• Quick note - the problem with Youtube videos not embedding on the forum appears to have been fixed, thanks to ZiprHead. If you do still see problems let me know.

What evidence is there for evolution being non-random?

mijopaalmc said:
I only ask this, because I am thoroughly disappointed in the evidence that I have received from the posters in this thread.
Click to expand...

By posters you mean Claus, everyone else has given up arguing with him

You'll just have to wait for everyone else in this thread to catch up with what your talking about, and then they'll start agreeing with you.
 
davidsmith73 said:
But there is a more pertinent problem with adaptive mutation as I see it. In another abstract I've read, bacteria that begin to starve increase their expression of an error-prone DNA repair enzyme, thus increasing the rate of mutation under conditions where mutations might generate novel metabolic processes and get the cells out of trouble. But the base pairs that the polymerase targets is still random. At least, this is what the experiments appear to be saying so far - I would not rule out targeted errors just yet because I don't believe anyone is asking this question in their experiments.
Click to expand...

This is what I remember from being taught about this last year. It just seems that bacteria under stress produce more replication-error-based mutations, which makes sense.
 
davidsmith73 said:
Probability surely is not a synonym for random in this context. If the environment can trigger a biochemical cascade that alters the probability distribution of the occurance of mutations, then we have a mechanistic link between adaptation and mutation. This goes agaisnt what I believe most people refer to when they talk about the generation of mutations being random with respect to their phenotypic function. I have to admit, I don't understand the usefullness and relevance of your usage of "random".
Click to expand...
As an example, a weighted six-sided die favours some results more than others. We have a mechanistic reason for the change in the distribution. Which side is the most likely is dependent on the weighting you apply, but most weighted die I've played with still come up with the other numbers on occasion. They are random, just not-uniformly distributed.

Also they don't talk about "mutations being random with respect to their phenotypic function" they talk about the "mutation being indepentent of their phenotypic function." An electron passing through an difraction grating has a random probability of hitting various points on a screen, but that probability at each point is dependent on the difraction grating.
But there is a more pertinent problem with adaptive mutation as I see it. In another abstract I've read, bacteria that begin to starve increase their expression of an error-prone DNA repair enzyme, thus increasing the rate of mutation under conditions where mutations might generate novel metabolic processes and get the cells out of trouble. But the base pairs that the polymerase targets is still random. At least, this is what the experiments appear to be saying so far - I would not rule out targeted errors just yet because I don't believe anyone is asking this question in their experiments.
Click to expand...
Awesome stuff, I might have to subscribe.
 
CFLarsen said:
What would it take, then?
Click to expand...

Claus, a female human does not give birth to the full range of (2*2)^23 children she could possibly have with her partner.
Some sub-sampling has to go on. As to which of these children do finally pop out, it is entirely random.

Because evolution is chaotic it is some instances the outcome is sensitive to these initial fluctuations, making it a genuinely random process.
 
Taffer said:
This is what I remember from being taught about this last year.
Click to expand...

Thats interesting that these experiments are now be taught. I remember first reading about it in the mid 90's and I think the experiments were treated with great suspicion.
 
Taffer said:
No, you don't get my point. You already know it's random, because I said it was a die. Instead, look at the results. Say we were looking at the results of 100 of those rolls. 99 were 4 or above, with only 1 being below. What would you conclude?
Click to expand...
Perhaps I should've been more specific. You stated you throw out everything below a 4. If you had said you throw out everything below a six, that would give you a non-random function. But you threw out everything below a four, so instead of having a random distribution from 1 to 6, we have a distribution from 4 to 6.
Some do. There are many different ways for mutations to arise. For example, indel mutations often come about from "slipping" of the DNA replication mechanisms.
Click to expand...
I am unsure of the mechanism of slipping. My quick google search turned up a few quotes saying the mechanism was poorly understood (it did use the language of probability, but that is just as much an indication of lack of knowledge of the underlying principles). However, my old knowledge (from a program I saw a while ago, which was already out of date) suggests that some sequence shifts had to do with bonds breaking between codons, and a new one being inserted and bonding in the break. This again involves molecular bonding.

In addition, for evolution to be random, not all mutations have to be random. As long as a mutation of the random sort can becomes successful in the population and subsequent generations of that species are thus influenced by the random mutation, it will be technically random, through in the fact that random mutation affect the environment (and thus selection) in other species the affect can be widespread and longlasting. If many said random mutations occur over time then the result after millions of years of evolution has many different outcomes that vary by a great deal. In that said it will be random by many layman's definitions. It's not like in one scenerio humanity has a smaller pinky finger; the thing is that there was no guarantee that a intelligent, planet altering, biped was going to emerge based on the state of thing in previous epochs.

If the disease decreased fitness of a population, then obviously the other population would get a similar boost in fitness.
Click to expand...
This is wear I think the "non-triviality" of selection comes in. If a random event causes the selection mechanism to "choose" one species over another, the affect will cascade through. As a result the random variation isn't just "things are slightly different", but you can come up with to totally different situations.
 
davidsmith73 said:
Thats interesting that these experiments are now be taught. I remember first reading about it in the mid 90's and I think the experiments were treated with great suspicion.
Click to expand...

It wasn't just those experiments. It was more as a "possible" thing, with special highlighting of current experiments into the same mechanism, IIRC.
 
Just re the controversy over the adaptive mutation experiements. I remember where I first read about it.

http://www.science-frontiers.com/sf064/sf064b07.htm

http://www.science-frontiers.com/sf100/sf100b07.htm

http://www.science-frontiers.com/sf096/sf096b08.htm


So it seems that the first few experiments claimed that specifically adaptive mutations were occuring more than others in response to enviromental stress. Its no wonder they stirred up controversy. But I wouldn't rely on the interpretation from science frontiers but rather on the primary sources. Perhaps the original experiments have been shown to be at fault by now, I'm not sure. Unfortunatley I don't have access to full articles.
 
Walter Wayne said:
Perhaps I should've been more specific. You stated you throw out everything below a 4. If you had said you throw out everything below a six, that would give you a non-random function. But you threw out everything below a four, so instead of having a random distribution from 1 to 6, we have a distribution from 4 to 6.
Click to expand...

Again, that's not what I'm talking about. Lets try again.

You flip a coin 1000 times. Each "heads" represent a mean increase in fitness of a population over time, with each tails representing a mean decrease in fitness of a population over time (this isn't quite correct, but it'll do for this discussion). You only observe heads. What can you conclude?

I am unsure of the mechanism of slipping. My quick google search turned up a few quotes saying the mechanism was poorly understood (it did use the language of probability, but that is just as much an indication of lack of knowledge of the underlying principles). However, my old knowledge (from a program I saw a while ago, which was already out of date) suggests that some sequence shifts had to do with bonds breaking between codons, and a new one being inserted and bonding in the break. This again involves molecular bonding.
Click to expand...

There are different forms of this mutation type. Others include the DNA plymerase simply including an incorrect basepair.

In addition, for evolution to be random, not all mutations have to be random. As long as a mutation of the random sort can becomes successful in the population and subsequent generations of that species are thus influenced by the random mutation, it will be technically random, through in the fact that random mutation affect the environment (and thus selection) in other species the affect can be widespread and longlasting. If many said random mutations occur over time then the result after millions of years of evolution has many different outcomes that vary by a great deal. In that said it will be random by many layman's definitions. It's not like in one scenerio humanity has a smaller pinky finger; the thing is that there was no guarantee that a intelligent, planet altering, biped was going to emerge based on the state of thing in previous epochs.
Click to expand...

Please, please, define how you're using "random".

This is wear I think the "non-triviality" of selection comes in. If a random event causes the selection mechanism to "choose" one species over another, the affect will cascade through. As a result the random variation isn't just "things are slightly different", but you can come up with to totally different situations.
Click to expand...

It depends on how you look at things, and it depends on your definition of a "random system". For example, if I were teaching evolutionary genetics, I would describe selection as working on variation within a population. The variation may very well be random. But selection is not. If I roll a die 1000 times, then discard all but rolls of "6", would you call that random? Perhaps you could say that the number of 6's in the 'population' is random, but then this analogy isn't very good.
 
davidsmith73 said:
Just re the controversy over the adaptive mutation experiements. I remember where I first read about it.

http://www.science-frontiers.com/sf064/sf064b07.htm

http://www.science-frontiers.com/sf100/sf100b07.htm

http://www.science-frontiers.com/sf096/sf096b08.htm


So it seems that the first few experiments claimed that specifically adaptive mutations were occuring more than others in response to enviromental stress. Its no wonder they stirred up controversy. But I wouldn't rely on the interpretation from science frontiers but rather on the primary sources. Perhaps the original experiments have been shown to be at fault by now, I'm not sure. Unfortunatley I don't have access to full articles.
Click to expand...

AFAIK, no papers have been publish which support the existance of that form of adaptive mutation. I could be wrong, of course, but I'd have thought we'd have been taught about that. All I remember is that the current experiments suggested that bacteria could increase the number of mutations in the presence of stress, which I (ha!) stress is not particularly surprising.
 
Jekyll said:
Claus, a female human does not give birth to the full range of (2*2)^23 children she could possibly have with her partner.
Some sub-sampling has to go on. As to which of these children do finally pop out, it is entirely random.
Click to expand...

How do you know that?

Jekyll said:
Because evolution is chaotic it is some instances the outcome is sensitive to these initial fluctuations, making it a genuinely random process.
Click to expand...

Whoa. Chaotic does not mean random.

"Some instances"? Which are those? When does it not apply?
 
Claus, it depends on their definition of "random". Let's wait and see, shall we?
 
Walter Wayne said:
He hasn't met a woman who has had 70 trillion children.
Click to expand...

Women do not have 70 trillion oocytes. ;)

If selection is chaotic, then a random input into it (mutation) will make the process as a whole random.
Click to expand...

Not to nag, but we really need to know how you're using "random". Especially with regard to a system over a process.
 
Taffer said:
Women do not have 70 trillion oocytes. ;)



Not to nag, but we really need to know how you're using "random". Especially with regard to a system over a process.
Click to expand...
Technically, from identical initial conditions, several results are possible. I am using the technical a little more narrowly in that I am suggesting there are significantly different results possible. So I don't include an electronic system where the voltage is 100V +/-10mV because of thermal noise. In the case of evolution I would say that even with the exact same non-biological inputs, they is no guarantee that the end of the age of dinosaurs guarantees a branch as intelligent as humans as one example of a "siginificantly different possibility".

I do that because I think that evolution is random even in many non-technical senses. I would agree that evolution is not random if using the layman's definition of uniform probability, but the sum of two dice is non-uniform, but in most ways I have seen it used in general speech I think it is random.

Edited to add: Sorry for the long answer, but I really think it is random in many senses of the word, trying to be specific about some of the senses that don't apply.
And also so people understand I don't mean it in the almost useless sense that anything made up of quarks and leptons (i.e. everything) is going to be random in a very strict sense.
 
Last edited:
CFLarsen said:
But selection is not chaotic.
Click to expand...
In what sense? and evidence?

Also, selection only needs to have variable results with inputs differing on the scale of the mutation (random input) for evolution to be random.

So a chaotic selection is, while sufficient, is not necessary to lead to a random evolution.
 
Walter Wayne said:
In what sense? and evidence?

Also, selection only needs to have variable results with inputs differing on the scale of the mutation (random input) for evolution to be random.

So a chaotic selection is, while sufficient, is not necessary to lead to a random evolution.
Click to expand...

If selection is chaotic - unpredictable - and evolution is random, how can you correctly predict that from a gene pool of exclusively blue eyed people, you will get blue eyed children?
 
You must log in or register to reply here.

Back
Top Bottom