Register | Sign In


Understanding through Discussion


EvC Forum active members: 64 (9164 total)
5 online now:
Newest Member: ChatGPT
Post Volume: Total: 916,832 Year: 4,089/9,624 Month: 960/974 Week: 287/286 Day: 8/40 Hour: 4/4


Thread  Details

Email This Thread
Newer Topic | Older Topic
  
Author Topic:   Explaining the pro-Evolution position
New Cat's Eye
Inactive Member


(1)
Message 24 of 393 (792263)
10-07-2016 3:58 PM
Reply to: Message 21 by Kleinman
10-07-2016 3:34 PM


In this example, it takes 5 mutations for the e coli to achieve resistance to this antibiotic. How do you compute the probability for those 5 mutations to occur, that is the probability is 1?
Ah, I get it. You don't understand what it means for a mutation to be random.
In Message 15 you wrote:
rmns is governed by the theorems of probability theory.
So really, it isn't. Mutations aren't necessarily purely random like Brownian Motion.
And the link you provided confirms that:
"Darwinian Evolution Can Follow Only Very Few Mutational Paths to Fitter Proteins"
The randomness of mutations is with respect to fitness, not randomness like Brownian Motion.
There can be predictable causes of mutations, its just that you can't predict them from the perspective of the phenotype. But down at the genotypic level, you very well may be able to.
So this game you're trying to play is to get to: "See, the probability is so low that those mutations couldn't have been random, so therefore the Theory of Evolution is wrong because it requires mutations to be random."
The fact that we'd all see through that baloney is why you have to take this convoluted route of asking questions that you already know the answers to and trying to "help" us figure out what you're getting at.

This message is a reply to:
 Message 21 by Kleinman, posted 10-07-2016 3:34 PM Kleinman has replied

Replies to this message:
 Message 25 by Kleinman, posted 10-07-2016 4:16 PM New Cat's Eye has replied

  
New Cat's Eye
Inactive Member


Message 26 of 393 (792268)
10-07-2016 4:20 PM
Reply to: Message 25 by Kleinman
10-07-2016 4:16 PM


Mutations are random events and rare events as well.
I don't believe you.
So how can natural selection change the probabilities so that the probability that 5 rare mutations can occur on a lineage, that is the probability is 1 to give drug resistance to an antibiotic?
Natural selection (NS) does not change the probabilities. (well, I mean it can, but that's irrelevant to this)
With respect to fitness (i.e. where NS operates), the mutations are random.
With respect to the genome, the mutations may not be truly random like Brownian Motion is.

This message is a reply to:
 Message 25 by Kleinman, posted 10-07-2016 4:16 PM Kleinman has replied

Replies to this message:
 Message 27 by Kleinman, posted 10-07-2016 4:38 PM New Cat's Eye has replied

  
New Cat's Eye
Inactive Member


(1)
Message 29 of 393 (792278)
10-07-2016 4:52 PM
Reply to: Message 27 by Kleinman
10-07-2016 4:38 PM


quote:
Mutations are random events and rare events as well.
quote:
I don't believe you.

Do you think that mutations are not random events?
I think that they are not rare events.
I think that they are random with respect to fitness.
I think they may or may not be truly random like Brownian Motion, depending on the mutation.
Can you predict when a mutation will occur?
Me personally? No.
Scientists? Maybe.
So what is the frequency of a mutation occurring at a particular site in a genome?
Which mutation and which site?
Natural selection is totally relevant. Natural selection must do something very specific in order to improve the probability that a beneficial mutation will occur.
No, that's complete nonsense. Natural selection doesn't do stuff.
There are factors which can alter mutation rates but these factors do not alter whether mutations are random or not.
I bet if I subjected my balls to some particular radiation that we could cause particular mutations that I could pass on to my offspring, but that's the kind of stuff that I'm calling irrelevant.
Mutations are always random.
No, not necessarily. And it depends what you mean by "random".
Like, there is a series of events that causes a particular mutation to happen. It may be that once that process is kicked off, that it is inevitable that the particular mutation will occur. That would not be random. That would be repeatable and testable.
And like the paper that you linked to said, for some mutations there are particular pathways that are more probable to occur than others.
quote:
Darwinian Evolution Can Follow Only Very Few Mutational Paths to Fitter Proteins.
"This implies that the protein tape of life may be largely reproducible and even predictable."
Does that sound random to you?

This message is a reply to:
 Message 27 by Kleinman, posted 10-07-2016 4:38 PM Kleinman has replied

Replies to this message:
 Message 33 by Kleinman, posted 10-07-2016 5:32 PM New Cat's Eye has replied

  
New Cat's Eye
Inactive Member


Message 93 of 393 (792469)
10-10-2016 10:12 AM
Reply to: Message 33 by Kleinman
10-07-2016 5:32 PM


They are rare events when a particular mutation must occur at a particular site in the genome to improve fitness.
And at that level is what I'm talking about when I say that you are wrong to look at these as purely random events on which you should be making probability calculations.
Having a particular mutation at a particular site may or may not be random.
And natural selection must do something specifically to improve the probability that a particular mutation will occur at a particular site in a genome to improve fitness.
Wrong. Natural Selection does not do stuff, it is an effect.
In any stochastic process, it is not possible to predict the outcome of any random experiment. What probability theory does is enable one to predict the relative frequencies of outcomes when the random experiment is done many times. A simple example is tossing a coin. You can not predict whether a head or a tail comes up with any toss. What you can predict is that if you toss the coin many times, about half the outcomes will be heads, the other half of the outcomes will be tails.
But coin tosses are completely independent. What if the odds of a particular mutation occurring are dependent on previous states?
quote:
So what is the frequency of a mutation occurring at a particular site in a genome?
quote:
Which mutation and which site?

Any particular mutation at any particular site.
Huh? That's like me asking you the frequency of rolling a 7 on a die, for any die with any number of sides for either a fair or unfair die...
What I'm trying to get at is the definition of mutation rate. The mutation rate is the probability that mutation will occur at a given site in the genome in a single replication.
No, it isn't. It is a measurement of how many mutations of a particular type one would expect to see over a given amount of time.
quote:
Natural selection is totally relevant. Natural selection must do something very specific in order to improve the probability that a beneficial mutation will occur.
quote:
No, that's complete nonsense. Natural selection doesn't do stuff.

It certainly does, natural selection changes the probabilities of particular mutations occurring by changing population sizes.
No, the environment changes the population sizes. Natural selection is the differential successes that the populations have because of the effect that the environment has on them.
And if NS did change the probabilities of mutations occurring, then how could you treat those mutations as random?
You are changing the mutation rate but you are not changing the fact that mutations are random events.
If I know what kind of radiation causes which mutations, and I do it on purpose, then those mutations would not be random.
You need to give us an empirical example where a mutation is not a random event.
Have you tried doing a Google Scholar search for "non random mutations"?
I'll tell you what; when you start providing evidence for your claims, I'll come back and start looking for sources to provide you.
Random means something that occurs by chance, not predictable.
Right, and if whether or not a mutation occurs can be affected by previous states, unlike a coin toss, then those mutations do not occur by chance alone, i.e. they are not truly random.
I'm pretty sure that I hit the nail on the head in my first reply to you: You're misunderstanding what the Theory of Evolution means by calling a mutation random.
It does not mean that all mutations happen purely by chance, like you are using it.
Natural Selection occurs at the phenotype, and from that perspective, mutations do appear to occur purely from chance (and even in this case there are exceptions). The ToE talks about mutations being random with respect to fitness.
That "with respect to fitness" is the important part you are missing, which is allowing you to make the mistake of treating mutations as a result of pure independent chance that you can then do coin-toss-like probability calculations on at the genetic level.
You can't treat all mutations as independent events and then provide a probability calculation to say that the odds of these particular mutations happening in series is impossible.

This message is a reply to:
 Message 33 by Kleinman, posted 10-07-2016 5:32 PM Kleinman has not replied

  
New Cat's Eye
Inactive Member


Message 218 of 393 (792681)
10-12-2016 11:09 PM
Reply to: Message 217 by Kleinman
10-12-2016 9:18 PM


Re: Mathematics cannot change reality but when done correctly can predict it
In case you don't recognize it, these probability equations are the general solution for rmns and apply to all replicators.
There's no reason to believe that at all.

This message is a reply to:
 Message 217 by Kleinman, posted 10-12-2016 9:18 PM Kleinman has replied

Replies to this message:
 Message 224 by Kleinman, posted 10-13-2016 4:41 PM New Cat's Eye has replied

  
New Cat's Eye
Inactive Member


Message 231 of 393 (792731)
10-13-2016 5:13 PM
Reply to: Message 224 by Kleinman
10-13-2016 4:41 PM


Re: Mathematics cannot change reality but when done correctly can predict it
quote:
In case you don't recognize it, these probability equations are the general solution for rmns and apply to all replicators.
There's no reason to believe that at all.
There is if you understand that rmns is a stochastic process and understand how to do probability calculations.
There's no reason to believe that your probability equations are the general solution for rmns, and there is no reason to believe that your probability equations apply to all replicators.
You have probability equations for particular replicators in particular situations. You've provided no evidence or argument for them applying as a general case or for all replicators.
The closest you've come to it is: "show me one that doesn't", and that is neither evidence nor argument.

This message is a reply to:
 Message 224 by Kleinman, posted 10-13-2016 4:41 PM Kleinman has not replied

  
Newer Topic | Older Topic
Jump to:


Copyright 2001-2023 by EvC Forum, All Rights Reserved

™ Version 4.2
Innovative software from Qwixotic © 2024