What is an ID proponent's basis of comparison? (edited)

Good point. However, as you would expect, I remain skeptical because if we are talking about one or two simple mutations, then I would have to agree with you. But when there are multiple coherent mutations that are required to produce new structures or to take large steps, I remain unconvinced. I must say that I don't know enough about the cell and genetics to really refute what you are saying.What I do know is that just after the first time I left this forum, I became a better debater. As my jiu-jitsu teacher taught me, "You will learn more when you risk exposing yourself than when you don't."

As evolutionary mechanisms are currently understood, new structures or large steps in a single generation would be very unlikely.Evolution produces only minute change in each generation, selecting traits from the pool of individuals who survived to reproduce. The procedure of selection and reproduction is repeated from scratch in each generation, each generation producing minute change. Over thousands of years the changes gradually accumulate. This tells me two things. One, your Jiu-Jitsu teacher was very wise. Two, keep disagreements with you on a verbal level. --Percy

Traderdrew, can you provide an example of this large step that has to be taken all at once? Personal incredulity does not provide much to argue about. While evolution can occur in somewhat large steps(frameshift mutations that can alter multiple amino acids) its almost always in small steps over many many generations. Can you provide me some instances of a super difficult mutation along with the mathematical probability that this would occur?

This is part of Darwinian theory, is it not?For one thing, a specific gene can be broken in multiple places. (missing amino acids, or insertions, substitutions are some examples) In order for a gene like that to regain its original function, there would have to be soe multiple mutations that are also the right kind of mutations "before" deleterious mutations can occur to that gene.I see that you are being cool. So in defense of Darwinism, I would suspect that multicellular organisms would more easily circumnavigate whatever hurdles it may encounter so it can continue to evolve.However, I don't think this would explain how to evolved IC systems without some type of blueprint. Also, how it would explain how functionally integrated proteins evolved through a step by step process. Multiple coherent mutations would be required just to get two proteins to bind to each other. Perhaps there is a theory that explains how the machinery, that makes these proteins, evolved through a Darwinian fashion.

I haven't explored mathematical probabilities of small hurdles in proteins. I casually have thought about it in greater proabilities but of course our minds tend to lose the concept of large numbers.According to information I got from Michael Behe, I would think a protein binding site would be a good example. There are many proteins in our body that work in teams. Just binding two proteins together would be around one chance in 10 to the 20th power.Of course, if the genetic code can't evolve a blueprint to make an IC system, what are the chances of evolving one through a Darwinian framework? Sure there are Darwinian models but do those models conceal any problems along the way?Edited by traderdrew, : No reason given.

Well, that kind of thing can happen. It all depends on the situation. It is possible that a mutation could occur that in the current environment is neutral, but if the environment changes, it's effect on fitness could become either positive or negative. It's also possible that a neutral mutation could interact with a new mutation in such a way as to give a positive or negative fitness effect. Mutations are all about context. A set of strictly neutral mutations will not be adaptive, but if their put in an environment that causes them to have a fitness effect, they can start to have effects on adaptive evolution. One effect of genetic drift is the accumulation of neutral alleles (another is the fixation or loss of alleles regardless of their fitness advantage) Genetic drift, however has no adaptive power. This has been thoroughly tested, and no scientist claims that genetic drift causes adaptive evolution. However, it does cause evolution in the sense that it can cause a change in the frequency of alleles in a population simply due to sampling error of finite populations.

---

We have many intuitions in our life and the point is that many of these intuitions are wrong. The question is, are we going to test those intuitions?
-Dan Ariely

quote:

---

You can't even remember what you said in previous posts. You really are dumbing down your cause.

---

You filthy scum, I meant a cause inside the organism.

quote:

---

No that is not what the random in random mutations means. Do you understand what the word 'random' means?

​

Random means it cannot be accurately predicted meaning we cannot accurately predict where genetic mutations will strike next in an organisms genome. Some mutational sources are more random than others. For example mutations cause by UV sources are almost completely random since nearly all portions of the genome are sucesptable to this radiation source and it would be nearly impossible to determine where exactly these point mutations could occur. Whereas, the areas of DNA which are suseptible to viral agents of mutation may be more predictable in there location of occurance.

---

We can't predict them 100% but we can predict them with probability. Since some mechanisms are know to induce mutations, than it's obvious that we can predict on which regions of the genome they are more likely to appear.

quote:

---

You have yet to show how this helps your case as I debunked your idea that these genetic mechanisms had to develop for any evolutionary changes to occur.

---

You debunked me? Where?Even if there were still random mutations present for that specific part of the genome, it shows that without LexA they are so insignificant that the can't get you resistance in a meaningful amount of time.

quote:

---

Randomness has absolutely nothing to do with causation. The concept you are thinking of is described by the term "spontaneous." Mutations are not spontaneous: they are random.

​

Randomness only deals with incidence and uncertainty, not causation. It implies that there are multiple options, no one of which is guaranteed to come to fruition, but any one of which could potentially happen.

​

This is a perfect description for bacterial mutations, with or without the inhibitor. Turning on the inhibitor increases the likelihood that a mutation will slip past the repair machinery, but it does not cause mutations to happen. The bacterium is still reliant on the usual causes of mutation to make mutations happen.

​

If the inhibitor is on, mutations are caused by DNA replication errors, chemical imbalances, radiation, etc.; if the inhibitor is off, mutations are caused by DNA replication errors, chemical imbalances, radiation, etc.

​

What you are calling "induction" is not causation, it is facilitation. This means that anything that happens under induction could happen without induction, but it would just happen more slowly.

---

You are confusing mechanism that let mutations through by shuting down repair systems, with mechanisms that induce mutations. And when they are shut down, the specific region of the genome can not evolve.

quote:

---

Except it's true. If it were false then models of the evolutionary algorithm couldn't produce new information. If it were false then you would be able to identify the location in the cell of the information for where to place the mutation and which base pairs to substitute.

​

I understand that you believe it is false. But believing and proving are two different things.

---

No, I already explained why it's false. And I gave links that describe the NFL theorem that explains why algorithms do no produce new information. Why didn't you read them?

quote:

---

The evidence we have says that the mutation rate goes up when the mutation repair mechanism is disabled. If you think a mutation inducing mechanism exists then you have to provide evidence of one.

---

I did, few posts ago. It's caleld LexA. When it is turned off no evolution is possible on the specific part of the genome.

quote:

---

Interesting idea that all mutations are deterministically induced, but there's no evidence for it.

---

Maybe they are, maybe not. The point is, LexA is evidence from which we can extrapolate to think that it could be true.

quote:

---

By random I mean the frame shift/base pair change or whatever IS random. If I induce a die to roll is the outcome no longer random?!

---

Teh outcome is random, but the induction is not.

quote:

---

Contradiction, anyone???

---

It's not a contradiction. I said that the mechanism has no specific goal in mind. When the bacteria acquires resistance the mechanism stops. But the mechanism doesn't know the exact sequence it needs. That's why it takes time. If it did know, it would change the DNA sequence in an instant.

quote:

---

Yes, they wait for it to randomly occur! While mutating randomly (like a die rolling) there is a probability that a mutation will arise that gives the bacterium an advantage (like getting the desired number up on the die). Even if this increased mutation rate is induced, each time there is a mutation, it's a random mutation. Most of these mutation will be neutral or harmful, but with enough generations (rolls of the die) the right one will come up, and the bacteria that have that mutation will out-compete the others. It's really quite simple, and you seem to have a hard time grasping what random really means in this context. I've tried to explain it to you many times. This is my last. I'll respond to another argument that you have, but I'm done with this random mutation part. I don't see how it could be any clearer.

---

The point is that the act of induction of mutations, their occurance itslef is not random. They are not happening randomly. The sequence you get when you mutate a specific part of the genome is random, but the act of mutating it is not random.

quote:

---

It seems that an assertion you're making, Smooth, is that evolutionary algorithms can't produce new complex specified information, because they are programmed in advance by humans, even their 'random' components...

---

Exactly.

quote:

---

My question: Does this limitation on evolutionary algorithms, in your view, apply to algorithms more generally? i.e. can any algorithms produce new complex specified information? If they can, which ones can and which ones can't? How do we tell the two kinds apart?

​

If no algorithms can generate CSI, then it would imply that 'complex specified information' is in technical terms non-computable. This would have interesting implications.

---

It is computable, the algorithm just can't generate it. They can only process it. It applies to all algorithms. It has been shown to be true. I have already posted a link here about the NFL theorem that says that algorithms do not produce new information. It really gets on my nerves to have to do it again and again.

Here at EvC Forum you're expected to produce arguments in your words and use links only as supporting references. This is from the Forum Guidelines:

5. Bare links with no supporting discussion should be avoided. Make the argument in your own words and use links as supporting references.

So what you need to do is explain how NFL theorems disallow the possibility of genetic algorithms producing new information.But I can save you some time. Genetic algorithms that model evolutionary behavior are already in use today producing novel designs. They're a practical reality. There really isn't much point to arguing that something that works before our very eyes doesn't really happen. And then how do you explain the design innovations produced by the algorithm since they're not figments.But we don't want to turn this thread into a discussion of GA's. The more relevant point is that new information is being created all the time throughout the universe, including through the process of mutation. If you think that the information identifying where mutations should occur and which base pairs should be involved is already part of the cell, then you have only to find the source of this information, and you have to find it for all possible mutations. Since mutations are known to occur anywhere throughout a genome, and since bacterial genomes range from around 500,000 base pairs up to 10 million base pairs, you need information somewhere in the bacterial cell for the production of each and every possible mutation, as well as the molecular triggers for each one. It has already been explained that this isn't true. What LexA does is control whether the genetic repair mechanism is enabled or not. When it is enabled then there are still mutations, just fewer of them. The average bacterial mutation rate of 10^-8 is when the repair mechanism is enabled. Note that it isn't 0.--Percy

quote:

---

So what you need to do is explain how NFL theorems disallow the possibility of genetic algorithms producing new information.

---

Read my next post, I will explain it there.

quote:

---

But I can save you some time. Genetic algorithms that model evolutionary behavior are already in use today producing novel designs. They're a practical reality. There really isn't much point to arguing that something that works before our very eyes doesn't really happen. And then how do you explain the design innovations produced by the algorithm since they're not figments.

---

I'm sorry but no. Are you talking about that NASA antenna they build with algorithms? Yes, I know about that. The problem is, that the algorithm itslef already had all the information to produce the design. Just like a computer has all the information it needs to produce anything you want it to. But its very long and a boring job, so scientists let to computer calculate whatever they put in it.And after the calculation is done, the optimized design is produced by the computer. But the point is that computer had all teh information it had to select the best design. It produced no new information. It only selected the best one.

quote:

---

The more relevant point is that new information is being created all the time throughout the universe, including through the process of mutation. If you think that the information identifying where mutations should occur and which base pairs should be involved is already part of the cell, then you have only to find the source of this information, and you have to find it for all possible mutations. Since mutations are known to occur anywhere throughout a genome, and since bacterial genomes range from around 500,000 base pairs up to 10 million base pairs, you need information somewhere in the bacterial cell for the production of each and every possible mutation, as well as the molecular triggers for each one.

---

The information is already in the genome. The mechanisms that the cell has helps the cell adapt. It selects the best possible expression of already existing information.

quote:

---

It has already been explained that this isn't true. What LexA does is control whether the genetic repair mechanism is enabled or not. When it is enabled then there are still mutations, just fewer of them. The average bacterial mutation rate of 10-8 is when the repair mechanism is enabled. Note that it isn't 0.

---

Wrong. I said it three times already. You are confusing the mutation repair and mutation inducing mechanisms. When LexA is turend ON there are mutations. When its turned OFF, there are no mutations.

quote:

---

The scientists also show that E. coli evolution could be halted in its tracks by subjecting the bacteria to compounds that block LexA. Interfering with this protein renders the bacteria unable to evolve resistance to the common antibiotics ciprofloxacin and rifampicin.

---

It' can't evolve while LexA is BLOCKED. In other words, it's turned OFF, not ON.To Stop Evolution: New Way Of Fighting Antibiotic Resistance Demonstrated By Scripps Scientists – Uncommon DescentEdited by Smooth Operator, : No reason given.

NFL Theorems... I will explain them here once and for all.The NFL theorems say that no algorithm can otperform any other on the average unless it takes advantage of prior information about the target.Meaning, that if you are in a house and you are searching for the keys. You search them for some time and than you find them. After that you write down the exact path you went through to find the keys.The question is: If somebody asks you to find their keys in his own house, to whih you know nothing about, and you try to find the keys, will your path you used last time, find the keys faster than the random search?The answer is NO it won't. Becasue if you know nothing about this new house, if you don't know the size of the house, the number of rooms, or where the keys certainly are not. You are not going to find them faster than the last time, or faster than the random search.To find them faster, you have to know at least something. The size of the house, the number of rooms, or where they keys certainly are not located. But this is the main point. Any of those informations you can't get unitll you started searching. So if someone does tell you in advance, they gave you PRIOR INFORMATION about the search problem!This basicly means that even the evolutionary algorithms, which have no knowledge about what they are looking for in advance, will not be any better than a random chance. And since random chance doesn't create new information, neither does an evolutionary algorithm.Some quotes: [quote]"... no operation performed by a computer can create new information."The [computing] machine does not create any new information, but it performs a very valuable transformation of known information.A "learner... that achieves at least mildly than better-than-chance performance, on average, ... is like a perpetual motion machine - conservation of generalization performance precludes it.Unless you can make prior assumptions about the ... [problems] you are working on, then no search strategy, no matter how sophisticated, can be expected to perform better than any other"The inability of any evolutionary search procedure to perform better than average indicate[s] the importance of incorporating problem-specific knowledge into the behavior of the [search] algorithm.[/quote]The Evolutionary Informatics Lab - EvoInfo.org

Again, this is an assertion. In your example, the random search of your house turns up other information you can use to build prior information for your next search in someone else's house.For example, you notice that nothing is sitting on the ceiling, that reduces the number of places you need to search next time. You can also say, well, since I found them on the bedside table last time, rather than going through all my previous steps, I'll start there next time. Even in a new house, that's a good place to start as it assumes people are generally the same. All of this is without using prior nowledge the first time, and using the new information the second time. If the situation is similar, the exact conditions don't matter, the process can still work faster than random.