutation B, that is a new branch on a phylogenetic tree. However, there is only one member to start with. This member must replicate for many generations so that there are a large number of members with mutation A, then there will be a reasonable probability that one of the members with mutation A on replication will get mutation B. The mathematics is done in an analogous manner as computing the probability of mutation A occurring in a population size "n" in "nGA" generations but in this case n->nA the number of members with mutation A, nGA->nGB, the number of generations the members with mutation A replicate. The calculation goes as follows:
P(B) = P(BeneficialB), complementary rule
P(Bc) = 1 - P(beneficialB), multiplication rule
P(Bc) = (1 - P(BeneficialB))^nA, again multiplication rule
P(Bc) = ((1 - P(BeneficialB))^nA)^nGB = (1 - P(BeneficialB))^(nA*nGB), and finally complementary rule
P(B) = 1 - (1 - P(BeneficialB))^(nA*nGB)
And to compute the joint probability of some member of the population getting both mutation A and B, we use the multiplication rule:
P(A)P(B)= (1 - (1 - P(BeneficialA))^(n*nGA))*(1 - (1 - P(BeneficialB))^(nA*nGB))
The calculation for the mutations C,D and E occurring is done in the exact same manner. What is seen that in order for this evolutionary process to occur, we have a cycle of beneficial mutation followed by amplification of that mutation in order to improve the probability of the next beneficial mutation. That rate of amplification is strongly dependent on environmental conditions, that is the other selection pressures that are acting on the population at that time.
The simplest example of how this cycle of beneficial mutation followed by amplification of the mutation can be disrupted is by applying a second selection pressure simultaneous with the first selection pressure. Not only does this make the evolutionary trajectory more complex, the amplification process for any beneficial mutation for one selection pressure is interfered with by selection pressures targeting other genetic loci.
One can easily substitute values into the above equations using a spreadsheet program using different values for the mutation rates and populations sizes and determine the probabilities.
An interesting extension of this mathematics is determining the probabilities when more than a single selection pressure are acting at the same time where it requires 2 (or more simultaneous) beneficial mutations in order to improve fitness. Anyone here besides me want to try to do that computation? Perhaps you can find a way to rescue your theory of evolution from the multiplication rule of probabilities. And feel free to substitute 1 for any of the P(Beneficial) terms.

This message is a reply to:
 Message 375 by Admin, posted 10-15-2016 8:48 AM Admin has replied

Replies to this message:
 Message 378 by Admin, posted 10-15-2016 9:48 AM Kleinman has not replied
 Message 379 by Dr Adequate, posted 10-15-2016 9:51 AM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 381 of 393 (792917)
10-15-2016 10:11 AM
Reply to: Message 373 by Modulous
10-14-2016 8:00 PM


Re: the equality of pressure?
quote:
quote:
... What I do remember is that they listed at least 8 genes necessary to be transformed. ...
Excuse me for coming in late in the discussion. I have received your pdfs but not had time to look them over yet. Can you answer a simple question for me, even if it has already been asked?
What is the probability that a mutation will be beneficial?
Your question hasn't been asked and my answer is I don't know. ...
If you don't know then you cannot invalidate evolution with your calculations.
... But this is not a number which you have to know to understand how rmns works.
Evolution works by beneficial mutations being selected by natural processes -- success in living and breeding. The probability of a mutation being beneficial would seem to be central to any mathematical approach trying to show that evolution is broken.
RAZD, which do you think will be larger, the mutation rate or the beneficial mutation rate? We know with mathematic certainty that the P(Beneficial) term can only have a value between 0 and 1. If you let P(Beneficial) = 1, you bracket the solution with the upper limit of the probability. Knowing or not knowing the exact value of P(Beneficial) does not have any mathematical significance on the physics of rmns.
quote:
quote:
I think we can all agree that mutations are random -- leaving aside for the moment that the probability of mutations varies with the section of DNA involved -- and that some are immediately deleterious or immediately beneficial, while others are immediately neutral and their relative deleterious\beneficial value can be important later.
We also have cases where a mutation is somewhat deleterious but leads later to beneficial results because of changing environmental conditions.
So how can we predict the probability of a mutation being beneficial?
Most people say that most mutations are neutral. ...
Which leaves the door open to later mutations that can build on them.
Certainly, environmental conditions have a significant impact in determining whether a mutation is beneficial or not. But it doesn't change the physics of rmns and that physics consists of nested binomial probability problems linked by the multiplication rule of probabilities.
quote:
... Mutations are fairly rare to begin with. ...
Yet every individual in every species has several.
That's fine, but you need to understand that in order for a lineage to accumulate a set of beneficial mutations, it must do so by overcoming a set of binomial probability problems linked by the multiplication rule. If you have an interest in solving medical problems such as less than durable cancer treatments, you need to understand the physics you are dealing with. This will become even more important as the use of targeted cancer therapies are developed. Targeted selection pressures give the easiest binomial probability problems for a replicator to solve by rmns. If you are going to gamble with replicators, you had better learn the rules of the game if you want to shift the odds in your favor.
quote:
... Most DNA replication is done with high fidelity. ...
For individuals that survive from zygote to born young. All those that die from birth defects are because of less than sufficient fidelity.
True as well. That's why any injury to cell in the embryonic state are transmitted to all the descendent cell. But even the normal cells in our body can develop minds of their own and become cancers because of the huge numbers of replication from zygote to adulthood. And these cells, if they are not driven to extinction by targeted therapy will still come back and kill the person. We see that with the use of estrogen blocking agents in the treatment of breast cancer. The cells are suppressed until a variant appears that's no longer estrogen sensitive. A second drug targeting these cells would slow if not stop this process.
quote:
... As you read my papers on rmns, you will see that I address the possibility that even though a mutation occurs at the correct site in a genome, it has to be the correct mutation to improve fitness. ...
This is confused. There is one mutation, it occurs randomly, and that means both location and format\type are part of the same mutation, not two separate problems. The same type in a different location would be a different mutation.
There is no "correct mutation" -- the mutation happens and then selection operates on that mutation, whether it is in location A or location B, whether it is type K or type L. Whether or not it is beneficial is that probability discussed above that you admit you don't know.
This seems to be the root of your problem, trying to make a single mutation event into a two event process.
It also seems from this that you are calculating the probability of a given mutation occurring in a second individual. Certainly when you go to two mutations occurring independently in different individuals the maths would give an extremely low probability for occurrence, but that is not how evolution works.
If you have no experience in the mathematics or probability theory, these papers will definitely confuse you. That's why when I looked for a journal to publish these papers, I wanted editors and peer reviewers who had experience with this type of mathematics. If you want to understand this mathematics, it's not that difficult but it requires a little training and some practice. If you want me to point you to sources, youtube has some good lectures on the subject. Master the mathematics of coin tossing and dice rolling and these calculations will become much clearer to you. If you have interest in improving cancer treatments and preventing antimicrobial drug resistance, learn this math.
quote:
... Just getting an accurate mutation rate is a challenging problem and then determining the fraction of the mutations which are beneficial, neutral and detrimental is even more challenging. ...
Yet we know that they all occur every generation of every species. All natural selection needs are some beneficial mutations and a low rate of death/fertility problems (where selection pressure enters the picture).
And these publications I sent you explains how natural selection does this. Mutations occur all the time on replication but in order to improve fitness to reproduce, it has to be the correct mutation occurring on the correct individual to accomplish that task.
quote:
... But the mutation rate is not the dominant factor in the rmns problem, it is the multiplication rule of probabilities that drives this phenomenon. ...
This too is confused. If I take a coin and toss it 53 times I end up with a pattern of heads and tails, and the probability of my getting that specific pattern is 1.
If I try to match that pattern with another 53 tosses the probabilities are, by the multiplication rule, extremely minute. You only use multiplication when the same steps need to be reproduced. Evolution does not work that way.
But in order to improve fitness requires specific patterns of mutations. Natural selection tries to select these pattern by a cycle of beneficial mutation followed by amplification of the beneficial mutation to improve fitness to reproduce.
quote:
... It is the joint probability that two or more beneficial mutation occur on a lineage which drives this problem.
Again, we get back to the question of the probability that a mutation will be beneficial, which you admitted you don't know ... certainly then you can't know the probability of a second mutation being beneficial, but that isn't the worst of your problem.
There are actual documented experiments (one involving E. coli) where a neutral mutation occurs in one generation and then in a later generation a second mutation occurs where the combination is beneficial, meaning that the original mutation is now beneficial. Calculating the probability that those two specific mutations would occur (the "correct mutations" at the "correct locations") would result in a very small number, but the probability that it occurred is 1: it happened.
Your model is wrong because there is an assumption of structure to the mutation process being necessary to evolution, and that assumption is false.
Again I ask you, which is larger, the mutation rate or the beneficial mutation rate? And before you say my model is wrong, you need to understand where it is wrong and right now you are confused. When my last paper was peer reviewed, I was asked to point out why Haldane's and Kimura's models were not correct. I studied and understood their models and calculations. In fact, I even wrote an exact solution to Haldane's model to check if his approximate solution was accurate. Haldane and Kimura don't have a problem in their mathematics, they have a problem in their physics. I explain the problem in the paper on rmns with multiple simultaneous selection pressures.

This message is a reply to:
 Message 373 by Modulous, posted 10-14-2016 8:00 PM Modulous has seen this message but not replied

Replies to this message:
 Message 383 by Admin, posted 10-15-2016 10:38 AM Kleinman has not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 382 of 393 (792918)
10-15-2016 10:29 AM
Reply to: Message 380 by Admin
10-15-2016 9:54 AM


Re: Mathematics cannot change reality but when done correctly can predict it
quote:
You've cited these inaccessible papers too many times. Until they're made publicly available, each time you cite them henceforth I will suspend you for 24 hours.
About doing "the first fundamental steps of the calculation," you haven't made any calculations based upon the single equation you presented.
I've posted the equations including the fundamental steps and the derivation of the equations in post 377. This is the physics and mathematics which governs rmns, it consists of nested binomial probability problems linked by the multiplication rule of probabilities. You don't need to go to the links to learn this mathematics, I've given it to you here. Now let's hear the repetitive arguments of evolutionists that it doesn't work this way despite the fact that all real, measurable and repeatable examples of rmns obeys this mathematics.

This message is a reply to:
 Message 380 by Admin, posted 10-15-2016 9:54 AM Admin has seen this message but not replied

  
Kleinman
Member (Idle past 335 days)
Posts: 2142
From: United States
Joined: 10-06-2016


Message 385 of 393 (792921)
10-15-2016 10:56 AM


Re: Mathematics cannot change reality but when done correctly can predict it
quote:
And to compute the joint probability of some member of the population getting both mutation A and B, we use the multiplication rule:
Shouldn't that read "some members"?
If the subpopulation with mutation A is large enough, you may get more than one member getting mutation B. But the real amplification process starts with that progenitor(s) with mutations A and B start replicating. A single member with mutation A and B in 30 generations of doubling will e9 members in the ideal case.
quote:
What is seen that in order for this evolutionary process to occur, we have a cycle of beneficial mutation followed by amplification of that mutation in order to improve the probability of the next beneficial mutation. That rate of amplification is strongly dependent on environmental conditions, that is the other selection pressures that are acting on the population at that time.
But your math doesn't model this amplification in any way. Unlike, for example, my simulation.
Amplification occurs when the number of replications increases for a particular variant, that doesn't necessarily happen when the relative frequency of variants in populations change. You can get fixation without any amplification just by killing off all variants except one. You need to recognize that in the stochastic process of rmns, the replication is the principle trial for this phenomenon.
quote:
Not only does this make the evolutionary trajectory more complex, the amplification process for any beneficial mutation for one selection pressure is interfered with by selection pressures targeting other genetic loci.
Again, where's the math? All your middle-school math deals with so far is the probability of two mutations arising in a given period of time, it says nothing about how they get amplified.
Now Doc, I take offense to you calling this middle-school math. This is elementary school math and don't forget it.
You are correct, these are not rate equations. These equations only predict how many replications are required for a particular mutation to occur on a particular individual in the population. The rate at which these probabilities change are strongly dependent on the selection conditions. The video of bacteria evolving resistance to an antibiotic shows a rapid rmns process for a single selection pressure targeting a single gene. The Lenski experiment on the other hand takes decades to evolve because of the single selection pressure targeting multiple genetic loci. Computing rates of change of relative frequencies of variants is not going to give you the correct time relationship.

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024