|
Users viewing this topic:
none
|
|
Login  | |
|
Evolution and Information - 8/28/2008 3:55:12 PM
|
|
|
essentialsaltes
Posts: 1024
Joined: 10/14/2007
From: Inglewood, CA
Status: offline
|
I don't know much about venus fly-traps except that they're neato. So I'm branching this discussion off before it arouses the ire of the mods. Refer to the More Proof thread for the lead-in.
Quoting DanJames:
quote:
you know, they actually said something like that in the class. "Mutations add information", and they then went on to list a thousand ways that information is damaged and possibly repaired, but never gave a mechanism for information to be added. Information is certainly regulated so that bulliform cells can be evacuated more explosively, or information can be damaged so a gland that would be producing a nectar is never turned off, thus producing a nectary (I made that up, I know of no such case). But in all cases information had to already be present in order for the damage or regulation to happen. Information is never added for natural selection to act on. If my nectary situation were to happen, the plant would loose the function of that gland, and we would have a generation of plants with less information, but perhaps, to the benefit of that plant.
I think this confuses two meaning of information. One is the information stored in the genome. The other is the information embodied in a particular protein. Obviously, there is some connection between the two. But they are not identical. Let's look first at the information in the genome. The information content of a message is defined with a particular mathematical formula that stresses the probability of knowing the next chunk of information. The more uncertain the next chunk is, the more information is in that text.
As an example, how much information is encoded in a sentence picked at random out of a book. You read the first word, "The...". That doesn't tell you much about what the next word is going to be, so the next chunk has a lot of information. "The quick..." not so many things could follow that, but there's still plenty of uncertainty, so the next chunk has some information content. "The quick brown..." Okay, now most of us have a pretty good idea what's coming next. Probably the odds that the next word is 'fox' is about 99%. I don't have to give you any more information, but you can already guess the entire content of the sentence. So this sentence doesn't have nearly as much information as a sentence that you'd have to read all the way to the end to determine its meaning.
Back to the genetic code, as you're reading along AGTCTAGCTA..., suppose you as a smart researcher recognize that as the start of the sequence for whateveritis-ase in some species. Suppose that there is only one version of the whateveritis-ase in the population. Then the rest of the message has no more information content, because there is no uncertainty in any of the rest of the message.
If we consider a different population with some neutral mutations, say some changes in nucleotides that don't affect the amino acid coding, then the whateveritis-ase is completely unchanged. So the information embodied in the protein has not changed, but the information in the genome has increased. At locations where these mutations have taken place, there is uncertainty about which 'letter' is going to be there, so the information of the code has increased.
If we look at a more realistic population, there may be some variant forms of whateveritis-ase, and each of these forms may have different neutral spellings. This population has an even greater information content in its total genome.
Every novel mutation at a specific location increases the uncertainty in knowing what that nucleotide is. Every novel mutation increases the information of the gene pool.
Of course, many of these increases in information are fatal. These novel messages will get weeded out by selection, eliminating some (but not necessarily all) of the new informational content of the total genome. Having a busted protein does not mean that the genetic information has decreased. Quite the opposite. It may not be useful, or beneficial, to that particular organism, but the information content of the species' genome has increased. (albeit briefly, if it's a fatal mutation)
And, on some occasions, these increases in information will be beneficial. Such as the nylon bug: "... the capacity to synthesize nylonase most probably developed as a single-step mutation that survived because it improved the fitness of the bacteria possessing the mutation."
Now if this gene formerly coded for whateveritis-ase, you might be tempted to say that it now has a busted whateveritis-ase protein and has less information. But mathematically, the genomic information has increased, and even better, the busted whateveritis-ase is maybe better described as a functional nylonase.
Prior to 1930, functional nylonase would not have been beneficial. That new information may have been spit up by mutation over and over again, and suppressed by selection. But now that new information has value in the current environment.
quote:
Let's not mince words. Unless the change occurs in a regulation zone, the information is damaged. Causing the DNA to code for another amino acid may cause the protein to look or behave differently, but this has an uncanny tendency to cause it to act more poorly to some degree. It would seem, at least in all the examples that I know of, that the protein is coded for in the DNA exactly how it should be to do the job that it was intended to in the best way possible.
It may have a tendency to be detrimental, but not a certainty. The nylon bug is an example. The AIMilano apoprotein mutation seems to have arisen in 18th century Italy. It involves an amino acid substitution. I don't know if the variant protein does its job any better or worse than the usual form, but it significantly reduces the risk of atherosclerosis. I think there are hundreds of variant versions of particular proteins, and some must have at least slightly better function than others. So many people are carrying around proteins that are not the 'best way possible'. It's possible they could pass on to their children a mutant version that would be one of the more beneficial variants.
quote:
As I hope to have demonstrated, nothing results in a pool with more information. A diploid organism heterozygous for a mutation has one copy of a good gene and one copy of a busted gene. The deciding factor for natural selection is whether or not the creature can cope with the busted gene. Information has been taken away, and the proteins coded for in that busted gene are no longer serving their function.
The busted proteins may serve an improved or different function. But regardless of their positive or negative character, every change away from uniformity increases the information in the gene pool. Mutations provide that increase in information; selection winnows that increased information, favoring beneficial information, culling unfavorable information.
_____________________________
"My object in all arguments is not to make any preconceived opinion of mine seem right, but merely to discover and establish the truth, whatever the truth may be."
-- HP Lovecraft, letter to Robert E. Howard 7/27-28/34
|
|
|
|
|
RE: Evolution and Information - 8/28/2008 4:12:24 PM
|
|
|
Jhud
Posts: 7589
Joined: 4/11/2005
From: Lake Wobegon
Status: online
|
I’ll get to the rest later, but I wanted to camp on this statement first:
quote:
Every novel mutation at a specific location increases the uncertainty in knowing what that nucleotide is. Every novel mutation increases the information of the gene pool.
In your idea of ‘information’, how is it that a change to a particular gene is an increase in information? We can all agree for example that a single point mutation represents a change in the information available, and that such a change might even evoke a useful modification to an organism, but is it necessarily an increase?
And even if it is, is increasing information help at all? In a radio transmission we can increase information in the form of noise, but that is bad. In computer code we can randomly add information, but that can cause inefficiencies and conflicts. What seems rally to be at question is an increase in useful information that is capable of working together with existing information or additional novel information in an organized fashion to produce a beneficial result, and observing this occur is a rare, if almost non-existent phenomena in nature.
_____________________________
Jack
I believe in Christianity as I believe that the sun has risen: not only because I see it, but because by it I see everything else.
- C.S. Lewis
|
|
|
|
|
RE: Evolution and Information - 8/28/2008 5:26:31 PM
|
|
|
essentialsaltes
Posts: 1024
Joined: 10/14/2007
From: Inglewood, CA
Status: offline
|
quote:
ORIGINAL: Jhud
I’ll get to the rest later, but I wanted to camp on this statement first:
quote:
Every novel mutation at a specific location increases the uncertainty in knowing what that nucleotide is. Every novel mutation increases the information of the gene pool.
In your idea of ‘information’, how is it that a change to a particular gene is an increase in information? We can all agree for example that a single point mutation represents a change in the information available, and that such a change might even evoke a useful modification to an organism, but is it necessarily an increase?
I think the correct way to think of this is at the species level, which is why I speak of the gene pool, rather than an individual. A novel mutation of a particular gene contributes a new allele (is that right?) to the gene pool that wasn't previously present.
A single animal can only have one nucleotide in a particular place, so there is no 'uncertainty': whatever comes next comes next. But with a gene pool, a single gene can have multiple spellings, and these (and their frequencies) determine the probabilities that go into the definition for information.
quote:
And even if it is, is increasing information help at all? In a radio transmission we can increase information in the form of noise, but that is bad. In computer code we can randomly add information, but that can cause inefficiencies and conflicts.
This is a much better question, and again it seems to drive at the two different uses of information. The Shannon definition from information theory is easy to apply to nucleotide sequences, since it's essentially a digital format. Noise is unwanted for ordinary human purposes, since it 'messes up' the intended message, but nevertheless noise represents an increase in information and complexity.
quote:
What seems rally to be at question is an increase in useful information that is capable of working together with existing information or additional novel information in an organized fashion to produce a beneficial result, and observing this occur is a rare, if almost non-existent phenomena in nature.
This 'useful information' and 'existing information' somehow exists in the proteins and all the other interconnected phenomena of living cells. I don't know how to calculate the information of that. Is there a biological definition of information that pertains to this? Is there a definition of 'useful information'? If we define useful information as information that can be used to create a particular form of whateveritis-ase (the intended message of this gene is whateveritis-ase) then any change in that gene leads to a loss in 'useful information'. But only because we have defined 'useful information' by assuming that whateveritis-ase is the only useful target. With this definition, nylonase is just an unintended error, no matter how useful it happens to be.
This ties in, I think, with the problems of calculating probabilities by drawing a circle around the current situation (the intended outcome), and neglecting the many other possible outcomes (excluded as unintended outcomes).
I'm not sure information is the right concept for these situations. Yes, biological systems are very complicated, and random tinkering is not likely to result in greater reproductive success. Agreed. But even rare beneficial results provide an engine for change. Whether this evolutionary engine can get us from A to B at the observed rates of change and the time allowed is a good question. But it is separate from the information question.
Looking at the information in the gene pool:
Novel mutations increase the available information.
Selection reduces and focuses that information toward beneficial information-types.
_____________________________
"My object in all arguments is not to make any preconceived opinion of mine seem right, but merely to discover and establish the truth, whatever the truth may be."
-- HP Lovecraft, letter to Robert E. Howard 7/27-28/34
|
|
|
|
|
RE: Evolution and Information - 8/29/2008 1:36:27 PM
|
|
|
DanJames
Posts: 679
Joined: 12/13/2007
Status: offline
|
quote:
ORIGINAL: essentialsaltes
This 'useful information' and 'existing information' somehow exists in the proteins and all the other interconnected phenomena of living cells. I don't know how to calculate the information of that. Is there a biological definition of information that pertains to this? Is there a definition of 'useful information'? If we define useful information as information that can be used to create a particular form of whateveritis-ase (the intended message of this gene is whateveritis-ase) then any change in that gene leads to a loss in 'useful information'. But only because we have defined 'useful information' by assuming that whateveritis-ase is the only useful target. With this definition, nylonase is just an unintended error, no matter how useful it happens to be.
This ties in, I think, with the problems of calculating probabilities by drawing a circle around the current situation (the intended outcome), and neglecting the many other possible outcomes (excluded as unintended outcomes).
I'm not sure information is the right concept for these situations. Yes, biological systems are very complicated, and random tinkering is not likely to result in greater reproductive success. Agreed. But even rare beneficial results provide an engine for change. Whether this evolutionary engine can get us from A to B at the observed rates of change and the time allowed is a good question. But it is separate from the information question.
Looking at the information in the gene pool:
Novel mutations increase the available information.
Selection reduces and focuses that information toward beneficial information-types.
I'm not enjoying the information metaphors all that much. I think we understand what DNA and mutations and such really are. The sequence ATC codes for isoleucine. This amino acid has a list of characterizations that make it ideal for a certain number of functions when placed into an enzyme such as being non-polar. Other amino acids are similar in some ways and different in others, but every amino acid can be thought to be unique. If, in the course of an organisms life, a change is made where the amino acid is replaced by a polar amino acid, the functionality of that a.a. is compromised. The protein is busted, but to what extent is determined by how different the protein becomes. This is the kind of information that is being added into the gene pool. The question is whether or not the organism can overcome its new information with the other set of genetic material. I think the point is that this process of busting our proteins (with the occasional example of a bacterium being able to process nylon) is not the kind of robust information-producing process needed to make a human being out of chemical soup.
|
|
|
|
|
RE: Evolution and Information - 9/2/2008 7:48:34 AM
|
|
|
BVZ
Posts: 306
Joined: 11/2/2006
Status: offline
|
Any mutation can be reversed by another mutation.
If a mutation increases information, reversing that mutation will decrease information.
If a mutation decreases information, reversing that mutation will increase information.
In other words, mutations CAN increase the information in a genome.
Easy.
|
|
|
|
|
RE: Evolution and Information - 9/2/2008 12:49:45 PM
|
|
|
DanJames
Posts: 679
Joined: 12/13/2007
Status: offline
|
quote:
ORIGINAL: BVZ
Any mutation can be reversed by another mutation.
If a mutation increases information, reversing that mutation will decrease information.
If a mutation decreases information, reversing that mutation will increase information.
In other words, mutations CAN increase the information in a genome.
Easy.
I don't think this position is really thought through. There is the possibility of one mutation masking another. A mutation for a flower having more petals than normal can be masked by a mutation for no petals at all. The possibility of a spontaneos mutation that turns off another mutation is extremely unlikely, and even if it did happen at some point in earth's history, that really does nothing at all to help your point. Your defenition of an increase in information being the spontaneous fixing of another mutation with a mutation will not be accepted by evolutionists or IDists. For evolutionists, you have no mathmatical potential for raising the amount of information above zero because if there are no mutations, there is no posibility to increase information. For the IDist... well it's just not a reflection of any naturaly known process. If you're proposing that the robust mechanism for bringing about complex organ systems is millions of years of reversing mutations, I'd say your theory is... wanting.
|
|
|
|
|
RE: Evolution and Information - 9/3/2008 8:41:21 AM
|
|
|
BVZ
Posts: 306
Joined: 11/2/2006
Status: offline
|
quote:
ORIGINAL: DanJames
quote:
ORIGINAL: BVZ
Any mutation can be reversed by another mutation.
If a mutation increases information, reversing that mutation will decrease information.
If a mutation decreases information, reversing that mutation will increase information.
In other words, mutations CAN increase the information in a genome.
Easy.
I don't think this position is really thought through. There is the possibility of one mutation masking another. A mutation for a flower having more petals than normal can be masked by a mutation for no petals at all. The possibility of a spontaneos mutation that turns off another mutation is extremely unlikely, and even if it did happen at some point in earth's history, that really does nothing at all to help your point. Your defenition of an increase in information being the spontaneous fixing of another mutation with a mutation will not be accepted by evolutionists or IDists. For evolutionists, you have no mathmatical potential for raising the amount of information above zero because if there are no mutations, there is no posibility to increase information. For the IDist... well it's just not a reflection of any naturaly known process. If you're proposing that the robust mechanism for bringing about complex organ systems is millions of years of reversing mutations, I'd say your theory is... wanting.
Are you saying that there can be a mutation that cannot be reversed by another mutation? Can you give me an example of such a mutation?
Because without showing that a mutation cannot be reversed by another, my argument stands.
|
|
|
|
|
RE: Evolution and Information - 9/3/2008 12:43:00 PM
|
|
|
DanJames
Posts: 679
Joined: 12/13/2007
Status: offline
|
quote:
ORIGINAL: BVZ
quote:
ORIGINAL: DanJames
quote:
ORIGINAL: BVZ
Any mutation can be reversed by another mutation.
If a mutation increases information, reversing that mutation will decrease information.
If a mutation decreases information, reversing that mutation will increase information.
In other words, mutations CAN increase the information in a genome.
Easy.
I don't think this position is really thought through. There is the possibility of one mutation masking another. A mutation for a flower having more petals than normal can be masked by a mutation for no petals at all. The possibility of a spontaneos mutation that turns off another mutation is extremely unlikely, and even if it did happen at some point in earth's history, that really does nothing at all to help your point. Your defenition of an increase in information being the spontaneous fixing of another mutation with a mutation will not be accepted by evolutionists or IDists. For evolutionists, you have no mathmatical potential for raising the amount of information above zero because if there are no mutations, there is no posibility to increase information. For the IDist... well it's just not a reflection of any naturaly known process. If you're proposing that the robust mechanism for bringing about complex organ systems is millions of years of reversing mutations, I'd say your theory is... wanting.
Are you saying that there can be a mutation that cannot be reversed by another mutation? Can you give me an example of such a mutation?
Because without showing that a mutation cannot be reversed by another, my argument stands.
No, it really doesn't. Every mutation could be undone with an equal and opposite mutation. Sure. Ask ES if your argument is valid. I'm sure he'll agree with me that what you're saying about DNA has no basis in reality and is not a helpful argument for UCD.
|
|
|
|
|
RE: Evolution and Information - 9/3/2008 1:32:50 PM
|
|
|
Veritas
Posts: 526
Joined: 4/19/2005
Status: online
|
quote:
ORIGINAL: DanJames
quote:
ORIGINAL: BVZ
Are you saying that there can be a mutation that cannot be reversed by another mutation? Can you give me an example of such a mutation?
Because without showing that a mutation cannot be reversed by another, my argument stands.
No, it really doesn't. Every mutation could be undone with an equal and opposite mutation. Sure. Ask ES if your argument is valid. I'm sure he'll agree with me that what you're saying about DNA has no basis in reality and is not a helpful argument for UCD.
You're missing the point. The point is not that an increase in information must be the reverse of another mutation. The point is that anytime you have mutations that decrease information, you must also have mutations that increase information. You can't have one without having the other.
|
|
|
|
|
RE: Evolution and Information - 9/3/2008 3:42:16 PM
|
|
|
essentialsaltes
Posts: 1024
Joined: 10/14/2007
From: Inglewood, CA
Status: offline
|
quote:
ORIGINAL: Veritas
quote:
ORIGINAL: DanJames
quote:
ORIGINAL: BVZ
Are you saying that there can be a mutation that cannot be reversed by another mutation? Can you give me an example of such a mutation?
Because without showing that a mutation cannot be reversed by another, my argument stands.
No, it really doesn't. Every mutation could be undone with an equal and opposite mutation. Sure. Ask ES if your argument is valid. I'm sure he'll agree with me that what you're saying about DNA has no basis in reality and is not a helpful argument for UCD.
You're missing the point. The point is not that an increase in information must be the reverse of another mutation. The point is that anytime you have mutations that decrease information, you must also have mutations that increase information. You can't have one without having the other.
Since DJ namechecked me, I'll jump in. The analogies with thermodynamics continue to plague my brain.
DJ may be envisioning what we might call a biosphere in genetic equilibrium. For every baby born with mutation A, another baby is born somewhere else with un-mutation -A. In this case, the total information of the species would remain the same. The fact that the mutation going one way increases information and going the other way it decreases the information is not enough to establish that the total information is increasing, or how all that information arose in the first place.
I would just add that we are not in 'genetic equilibrium'. Even counting only the alleles that currently exist in the human genepool, I imagine there are, technically speaking, zillions more 'potential' humans than actual humans that have ever been born. And since every baby has umpteen mutations, its DNA is not only unique, but it's virtually guaranteed to be exploring new genetic 'spaces' that have not yet been tried, adding new possible spellings of whateveritis-ase and increasing the information of the species' genepool.
There are more mutations adding information (going from common, known alleles to novel, rare alleles) than subtracting information (going from rare, previously-novel alleles back to a common, known allele). With random mutations, it's far more likely to spell something differently than to spell something the same.
_____________________________
"My object in all arguments is not to make any preconceived opinion of mine seem right, but merely to discover and establish the truth, whatever the truth may be."
-- HP Lovecraft, letter to Robert E. Howard 7/27-28/34
|
|
|
|
|
RE: Evolution and Information - 9/3/2008 6:24:45 PM
|
|
|
DanJames
Posts: 679
Joined: 12/13/2007
Status: offline
|
quote:
ORIGINAL: essentialsaltes
quote:
ORIGINAL: Veritas
quote:
ORIGINAL: DanJames
quote:
ORIGINAL: BVZ
Are you saying that there can be a mutation that cannot be reversed by another mutation? Can you give me an example of such a mutation?
Because without showing that a mutation cannot be reversed by another, my argument stands.
No, it really doesn't. Every mutation could be undone with an equal and opposite mutation. Sure. Ask ES if your argument is valid. I'm sure he'll agree with me that what you're saying about DNA has no basis in reality and is not a helpful argument for UCD.
You're missing the point. The point is not that an increase in information must be the reverse of another mutation. The point is that anytime you have mutations that decrease information, you must also have mutations that increase information. You can't have one without having the other.
Since DJ namechecked me, I'll jump in. The analogies with thermodynamics continue to plague my brain.
I consider them a plague as well. I didn't mean to call you out, it's just that... well, it's your thread.
quote:
DJ may be envisioning what we might call a biosphere in genetic equilibrium. For every baby born with mutation A, another baby is born somewhere else with un-mutation -A. In this case, the total information of the species would remain the same. The fact that the mutation going one way increases information and going the other way it decreases the information is not enough to establish that the total information is increasing, or how all that information arose in the first place.
Ok, I honestly have no idea what you're talking about. I don't envision a biosphere (you mean population perhaps?) with any kind of equilibrium. I envision a population with a set number of mutations, with more popping up all the time. Is that what you mean? Obviously you're not saying that just because a baby is born with Huntington's disease that another baby must be born without it.
quote:
I would just add that we are not in 'genetic equilibrium'. Even counting only the alleles that currently exist in the human genepool, I imagine there are, technically speaking, zillions more 'potential' humans than actual humans that have ever been born. And since every baby has umpteen mutations, its DNA is not only unique, but it's virtually guaranteed to be exploring new genetic 'spaces' that have not yet been tried, adding new possible spellings of whateveritis-ase and increasing the information of the species' genepool.
A child born with an addition or a deletion would suffer great effects if it occurs in the middle of a protein. A point change mutation would also cause harmful effects on that protein (or no effect at all if it occurs on a wobble base pair). The point being that changing isoleucine, with all of it's characteristics, into something else with all of its characteristics will almost certainly make an enzyme or protein that is less functional. I don't know of any examples where this is not the case, but perhaps you do. Changes in regulators are the only ways that I am aware of where you can get better information that can be selected against as favorable for the environment.quote:
There are more mutations adding information (going from common, known alleles to novel, rare alleles) than subtracting information (going from rare, previously-novel alleles back to a common, known allele). With random mutations, it's far more likely to spell something differently than to spell something the same.
I don't know how we can know for sure that there are more mutations adding information as you have defined it than there are subtracting from it, but if you're suggesting that a mutation (perhaps an addition of an extra base pair in the middle of a protein) that is turned off by another mutation (perhaps a subtraction in the exact proximity of the previous addition such that the mutation is canceled) can be ruled out as a possibility, and I don't see why it would help or be considered an increase of information as BVZ has put it. Perhaps I'm not understanding this concept correctly.
|
|
|
|
|
RE: Evolution and Information - 9/3/2008 10:31:06 PM
|
|
|
essentialsaltes
Posts: 1024
Joined: 10/14/2007
From: Inglewood, CA
Status: offline
|
quote:
ORIGINAL: DanJames
Ok, I honestly have no idea what you're talking about. I don't envision a biosphere (you mean population perhaps?) with any kind of equilibrium. I envision a population with a set number of mutations, with more popping up all the time. Is that what you mean?
That's not what I meant when I tried to mean what I thought you meant. Wait, let's just stop that. But if it's what you mean... more mutations popping up all the time increases the information in the gene pool.
quote:
A child born with an addition or a deletion would suffer great effects if it occurs in the middle of a protein.
Yes, yes, yes. This is what I meant about focusing on the 'information' embodied in a particular protein. I don't know how to count that information. The information I can count is in the DNA. Mutations add information. Novel DNA sequences add information. "Of course, many of these increases in information are fatal. These novel messages will get weeded out by selection, eliminating some (but not necessarily all) of the new informational content of the total genome. Having a busted protein does not mean that the genetic information has decreased. Quite the opposite. It may not be useful, or beneficial, to that particular organism, but the information content of the species' genome has increased. (albeit briefly, if it's a fatal mutation)"
quote:
The point being that changing isoleucine, with all of it's characteristics, into something else with all of its characteristics will almost certainly make an enzyme or protein that is less functional. I don't know of any examples where this is not the case, but perhaps you do.
A possible example is the Milano apoprotein mutation. The only known effect it has, so far as I know, is to reduce the risk of atherosclerosis. If it does 'its job' whatever that is at least roughly as well as the more common variant, but also confers protection against a serious disease, this is a net benefit to the organism.
quote:
I don't know how we can know for sure that there are more mutations adding information as you have defined it than there are subtracting from it
In the whole history of mankind, we have not exhausted the genetic diversity even within 'normal' (unmutated) human alleles. Barring extreme coincidence, every baby born expands the genetic diversity of the human genome. This produces an increase in information. (This is tied not just to the general increase in population, but also to the 'additional' diversity from novel mutations) Despite every baby having umpteen mutations, relatively few babies/zygotes die rapidly from genetic defects, so most mutations must be approximately or functionally neutral. So babies continue to add genetic diversity to the species' gene pool, and this results in an increase in the information content of the gene pool.
quote:
but if you're suggesting that a mutation (perhaps an addition of an extra base pair in the middle of a protein) that is turned off by another mutation (perhaps a subtraction in the exact proximity of the previous addition such that the mutation is canceled) can be ruled out as a possibility
No, that's not what I was aiming at.
----
Another analogy, just because I thought of it the other day.
Poor William Congreve composed a pithy aphorism, more than 300 years ago:
Music has charms to sooth the savage breast.
And yet, in a Googlefight, Congreve loses out to a point mutation. There are more than twice as many hits for:
Music has charms to sooth the savage beast.
Apparently, this mutation is more successful than the original meme. Congreve is spinning in his grave, so to him this is no doubt a 'busted' aphorism. But this busted form is more successful at having itself propagated. The environment prefers the mutated form. The existence of both forms demonstrates an increase in information (in the information theory sense) - since there is uncertainty in the letter following 'b'. You don't know how this meme ends until you've read that letter.
_____________________________
"My object in all arguments is not to make any preconceived opinion of mine seem right, but merely to discover and establish the truth, whatever the truth may be."
-- HP Lovecraft, letter to Robert E. Howard 7/27-28/34
|
|
|
|
|
RE: Evolution and Information - 9/4/2008 5:45:07 AM
|
|
|
BVZ
Posts: 306
Joined: 11/2/2006
Status: offline
|
quote:
ORIGINAL: DanJames
quote:
ORIGINAL: BVZ
Are you saying that there can be a mutation that cannot be reversed by another mutation? Can you give me an example of such a mutation?
Because without showing that a mutation cannot be reversed by another, my argument stands.
No, it really doesn't. Every mutation could be undone with an equal and opposite mutation. Sure.
So you agree then. Excellent.
quote:
Ask ES if your argument is valid. I'm sure he'll agree with me that what you're saying about DNA has no basis in reality and is not a helpful argument for UCD.
Why would I ask ES?
YOU are the one saying that my argument is not a valid one. YOU should show why this is the case.
Since you admit that any mutation can be reversed by another, it follows that any mutation that DECREASES information (no matter how you choose to measure it) can be reversed by another mutation.
And since the reverse of a mutation that decreases information MUST increase information, it follows that mutations CAN increase the information in any genome where mutations are possible.
My argument stands, and you have not even begun to attack it.
|
|
|
|
|
RE: Evolution and Information - 9/4/2008 3:28:03 PM
|
|
|
DanJames
Posts: 679
Joined: 12/13/2007
Status: offline
|
quote:
ORIGINAL: essentialsaltes
quote:
ORIGINAL: DanJames
Ok, I honestly have no idea what you're talking about. I don't envision a biosphere (you mean population perhaps?) with any kind of equilibrium. I envision a population with a set number of mutations, with more popping up all the time. Is that what you mean?
That's not what I meant when I tried to mean what I thought you meant. Wait, let's just stop that. But if it's what you mean... more mutations popping up all the time increases the information in the gene pool.
quote:
A child born with an addition or a deletion would suffer great effects if it occurs in the middle of a protein.
Yes, yes, yes. This is what I meant about focusing on the 'information' embodied in a particular protein. I don't know how to count that information. The information I can count is in the DNA. Mutations add information. Novel DNA sequences add information. "Of course, many of these increases in information are fatal. These novel messages will get weeded out by selection, eliminating some (but not necessarily all) of the new informational content of the total genome. Having a busted protein does not mean that the genetic information has decreased. Quite the opposite. It may not be useful, or beneficial, to that particular organism, but the information content of the species' genome has increased. (albeit briefly, if it's a fatal mutation)"
quote:
A possible example is the Milano apoprotein mutation. The only known effect it has, so far as I know, is to reduce the risk of atherosclerosis. If it does 'its job' whatever that is at least roughly as well as the more common variant, but also confers protection against a serious disease, this is a net benefit to the organism.
Well before we start giving people viral treatment to cause this mutation, let's not forget that the principal effect of mutations is to break genes that you want. Perhaps a new thread could be started to encourage people to look up this gene, since it is what you would call a "beneficial mutaiton."quote:
In the whole history of mankind, we have not exhausted the genetic diversity even within 'normal' (unmutated) human alleles. Barring extreme coincidence, every baby born expands the genetic diversity of the human genome. This produces an increase in information. (This is tied not just to the general increase in population, but also to the 'additional' diversity from novel mutations) Despite every baby having umpteen mutations, relatively few babies/zygotes die rapidly from genetic defects, so most mutations must be approximately or functionally neutral. So babies continue to add genetic diversity to the species' gene pool, and this results in an increase in the information content of the gene pool.
I can see how you would consider a mutation an increase in information. But it seems to me that you're of the opinion that a kid with more pronounced cheek bones has this feature as a result of a mutation. Which caused me to realize that I have no idea what goes into diversity. We are a combination of the combination of our parents' parents' genomes. Guess I have more questions to ask.
quote:
quote:
but if you're suggesting that a mutation (perhaps an addition of an extra base pair in the middle of a protein) that is turned off by another mutation (perhaps a subtraction in the exact proximity of the previous addition such that the mutation is canceled) can be ruled out as a possibility
No, that's not what I was aiming at.
----
Another analogy, just because I thought of it the other day.
Poor William Congreve composed a pithy aphorism, more than 300 years ago:
Music has charms to sooth the savage breast.
And yet, in a Googlefight, Congreve loses out to a point mutation. There are more than twice as many hits for:
Music has charms to sooth the savage beast.
Apparently, this mutation is more successful than the original meme. Congreve is spinning in his grave, so to him this is no doubt a 'busted' aphorism. But this busted form is more successful at having itself propagated. The environment prefers the mutated form. The existence of both forms demonstrates an increase in information (in the information theory sense) - since there is uncertainty in the letter following 'b'. You don't know how this meme ends until you've read that letter.
Well, I see how you would consider this an increase of information. On the one hand we had one sentence that pops up in Google, now we have two. The amount of information in the gene pool of the population of sentences is increased. Nevertheless, this analogy breaks down in light of the fact that the principle effect of a mutation is to break a protein, not to turn it into another word that fits into the context of a sentence. When this organism has kids, those might inherit this busted protein and the information in the gene pool increases (as you have put it). But what is introduced is a broken protein that the gene pool must overcome. This is NOT the kind of robust information producing machinery sufficient to explain the arrising of novel, complex organ systems. In my opinion.
|
|
|
|
|
RE: Evolution and Information - 9/4/2008 3:37:52 PM
|
|
|
DanJames
Posts: 679
Joined: 12/13/2007
Status: offline
|
quote:
ORIGINAL: BVZ
quote:
ORIGINAL: DanJames
quote:
ORIGINAL: BVZ
Are you saying that there can be a mutation that cannot be reversed by another mutation? Can you give me an example of such a mutation?
Because without showing that a mutation cannot be reversed by another, my argument stands.
No, it really doesn't. Every mutation could be undone with an equal and opposite mutation. Sure.
So you agree then. Excellent.
Yeah, I guess so, BVZ. I would also agree that every atom in the building in which I'm sitting could spontaneously slip into a fold in space, leaving me enjoying the beautiful, Fall, Nebraska weather. Both are possible, neither are likely.
quote:
quote:
Ask ES if your argument is valid. I'm sure he'll agree with me that what you're saying about DNA has no basis in reality and is not a helpful argument for UCD.
Why would I ask ES?
YOU are the one saying that my argument is not a valid one. YOU should show why this is the case.
Since you admit that any mutation can be reversed by another, it follows that any mutation that DECREASES information (no matter how you choose to measure it) can be reversed by another mutation.
And since the reverse of a mutation that decreases information MUST increase information, it follows that mutations CAN increase the information in any genome where mutations are possible.
Why do you keep saying this? I'm not interested in new defenitions to what is considered an increase of information. I've got ES's to deal with. Are you just screwing around?
Assuming you're being serious: Mutations are not counteracted by subsequent mutations. It was unlikely that the mutation would happen in the first place thanks to a very robust mutation-fixing process. (bet nobody praised God for that gem today) Point being, mutations are extremely unlikely, and if you're the base pair that gets affected, you are an unlucky base pair indeed. For that same base pair to undergo an equal and opposite mutation is zero. It's not going to happen. Or for a frame shift mutation to undergo an equal and opposite mutation such that the protein is fixed with NO deleterious effects, I would say, is quite near zero.
I must not be understanding this concept well. If I'm not addressing it can someone help explain it to me? Or can I get a witness?
|
|
|
|
|
RE: Evolution and Information - 9/4/2008 4:38:30 PM
|
|
|
essentialsaltes
Posts: 1024
Joined: 10/14/2007
From: Inglewood, CA
Status: offline
|
quote:
ORIGINAL: DanJames
Nevertheless, this analogy breaks down in light of the fact that the principle effect of a mutation is to break a protein ... But what is introduced is a broken protein that the gene pool must overcome.
Not all mutations are harmful. Not all proteins get busted when they're spelled 'wrong'. Some mutations can enhance protein function.
Even if rare, novel beneficial mutations should tend to survive, increasing the information content of the gene pool.
_____________________________
"My object in all arguments is not to make any preconceived opinion of mine seem right, but merely to discover and establish the truth, whatever the truth may be."
-- HP Lovecraft, letter to Robert E. Howard 7/27-28/34
|
|
|
|
|
RE: Evolution and Information - 9/4/2008 5:06:58 PM
|
|
|
essentialsaltes
Posts: 1024
Joined: 10/14/2007
From: Inglewood, CA
Status: offline
|
quote:
ORIGINAL: DanJames
Are you just screwing around?
No, BVZ's not. Let's go back to something you originally said in the previous thread: "Information is never added for natural selection to act on. If my nectary situation were to happen, the plant would loose the function of that gland, and we would have a generation of plants with less information"
This suggests that you are making the following claim:
Mutation only causes a loss of function, i.e. a loss of information.
Even though I don't like it, for this comment I will adopt the idea that function equals information.
Suppose that to bust a protein requires a C to change into an A. Ok, if this represents a loss of information, then going from an A to a C would represent a gain in information. Random mutations don't care which way they go, so a mutation can produce an increase in information, contrary to your claim.
Now hold your horses, you exclaim (or some other phrase). Mutations only bust things because the genome is already perfect.
I've already talked about cases where it appears that 'busted' proteins have enhanced function. But here's another angle.
OK, now most people acknowledge microevolution. Giraffes can get longer necks, or critters can change their size to take advantage of cooler or warmer environments, or pepper moths can turn dark or light, as the environment dictates. But the environmental pressure can press both ways. When smog got worse, was a mutation to a dark allele adding information to moths? When smog got better, was the same mutation now subtracting information? The functional success of the mutation depends on the environment. And the environment is not fixed. If function = information, then the exact same mutation can either add or subtract function/information depending on the context.
Or perhaps a better way to put it, when the environment changes, a species may want its proteins busted (in the right way) in order to enhance function/increase information.
quote:
It was unlikely that the mutation would happen in the first place thanks to a very robust mutation-fixing process. (bet nobody praised God for that gem today)
We established before that "In contrast to DNA damage, a mutation is a change in the base sequence of the DNA. A mutation cannot be recognized by enzymes once the base change is present in both DNA strands, and thus a mutation cannot be repaired."
_____________________________
"My object in all arguments is not to make any preconceived opinion of mine seem right, but merely to discover and establish the truth, whatever the truth may be."
-- HP Lovecraft, letter to Robert E. Howard 7/27-28/34
|
|
|
|
|
RE: Evolution and Information - 9/5/2008 2:12:11 PM
|
|
|
DanJames
Posts: 679
Joined: 12/13/2007
Status: offline
|
quote:
ORIGINAL: essentialsaltes
quote:
ORIGINAL: DanJames
Nevertheless, this analogy breaks down in light of the fact that the principle effect of a mutation is to break a protein ... But what is introduced is a broken protein that the gene pool must overcome.
Not all mutations are harmful. Not all proteins get busted when they're spelled 'wrong'. Some mutations can enhance protein function.
Even if rare, novel beneficial mutations should tend to survive, increasing the information content of the gene pool.
Reading the full text of the article that you posted has had a profound impact on my understanding of what I understand... understand? I realized that I don't have the first clue as to what is involved in blood clotting... even after looking it up on Wikipedia, I still don't feel like I understand it. Apparently this protein that is involved in this study has had been induced with a range of point mutations using a PCR method. This is a common way to find out how a protein works. You mutate it. Sequence the new mutant, see what happens in vitro and compare the results to the wild type. In this case they made the incredible discovery of an improvement to what they think is the function of the protein! Though they don't appear to know exactly how it works:
quote:
Although the actual site of membrane contact is still unknown, it can be noted that this site is adjacent to position 10, one of the three residues that appeared to regulate most of the diversity in membrane affinity among the naturally occurring proteins (sites 10, 32, and 33 (2))
They do appear to have increased the rate of a function of this protein.
I would note a couple of things. One: I understand that scenarios can be imagined where this kind of process can make an organism more suited to its environment. Though this is not necessarily one of them, as the author mentioned:
quote:
The calcium concentration under in vivo conditions is not known, and whole animal experiments may be needed to determine whether improvements in calcium binding to the protein can aid function.
It doesn't take a wild imagination to come up with a scenario where this kind of mutation can allow an organism to out-compete its peers. So, I'll stop saying that a mutation can't improve function unless someone else can tell me how this is not an improvement to a function. It does not appear to be a mutation in a regulatory gene.
Two: This has not created a new function, or a new structure. This mutation was induced in a lab, and selected for by scientists. If this is the process that is supposed to have happened in nature over the last billion years or so to turn RNA into humans, we're a far cry from that place. We assume that the Human Protein C started off as less efficient, and a mutation was induced to make it more efficient, but this same process will not produce the Human Protein C itself, and certainly not the mind-bogglingly complex process of blood clotting in general.
|
|
|
|
|
RE: Evolution and Information - 9/5/2008 5:47:21 PM
|
|
|
essentialsaltes
Posts: 1024
Joined: 10/14/2007
From: Inglewood, CA
Status: offline
|
quote:
ORIGINAL: DanJames
Reading the full text of the article that you posted has had a profound impact on my understanding of what I understand...
Again I appreciate your willingness to check this out to your satisfaction. I made it through the abstract, but that was about as far as I was likely to understand things.
quote:
In this case they made the incredible discovery of an improvement to what they think is the function of the protein! Though they don't appear to know exactly how it works:
Actually, I quite like the detail that they don't know how it works. It demonstrates that this improvement in function was not due to intelligent design. Certainly human agency was required to cause these mutations, but they were not shooting for a design that they knew beforehand. This is the kind of unintelligent design that evolution produces.
quote:
So, I'll stop saying that a mutation can't improve function unless someone else can tell me how this is not an improvement to a function. It does not appear to be a mutation in a regulatory gene.
Two: This has not created a new function, or a new structure. This mutation was induced in a lab, and selected for by scientists. If this is the process that is supposed to have happened in nature over the last billion years or so to turn RNA into humans, we're a far cry from that place.
Understood, but at least we're getting somewhere. Microevolution is possible. Current alleles are not necessarily optimal. Mutations in coding sequences (rather than regulatory sequences) can improve function.
So we have an 'engine' that provides evolutionary pressure. The question is now, "How far can it push?"
Can it teach an old protein new tricks?
I'm scraping the bottom of my biological knowledge, so I probably run the risk of sounding like one of those 'just so stories'. But a gene for protein A can get doubled, and the two genes can evolve separately. Even if one copy gets 'turned off' and/or 'busted,' random mutations still act on the sequence, and it might spit up something useful. No one says unintelligent design is efficient, but if you have millions of years, the job can get done.
As an example (just completely forget my just so story if it offends you), researchers found that 7 mutations in a gene for a mineralocorticoid receptor, convert it into a gene for glucocorticoid receptor, important in vertebrates for cortisol/stress response.
OK, ok, so it goes from one receptor to a different receptor, not a huge change in function. But this is definitely a change in function, not an improvement in an existing function. So our engine maybe can push a little further than we just established.
Can it push from goo to you (given billions of years)?
Now I'm definitely out of my depth. All I can say is that the fossil record demonstrates to my satisfaction that living creatures have changed-over-time. Evolutionary theory offers the best scientific explanation.
_____________________________
"My object in all arguments is not to make any preconceived opinion of mine seem right, but merely to discover and establish the truth, whatever the truth may be."
-- HP Lovecraft, letter to Robert E. Howard 7/27-28/34
|
|
|
|
|
RE: Evolution and Information - 9/6/2008 3:38:03 PM
|
| | | |
Forum Tools
Printable Version
