A challenge to Creationists.

[elijah115]

From what I can gather the general consensus on both sides of the debate is that there is a thick shroud of mystery that hangs over this question.

So why are we arguing over evolution if no-one can answer that question?

 

[Micaiah]

So why are we arguing over evolution if no-one can answer that question?

We all agree I think that God is the only one who really knows precisely what happened during Creation. Some of us believe He has given us some facts about creation in Genesis, others find this difficult to accept. How about yourself? Have you given the topic much thought.

 

[shernren]

I am looking for general definitions of the terms mentioned. Thanks for your attempt anyhow. As you may appreciate this is but the tip of the iceberg. We haven't started on the mathematics yet. This was really directed at KerrMetric. I am looking for a response in his own words or an admission that even if I was able to provide a mathematic method for measuring information gain or loss it would be meaningless to him and most other people on this forum.

Actually, it wouldn't be too hard. Let's say that of the four possible base pairs in a molecule of DNA, on average the frequency of each base pair is equal i.e. 25% each. This means that the most efficient way to encode them would be to use 2 bits per base pair. Then, if the length of the DNA strand is n base pairs, the least amount of bits that can be used to encode it is 2n bits. We can take as a crude measure of the amount of information the number of bits, i.e. 2n bits of information.

So, does a polyploidy increase information? Yes and no. Let's say that I pass you a stream of 200bits. You read through the first 100 bits and put together a 50 base pair sequence. Then you listen to the next 100 bits. Wait a minute! They're the exact same sequence! How can you encode them? Well, the easiest way would be to send a 100 bit stream, and then attach, "Message repeats". There is additional information: the information that tells the recipient that the message repeats (and as a result the message will need, say, 105 bits, instead of the original 100). Furthermore, if a point mutation happens in the second segment, then you will have to use a full 200 bit encoding to represent it.

So before realizing that it was a polyploidy, the polyploidy represented double the amount of info. After realizing that it was a polyploidy, the polyploidy represented slightly more than the original amount of info. Did AiG ever tell you that?

Under this scheme, evolution would indeed require an increase in information content. But mutations would also increase information content, insertion mutations in particular. Now, if there are many ways to quantify a process, the actual direction of the process doesn't change even if the process is quantified differently. (Whether I use algebra or differentiation to find the minimum of a quadratic, the minimum is the same.) Therefore evolution is not impossible from the angle of information content. If one defines information in such a way that it is impossible for mutations to increase information, the chances are that by that same definition evolution would not require an increase in information.

 

[shernren]

interesting read, how did the first inanimate-to-living-thing transformation occur anyway?

Well, for starters, it is believed that amino acids and RNA can arise spontaneously from basic organic compounds. This has been laboratory demonstrated (not just in Miller-Urey, mind you), although of course we may never be exactly sure just what combination took life. Also, membranes resembling cell membranes can form spontaneously from lipids, which can perform functions that life performs such as swelling or shrinking based on the environment and even storing energy as potential across the membrane. All said and done, though, there's still a lot we don't know. For both creationist and evolutionist.

 

[elijah115]

^^what paper demomstrated this? (amino acid or rna generation) i'm asking because idk

 

[elijah115]

Actually, it wouldn't be too hard. Let's say that of the four possible base pairs in a molecule of DNA, on average the frequency of each base pair is equal i.e. 25% each. This means that the most efficient way to encode them would be to use 2 bits per base pair. Then, if the length of the DNA strand is n base pairs, the least amount of bits that can be used to encode it is 2n bits. We can take as a crude measure of the amount of information the number of bits, i.e. 2n bits of information.

So, does a polyploidy increase information? Yes and no. Let's say that I pass you a stream of 200bits. You read through the first 100 bits and put together a 50 base pair sequence. Then you listen to the next 100 bits. Wait a minute! They're the exact same sequence! How can you encode them? Well, the easiest way would be to send a 100 bit stream, and then attach, "Message repeats". There is additional information: the information that tells the recipient that the message repeats (and as a result the message will need, say, 105 bits, instead of the original 100). Furthermore, if a point mutation happens in the second segment, then you will have to use a full 200 bit encoding to represent it.

i'm not a biologist but wouldn't you need information to tell the recipient when the message repeat terminates?

 

[Micaiah]

Actually, it wouldn't be too hard. Let's say that of the four possible base pairs in a molecule of DNA, on average the frequency of each base pair is equal i.e. 25% each. This means that the most efficient way to encode them would be to use 2 bits per base pair. Then, if the length of the DNA strand is n base pairs, the least amount of bits that can be used to encode it is 2n bits. We can take as a crude measure of the amount of information the number of bits, i.e. 2n bits of information.

So, does a polyploidy increase information? Yes and no. Let's say that I pass you a stream of 200bits. You read through the first 100 bits and put together a 50 base pair sequence. Then you listen to the next 100 bits. Wait a minute! They're the exact same sequence! How can you encode them? Well, the easiest way would be to send a 100 bit stream, and then attach, "Message repeats". There is additional information: the information that tells the recipient that the message repeats (and as a result the message will need, say, 105 bits, instead of the original 100). Furthermore, if a point mutation happens in the second segment, then you will have to use a full 200 bit encoding to represent it.

So before realizing that it was a polyploidy, the polyploidy represented double the amount of info. After realizing that it was a polyploidy, the polyploidy represented slightly more than the original amount of info. Did AiG ever tell you that?

Under this scheme, evolution would indeed require an increase in information content. But mutations would also increase information content, insertion mutations in particular. Now, if there are many ways to quantify a process, the actual direction of the process doesn't change even if the process is quantified differently. (Whether I use algebra or differentiation to find the minimum of a quadratic, the minimum is the same.) Therefore evolution is not impossible from the angle of information content. If one defines information in such a way that it is impossible for mutations to increase information, the chances are that by that same definition evolution would not require an increase in information.

Thanks. An interesting approach. From my understanding you have oversimplified the problem.

 

[shernren]

Elijah: here's a good starter link: http://www.utdallas.edu/~cirillo/nats/day18.htm and go Google any of the terms there if you want to go indepth.

well, we can assume the recipient would know. It's a theoretical example. When it comes to DNA, the message would "end" with the reading of a stop codon.

Micaiah: I'm deliberately being simplistic. In fact a protein could be quantified in a similar way, say by taking 5 bits to encode each of the 20 amino acids, since a protein is essentially a very long chain of amino acids. But then you would find proteins which are very very long, in other words having a high information content, and yet when actually synthesized cannot do any useful function at all. One would perhaps define a sort of "functional information content", or the amount of information required to encode the functional characteristics and behaviour of a protein. But then it would be demonstrable that mutations change the behaviour of a protein and therefore add function. The link between "change" and "add" in that last sentence was vague, of course, but then again the definition is vague to begin with.

 

[random_guy]

and where did I say this? I haven't even talked about information theory and how it applies to DNA. I think you got the wrong person.

Whoops, I'm sorry. I got you confused with Micaiah. I think he was the one that argued using information theory. Again, my apologies.

 

[elijah115]

Elijah: here's a good starter link: http://www.utdallas.edu/~cirillo/nats/day18.htm and go Google any of the terms there if you want to go indepth.

er thanks.... but I meant ..... "Author, Journal, Year, Vol, Page" or "Author, Year Book, Publisher, Location" if appropriate

 

[Micaiah]

Whoops, I'm sorry. I got you confused with Micaiah. I think he was the one that argued using information theory. Again, my apologies.

TE's have made the claim there are examples of the type of mutations required for 'slime to scientist' evolution to occur. If TE's want a mathemaitical model to prove this occurs, they are the ones who should provide the model.

If we could witness a mutation that inserted a complete new gene; or a copy of a gene was inserted into the DNA and this was changed into a completely new gene and all changes occured randomly I'd accept this as information gain of the type required for evolution.

The contention over definitions occurs because we don't have clear examples such as these. Would such examples prove evolution. It would certainly add weight to the evolutionist's claim, but it should be kept in mind that showing that something could happen and showing that something did happen are two different things.

 

[Critias]

Well, for starters, it is believed that amino acids and RNA can arise spontaneously from basic organic compounds. This has been laboratory demonstrated (not just in Miller-Urey, mind you), although of course we may never be exactly sure just what combination took life. Also, membranes resembling cell membranes can form spontaneously from lipids, which can perform functions that life performs such as swelling or shrinking based on the environment and even storing energy as potential across the membrane. All said and done, though, there's still a lot we don't know. For both creationist and evolutionist.

This is the type of thing I was speaking about in another thread here. Just because in a lab, with specific conditions and environments set, doesn't mean that it is true in a nature setting where conditions and environments are not controlled like the labs.

 

[Critias]

Whoops, I'm sorry. I got you confused with Micaiah. I think he was the one that argued using information theory. Again, my apologies.

No worries, I figured it was a mistake. If it wasn't, I was curious why I needed to support something I never said, hehe.

 

[shernren]

The Nylon Bug! The example has not been disproven yet. The nylon-digesting enzyme was completely new, and not a degenerate modification of an already-existing enzyme. Even Lee Spetner (IIRC) admitted that it amounted to an increase in information.

As for the "contention over definitions", I have already put forwards one rigorous though simple model that supports evolution and another vague but intuitive model that also supports evolution. I would imagine the onus is now on the creationists to provide their information model that disproves evolution.

Well, elijah, here's a link which has quite a number of quoted papers at the bottom.

http://www.talkorigins.org/faqs/abioprob/abioprob.html

 

[random_guy]

TE's have made the claim there are examples of the type of mutations required for 'slime to scientist' evolution to occur. If TE's want a mathemaitical model to prove this occurs, they are the ones who should provide the model.

If we could witness a mutation that inserted a complete new gene; or a copy of a gene was inserted into the DNA and this was changed into a completely new gene and all changes occured randomly I'd accept this as information gain of the type required for evolution.

The contention over definitions occurs because we don't have clear examples such as these. Would such examples prove evolution. It would certainly add weight to the evolutionist's claim, but it should be kept in mind that showing that something could happen and showing that something did happen are two different things.

Two problems. Biologists do have a method for propagating changes, mutations and natural selection. If you want to argue against evolution uses information theory, you need to come up with a metric, not the biologists.

Second, We have things like new genes being inserted. Transposons, or jumping genes, for example are used in human immunity systems to create different antibodies. However, you refuse to even define information gain in DNA, so why should we even take your argument seriously?

On one hand, we have a mechanism for change, and no evidence of any barriers that prevent a kind from changing into another kind, on the other hand, we have your word that DNA can't increase in information, you can't even define what "increase" means, and you haven't even presented any evidence.

Pretty obvious evolution is considered a science and Creationism is not.

 

[mixin machine]

If we could witness a mutation that inserted a complete new gene; or a copy of a gene was inserted into the DNA and this was changed into a completely new gene and all changes occured randomly I'd accept this as information gain of the type required for evolution.

That's true but statistically what are the odds of a mutation happening continuiously. Not one example, more like thousands of examples...

Andrew

 

[shernren]

Not really ... It's not as if there is a pre-specified order of mutations to be worked through and if even one goes missing the whole story of life is all wrong. A beneficial mutation, once produced, will persist in the population for as long as the environment allows it to be a beneficial mutation. The only real requirement (and not a very rigid one, too) for two mutations to interact is that the second mutation has to be produced after the first one and while the environment is still favourable for the first one, or vice versa.

In any case, the chances of any one series of mutations happening might be unlikely in itself, but over the set of all possible series of mutations (which creationists have not processed completely to find that all possible series of mutations do not allow evolution) one has to occur, and by God's providence it happened to be the right one.

 

[gluadys]

If we could witness a mutation that inserted a complete new gene; or a copy of a gene was inserted into the DNA and this was changed into a completely new gene and all changes occured randomly I'd accept this as information gain of the type required for evolution.

Genes are normally composed of thousands of DNA base nucleotides. Many mutations affect a single nucleotide or a small group of nucleotides. Of course, the insertion or deletion of a single base nucleotide can affect the sequence of hundreds of codons (the three-base nucleotides that code for amino acids). On the other hand when insertions or deletions, even large ones, occur in the non-coding section of a gene, you can get a lot of change without any effect on the protein product.

So what is a "completely new gene"? Is it one in which many changes occur, but with little or no effect on the protein product? Or one in which only a few changes occur with a major effect on the protein product--i.e. a different protein is produced? Or one in which a particular % of the base nucleotides have changed whether or not there is a change in the protein product?

 

[Micaiah]

In my post I used the term gene to include the commonly understood idea that the mutation that occured resulted in some new heretitable characteristic in offspring. The mutation could occur in a single block, or it could be a sequence of nucleotide changes over a long period of time. Either way, most scientists today conclude mutations are random, a thing which Gluadys doesn't yet seem to understand. It is most scientists today conclude mutations are random also generally recognised that large scale mutations will almost inevitaby result in the death of the host since the chance of the mutation resulting in a change that is beneficial becomes increasing unlikely as the number of nucleotides in the mutation increases. That is why the NDT proposed that variation is the result of single nucleotide substituations also known as point mutations.

Spetner notes the following (1)

Geneticists have found that inversions, deletions, insertions, and transpositions are are not just haphazard events. Special pieces of DNA that jump around in the chromosome cause these genetic changes. Short pieces of DNA, called transposons, have been found to jump from place to place in the chromosome. They have also been found to activate other special, shorter pieces and make them jump as well. A piece of DNA that a transposon has activated is called an insertion sequence (IS). It is so called because it can be inserted into a gene. It is taken from one place in the genome and inserted into another. An IS, once inserted, can be deleted again. It can be amplified into many copies and it can be turned around to make an inversion. Transposon's and IS's can jump not only from one place to another [Darnell et al. 1986, Federoff 1984]. An IS even from a plasmid (a small structure in the cell containing DNA that is not in the chromosome) can be inserted into the chromosome. Recombination between two identical IS's can delete the segment between them, duplicate it, or invert it [Stryder 1988]. In this way the IS's can be responsible for deletions, duplications, and inversions.

A transposon has in its sections of DNA that encode two of the enzymes it needs to carry out its job. The cell itself contributes the other necessary enzymes. The motion of these genetic elements about to produce the above mutations has been found to be a complex process and we probably haven't yet discovered all the complexity. But because no one knows why they occur, many geneticists have assumed they occur only by chance. I find it hard to believe that a process as precise and as well controlled as the transposition of genetic elements happens only by chance. Some scientists tend to call a mechanism random before we learn what it really does. If the source of the variation for evolution were point mutations, we could say the variation is random. But if the source of the variation is the complex process of transposition, then there is no justification for saying that evolution is based on random events. I'll return to this subject in Chapter 5 and again in Chapter 7.

The paragraphs quoted above sum up one of the main points of the book. Namely that the large scale genetic variations that we see do not happen randomly and therefore do not meet the requirements of Neo Darwinian evolution. Spetner [21-21] notes that the Modern Synthetic Theory of Evolution now generally called the Neo-Darwinian Theory of Evolution is built on the assumption that variation is the result of random mutation of genes, and not by environmental influences most scientists today conclude mutations are random or acquired characteristics. Darwin had previously concluded the variation did result from environmental influences.

I do not know if Spetner would call himself a Christian, but he has the credentials to make comment of this topic. For those interested, I suggest you get a hold of the book and have a close look at what he says. If nothing else it is thought provoking. It is not what you'd call light reading, but you may take some comfort from his comment that he most scientists today conclude mutations are random left out the mathematics in a bid to make it more readable. Having said that, the maths is available.

1. Spetner, L. (1997). Not by Chance,The Judaica Press, 42-44

 

[gluadys]

In my post I used the term gene to include the commonly understood idea that the mutation that occured resulted in some new heretitable characteristic in offspring.

ok, as long as you realize that there are also many mutations (possibly the majority of mutations) which do not have the effect of producing a new heritable characteristic.

I would now like to ask what you mean by a "new" heritable characteristic. Occasionally we do see something quite new, like the mutation which gave a bacterium the capacity to digest nylon. But more often we see a variation in a characteristic the species already possesses e.g. a reduction in the number of limbs, a change in the shape of a bone, etc. Are these still "new"?

The mutation could occur in a single block, or it could be a sequence of nucleotide changes over a long period of time. Either way, most scientists today conclude mutations are random, a thing which Gluadys doesn't yet seem to understand. It is most scientists today conclude mutations are random

I understand that perfectly and have never denied it. What I have said is that evolution is not random. Btw, when we speak of mutations being random, what does it mean? Random with respect to what?


Spetner notes the following (1)--excerpt

The motion of these genetic elements about to produce the above mutations [transpositions] has been found to be a complex process and we probably haven't yet discovered all the complexity. But because no one knows why occur, many geneticists have assumed they occur only by chance. I find it hard to believe that a process as precise and as well controlled as the transposition of genetic elements happens only by chance. Some scientists tend to call a mechanism random before we learn what it really does. If the source of the variation for evolution were point mutations, we could say the variation is random. But if the source of the variation is the complex process of transposition, then there is no justification for saying that evolution is based on random events. I'll return to this subject in Chapter 5 and again in Chapter 7.

This again goes into the precise meaning of "random" and "chance". I think a lot of people confuse description and cause when listening to scientists on this matter. As Spetner notes above the activity of transposons is "a complex process and we probably haven't yet discovered all the complexity."

He then implies that geneticists have prematurely said it "happens only by chance". In short, he suggests that chance is being invoked as a causative agent.

But it is not. Geneticists are well aware of many causative agents that produce mutations. The production of mutations is probably not a chance process. What scientists usually mean by "chance" is not the absence of an understandable causative agent, but the unpredictability of when and where that causative agent will have an effect due to the complexity of the process. "Chance" in this sense is not so much something that occurs in nature as a failure of the human mind to be able (as yet) to grasp what is happening in nature. In short, it is a confession of ignorance, not an attribution of the event to the goddess Fortuna.

So when Spetner notes that:
"Some scientists tend to call a mechanism random before we learn what it really does. "

he misunderstands the situation. A mechanism is called "random" precisely because scientists have not figured out yet what it does. If we fully understood the mechanism, it might not look random at all. I very much doubt that any scientist referring to a mechanism as "random" really thinks it has no understandable cause and process---only that we don't know enough about it to predict specific outcomes. **

The paragraphs quoted above sum up one of the main points of the book. Namely that the large scale genetic variations that we see do not happen randomly and therefore do not meet the requirements of Neo Darwinian evolution.

Where did you get the idea that evolution, including neo-Darwinian evolution, requires genetic variations to happen randomly? The fact that we observe genetic variations to occur randomly does not make random occurrences a necessity. Natural selection would work as well on non-random variations as random ones--in fact it would probably work more efficiently with non-random variations.

Remember it is neither mutations per se, nor the resulting variations, that drive evolution. It is natural selection. That is why evolution is not random, even though it builds on random events. Natural selection removes the random consequences of mutations and produces a non-random change in the species.

Spetner [21-21] notes that the Modern Synthetic Theory of Evolution now generally called the Neo-Darwinian Theory of Evolution is built on the assumption that variation is the result of random mutation of genes, and not by environmental influences most scientists today conclude mutations are random or acquired characteristics. Darwin had previously concluded the variation did result from environmental influences.

This is a distortion of both neo-Darwinian theory and Darwin's original theory.

There are two senses in which "environmental influence" can be understood. One is the Lamarckian sense in which the environment induces a change in the species because of its behavioural response to the environment e.g. the giraffe reaching for the highest leaves on the tree stretches its neck and eventually giraffes are born with longer necks. While Darwin did not fully reject this idea, his concept of natural selection was the reverse of the Lamarckian process. Darwinian theory pre-supposes that variations already exist--who knows why--and that some variations are non-randomly selected for broader distribution in the next generation, thus becoming more common in the species than other variations. Eventually the selection process will lead to all organisms in the species showing the selected variation.

But what is the factor that "selects" the favoured variation? It is environmental influence. The difference between Lamarkism and Darwinism is not that one includes environmental influence and the other does not, but with the placement of environmental influence in the process. Lamarck makes it the agent of variation--evolution begins with environmental influences causing variations in adult organisms which are passed to their offspring. Darwin places environmental influence subsequent to variation as a selective agent discriminating among variations which already exist.

So does neo-Darwinism. What neo-Darwinism adds to Darwin's original thesis is a cause for variation. Darwin knew that species varied, but he did not know why. Thanks to geneticists, we now know why. Variations are caused by genetic mutations.

We even know some of the causes of genetic mutations.

What we don't know about mutations are:

what all the factors involved in mutations are
how to predict when and where a mutation will happen

There is also a lot we don't know about genes yet. We know in general that they produce proteins which in turn create, control and sustain the characteristics of the cell and organism. But we are pretty vague yet on the specifics. We know it is not the simplistic one gene-one characteristic model. But working out the precise relationships between genes and physical characteristics is an immensely complicated task.

So although the neo-Darwinian theory solves one question (where do variations come from) there are a lot it still leaves unanswered. What it does do is completely reject the Lamarkian concept of environmental influence in the initial stage of producing variation.

However, it still includes environmental influences in other important ways:

1. Environmental influences are sometimes the cause of mutations e.g. radiation in the environment causes mutations--and some of these in turn will cause variations. Some chemicals as well are mutagenic.

2. Environmental influences can affect the expression of a gene. Some genes require a particular environmental factor to be present in order to activate them.

3. Environmental influences are the agent of natural selection. By favoring certain variations, natural selection also favours the genetic patterns that produced them.

The excerpt from Spetner above concludes with this observation:

"If the source of the variation for evolution were point mutations, we could say the variation is random. But if the source of the variation is the complex process of transposition, then there is no justification for saying that evolution is based on random events."

This overlooks the fact that there was never any justification for saying that evolution is based on random events in the first place. Mutations and variations in and of themselves never produce evolution. They simply increase the range of variability within the species. Evolution is the product of natural selection acting on variation (and indirectly on mutations) and that is not a random process.

**With the exception of quantum events which really are random.