bob b, Re: 'Increasing Genetic Information'

[Johnny]

Whether a single mutation causes a cascade of changes or not does not give you the required new system.

I'm not sure what you mean by "required new system".

In fact, if you have a cascade of changes you need more control to insure that the changes made don't remove a required function, and you also need more positive changes to account for a larger dynamic system

Indeed. Which is why multiple copies of sequences is relevant.

As to your second and third statements, you may be on to something. Can you tell me why mutiple copies and mutiple identical (or nearly identical functions) has any relevance?

Yes, it means that new functions can evolve without destroying old ones.

So what... novel proteins and novel protein interaction is only a way of digging your grave. I've never denied it.

To clarify, when I refer to novel proteins with novel function, I'm referring to new proteins which are functional. Novel interactions mean new, functional interactions.

Ah, yes, great argument: "obviously evolution is true; we're here!"

Eh, my argument is that it can be observed in a lab. Not that it's true because we're here.

 

[Greenrage]

First, it is suspected (and has been observed) that proteins often evolve from frame-shift mutations of internally repetitious sequences. Second, many proteins have multpile copies on the genome. Third, many proteins have different isoforms which function identically or almost identically.

Here's the bottom line: People can make whatever claims about how unlikely it is for a beneficial mutation to happen or for a system to evolve, but it can be demonstrated in a laboratory setting that novel protein and novel protein interactions do evolve. It's like someone arguing over the statistical improbability of cars working ignoring the 500 cars that pass his window daily. Seriously, examples of protein evolution is so well studied and documented that it takes an exceptionally important example to even be published in a big name journal (science, nature, journal of evolutionary biology) these days. You'd have 10,000 unimpressed and unsurpised biologists asking why it is significant.

Best post of the month! Thanks for rebutting the ID/creationists doubletalk.

 

[Yorzhik]

I'm not sure what you mean by "required new system".

The mutation creates something new that enhances the survivability of the organism. Whatever that something is that enhances survivability will require that the system be in place in the DNA. Thus, a "required new system".

Indeed. Which is why multiple copies of sequences is relevant.

Good luck with that. You seem to think if a program has multiple copies of the same code that the redundant code will automatically be referenced if the first set being used stops working. That can only happen if the system is set up before hand to handle such. Thus, if it does happen it would be a break down of the system, not an enhancement.

Yes, it means that new functions can evolve without destroying old ones.

ditto

To clarify, when I refer to novel proteins with novel function, I'm referring to new proteins which are functional. Novel interactions mean new, functional interactions.

I thought so. But this is similar to ThePhy's white hare (I think defending evo is causing him stress). You can set up a test to design a positive change, or you can find a positive change in nature, but if you try to see if mutations will turn one body type into another body type, you find the odds will be that things break before they create new types.

Eh, my argument is that it can be observed in a lab. Not that it's true because we're here.

No; this quote: "It's like someone arguing over the statistical improbability of cars working ignoring the 500 cars that pass his window daily." - says evolution is true because we are here. It's so simplistic that the analogy you are drawing will only fit that argument. Now if you were to say "It's like arguing the statistical improbability of cars turning into [working] submarines when there are hundreds of submarines that come from cars sucked into tornadoes every year." then your quote would be relevant. I'm going to guess that you think you're making a point when someone says "evolution is just a theory" you reply "gravity is just a theory too".

 

[Yorzhik]

Best post of the month! Thanks for rebutting the ID/creationists doubletalk.

It says more about you than Johnny that you think it was a good post.

 

[Johnny]

The mutation creates something new that enhances the survivability of the organism. Whatever that something is that enhances survivability will require that the system be in place in the DNA.

I'm still not clear on what you mean. Here's an example: the Na/K+ ATPase in the nephron is essential for maintaining a gradient that ultimately ensures proper renal function. A point mutation causes the Na/K+ ATPase to bind sodium 100 fold stronger intracellularly and 1000 fold weaker in the interstitium. Ultimately, this facilitates better renal handling of sodium and thus better regulation of plasma osmolality. This confers a survival advantage in a number of ways. What system must be in place in the DNA?

You seem to think if a program has multiple copies of the same code that the redundant code will automatically be referenced if the first set being used stops working. That can only happen if the system is set up before hand to handle such. Thus, if it does happen it would be a break down of the system, not an enhancement.

Perhaps you should familiarize yourself with how transcription actually takes place before you attempt to address the topic. If there are duplicate copies of a gene on the genome, they do not sit quietly in the corner. They have the same promoter and enhancers, and they are activated by the same cellular signals. If you change one, then it simply means the other copies simply have a greater chance of being activated. Imagine four identical genes which all have the potential to be induced by a steroid hormone. Imagine that a single steroid hormone molecule enters the cell. It could activate any one of the four genes it happens to interact with. Now take away one of the four. No reprogramming or preprogramming of any sort is required to ensure that the genes can still be activated. It simply means that the steroid hormone now has three potential interactions instead of four.

You can set up a test to design a positive change, or you can find a positive change in nature, but if you try to see if mutations will turn one body type into another body type, you find the odds will be that things break before they create new types.

Which happens to be the basis for natural selection.

No; this quote: "It's like someone arguing over the statistical improbability of cars working ignoring the 500 cars that pass his window daily." - says evolution is true because we are here.

No, it says that evolution on the molecular level is observable and thus it is absurd to speculate over the relative impossibility of it occuring naturally.

I'm going to guess that you think you're making a point when someone says "evolution is just a theory" you reply "gravity is just a theory too".

It becomes necessary to remind a creationists what a theory actually is, lest they forget in their unfamiliarity with science.

 

[SUTG]

I'm going to guess that you think you're making a point when someone says "evolution is just a theory" you reply "gravity is just a theory too".

If someone says "evolution is just a theory", they need to be brought up to speed on what that entails. So, pointing out that gravity is "just a theory" is the standard response. So, yeah, I think that is making a point.

Do you think the "evolution is just a theory" argument has merit?!

Is there anything that could convince you that the Theory of Evolution is correct?

 

[Greenrage]

. I'm going to guess that you think you're making a point when someone says "evolution is just a theory" you reply "gravity is just a theory too".

Creationists seem to think that calling a scientific theory a theory somehow is an insult. Of course ET is a theory, one that has been tested for about 150 years and during that time not one observable fact contradicts it. That means it is very likely a valid theory. I know, I know, you're going to say, but you can't proof it. And of course the answer is, you can't prove any scientific theory, and nobody is trying. Scientific theories aren't provable (though they can easily be disproven by the discovery of a fact contradicting the theory, requiring abandonment or modification), but they are testable.

Tomorrow a fact could be discover that contradicts ET. But don't hold your breath. Given the volume of evidence that supports ET, it is unlikely that it is not valid.

 

[Yorzhik]

I'm still not clear on what you mean. Here's an example: the Na/K+ ATPase in the nephron is essential for maintaining a gradient that ultimately ensures proper renal function. A point mutation causes the Na/K+ ATPase to bind sodium 100 fold stronger intracellularly and 1000 fold weaker in the interstitium. Ultimately, this facilitates better renal handling of sodium and thus better regulation of plasma osmolality. This confers a survival advantage in a number of ways. What system must be in place in the DNA?

Or another example would be the possible mutation that causes white hair in snowbound rabbits. Both are still undirected changes that when added up (despite some like these) tend to break down the DNA. This doesn't result in required new systems that it would take to make one body type into another body type.

Perhaps you should familiarize yourself with how transcription actually takes place before you attempt to address the topic. If there are duplicate copies of a gene on the genome, they do not sit quietly in the corner. They have the same promoter and enhancers, and they are activated by the same cellular signals. If you change one, then it simply means the other copies simply have a greater chance of being activated. Imagine four identical genes which all have the potential to be induced by a steroid hormone. Imagine that a single steroid hormone molecule enters the cell. It could activate any one of the four genes it happens to interact with. Now take away one of the four. No reprogramming or preprogramming of any sort is required to ensure that the genes can still be activated. It simply means that the steroid hormone now has three potential interactions instead of four.

It doesn't matter. Either you've lost one of your copies (a breakdown). Or useless copies keep being made (more breakdown). Or more than one copy activates (again, breakdown). And are you saying that every copy that we can find will be a redundant part in case the one before it fails? What I'd like to know is the ratio of newly created copies that will act as a backup to those that don't. And then next we need to know if how often this copying keeps going on until it starts to become a burden instead of a help.

You can set up a test to design a positive change, or you can find a positive change in nature, but if you try to see if mutations will turn one body type into another body type, you find the odds will be that things break before they create new types.

Which happens to be the basis for natural selection.

Destroyed DNA is the basis for natural selection? I don't think you've thought that through.

No, it says that evolution on the molecular level is observable and thus it is absurd to speculate over the relative impossibility of it occuring naturally.

It's absurd to believe that mutations, even some good ones, prove that evolution from protocell to human cell occurred? That isn't absurd. I'm only going as far as the science allows. You are going from "see, mutations exist and we can imagine a protocell" to "humans exist therefore evolution went from protocell to human". It's just proving evolution because we are here.

It becomes necessary to remind a creationists what a theory actually is, lest they forget in their unfamiliarity with science.

And which theory of gravity do you adhere to? How about most scientists? Do you know how many theories of gravity are still considered strong possibilities?

 

[Yorzhik]

If someone says "evolution is just a theory", they need to be brought up to speed on what that entails. So, pointing out that gravity is "just a theory" is the standard response. So, yeah, I think that is making a point.

Do you think the "evolution is just a theory" argument has merit?!

Is there anything that could convince you that the Theory of Evolution is correct?

I guess I'll ask you the same questions: And which theory of gravity do you adhere to? How about most scientists? Do you know how many theories of gravity are still considered strong possibilities?

 

[Yorzhik]

Creationists seem to think that calling a scientific theory a theory somehow is an insult. Of course ET is a theory, one that has been tested for about 150 years and during that time not one observable fact contradicts it. That means it is very likely a valid theory. I know, I know, you're going to say, but you can't proof it. And of course the answer is, you can't prove any scientific theory, and nobody is trying. Scientific theories aren't provable (though they can easily be disproven by the discovery of a fact contradicting the theory, requiring abandonment or modification), but they are testable.

Tomorrow a fact could be discover that contradicts ET. But don't hold your breath. Given the volume of evidence that supports ET, it is unlikely that it is not valid.

And for Greenrage something I hope is simpler: do a search and see if you can find out (generally) how many theories of gravity are currently the favorite possibilities. We only need the number. We don't need you trying to confuse yourself with explanations.

 

[Johnny]

Before we continue, can you summarize your understanding of cellular replication, DNA reproduction, and DNA transcription and translation? Just a short paragraph will suffice. Rereading one of your older posts I kind of felt like you were missing something. You said, "The odds are against the idea that a pure strain will survive generation through generation, stacking good mutation upon good mutation in their DNA."

Both are still undirected changes that when added up (despite some like these) tend to break down the DNA.

How? You haven't explained this.

This doesn't result in required new systems that it would take to make one body type into another body type.

Can you give me an example of a "new system" that would be required to make one body type into another body type?

It doesn't matter.

No, it does matter. You said, "Because not only will a good mutation have to be changed in whatever point it did change, but it will have to also come with changes (or wait for changes) to other systems that are dependent on that other single change." I provided a mechanism by which a mutation would not have to come with changes or wait for changes in other systems dependent on that single change.

Either you've lost one of your copies (a breakdown). Or useless copies keep being made (more breakdown). Or more than one copy activates (again, breakdown).

It appears you're defining any change away from the "starting" genotype as a "breakdown". In that case, I can breakdown all the way from a monkey to a man and your definition of breakdown (whatever it may be) becomes meaningless.

And are you saying that every copy that we can find will be a redundant part in case the one before it fails?

No, I don't think you understood what I was saying. You can't classify multiple copies as redundant or primary--they are functional equivalents. It's like having four copy machines in a room that do the same thing, and each time you need one you pick one randomly. You can't classify one as a primary machine and the rest as secondary or tertiary backups--they are all primary machines. If you knock out one, you haven't lost the ability to make copies. Which means that your clients who rely on that copy still get their papers.

Destroyed DNA is the basis for natural selection? I don't think you've thought that through.

Mutations are either good or bad. The bad ones get eliminated, the good ones live on. Thus, bad mutations are just as important for natural selection as good mutations.

It's absurd to believe that mutations, even some good ones, prove that evolution from protocell to human cell occurred?

Mutations do not prove evolution from a protocell, they provide a mechanism.

You are going from "see, mutations exist and we can imagine a protocell" to "humans exist therefore evolution went from protocell to human". It's just proving evolution because we are here.

I am directly telling you right now that I am making no such assertion. The assertion I was trying to make was that it's absurd to speculate over the impossibility of a good mutation happening when it happens all the time.

And which theory of gravity do you adhere to?

Einstein's theory.

How about most scientists?

Einstein's theory.

Do you know how many theories of gravity are still considered strong possibilities?

Quantum theory is probably the only viable competitor at the moment; but there is no empirical support for the graviton.

When a creationist says evolution is only a theory, what he's really saying (much to his dismay) is "evolution is a logically self-consistent description of natural phenomena which is supported by experimental evidence and is predictive, logical, and testable." Wikipedia says, "A [scientific] theory is a logically self-consistent model or framework for describing the behavior of a related set of natural or social phenomena. It originates from and/or is supported by experimental evidence (see scientific method). In this sense, a theory is a systematic and formalized expression of all previous observations that is predictive, logical and testable. In principle, scientific theories are always tentative, and subject to corrections or inclusion in a yet wider theory."

 

[sentientsynth]

Is there anything that could convince you that the Theory of Evolution is correct?

I'd like to see experimentation exploring the boundaries of DNA modification. Is it even possible to make gradual changes to a dog until it turns into a horse (or something, anything remotely similar) ?

I'm not talking about modifying yeast so that it won't produce ethanol. I mean big, bold, dirty changes. Something any hick on the street like me would recognize as a clear, trans-species change.

I just want to see if it's even possible.

Ya catch my drift?

 

[Yorzhik]

SS, go for body type change - that's pretty clear. Hey, let's try reptile to bird! that one they say has already been done.

 

[Johnny]

I'd like to see experimentation exploring the boundaries of DNA modification. Is it even possible to make gradual changes to a dog until it turns into a horse (or something, anything remotely similar) ?

If nucleotide arrangement dictates morphology, then why should it not be possible?

 

[sentientsynth]

If nucleotide arrangement dictates morphology, then why should it not be possible?

That's my bone of contention. Get your own arguments, Johnny. :chuckle:

It seems like once we have a complete genome of a species, we could compare it to that of a "relative" species and see how we could make slight changes to convert the one to the other. The question, then, would be just how slight can these modifications be?

A single nucleotide switch sometimes does nothing, as multiple codons can translate into a common amino acid, and hence an identical protein. So...how many amino acids do we have to modify at once? One? Fifty? Two thousand?

Frankly, I have no clue.

These sorts of questions are best worked out in a laboratory environment.

So...

Why haven't we seen experiments with this as its guiding principle?

Have there been any? You're better at juicing PubMed than I am, Johnny.

 

[noguru]

Selective breeding to create a horse from a canine is highly unlikely, given the current state of affairs. The processess that produced a horse are not entirely clear, and any husbandry change from canine to a horse would follow a different path than the one followed by nature. What would result would be what we call convergent evolution. Some animals like canivores are so distinct from hooved herbivores like horses that turning one into the other is virtually impossible. There common ancestor is so far in the past that we cannot expect their characteristics to be readily interchangeable. You have now raised the bar so high that it cannot possibly be reached. What we need to look at is the most plausible explanation for what we do see. We know that there are small changes in DNA that occurr. We know the physical mechanisms that are responsible. Can you point to a physical mechanism that would inhibit the build up of small DNA changes from becoming large DNA changes that would seperate horses from canines? If you cannot, then logic would dictate that small changes can and do build up and that they lead to the biodiversity we see today.

 

[noguru]

SS, go for body type change - that's pretty clear. Hey, let's try reptile to bird! that one they say has already been done.

Here again you have the same situation as trying to turn a dog into a horse. Birds and modern reptiles share common ancestry. But their common ancestor is so far in the past that it would be impossible to turn one into the other, given our current knowledge. However, some good evidence that they share a common ancestor can be gained from research into genetics. Here is an article that may help you understand.

Scales and feathers

 

[sentientsynth]

Selective breeding to create a horse from a canine is highly unlikely, given the current state of affairs. The processess that produced a horse are not entirely clear, and any husbandry change from canine to a horse would follow a different path than the one followed by nature.

I think I see what you're saying. You're saying that is wasn't a dog that became a horse. There was a common ancestor to both the dog and the horse which they both evolved from. Going from today's dog to today's horse is a different affair. It's a legitimate distinction, I think. But still one that may be reproducible in a lab.

Thing is, you could hop straight from a dog to a horse with, say, several tens of thousands of specific base substitutions. My question is, just how much can we tinker with existing DNA? How much can we change it? How little can we change and get away with it? Are there different cut-offs for different species?

Exciting questions, to me. But I keep reminding myself that this is the sort of thing I've always seen in those cheesy 80's science B-movies. Well, Island of Dr. Moreau was pretty good. Freaky, though. And exactly like what I'm talking about doing. Well, minus the human factor.

Anyway, back to the subject. I wouldn't even demand that it be as big, bold, and nasty as a dog to a horse. How about one species of kingdom Archeabacteria to another within the same kingdom? Swap things that are very similar genomics-wise. Then slowly increase the species differential. How far can you go, making whatever slight modifications are necessary? Which would lead us to the final question, How slight or how large does a modification have to be for a given species, for a given phenomic modification?

What would result would be what we call convergent evolution. Some animals like canivores are so distinct from hooved herbivores like horses that turning one into the other is virtually impossible.

I love it when people say things are impossible with science.

There common ancestor is so far in the past that we cannot expect their characteristics to be readily interchangeable. You have now raised the bar so high that it cannot possibly be reached.

Then let's lower it. What you said sounded reasonable. But...just how high is the bar for evolution? That's what I want to get at. Not to find out how we got here, but if it's even possible. A pure science endeavor.

What we need to look at is the most plausible explanation for what we do see.

Occam's razor has its exceptions. But I'm sure you know that.

We know that there are small changes in DNA that occurr. We know the physical mechanisms that are responsible. Can you point to a physical mechanism that would inhibit the build up of small DNA changes from becoming large DNA changes that would seperate horses from canines?

That's pretty much my point. But I'm not satisfied with an argument from ignorance. I want data. Something. Anything that answers my questions.

If you cannot, then logic would dictate that small changes can and do build up and that they lead to the biodiversity we see today.

Well, believe that if you like. But if you were arguing the mechanism of a certain, unknown chemical reaction and you presented argumentation analogous to what you've shown here, I'd be forced to kick you off my staff. Sorry. I've been accused of mixing up the scientific method and formal logic. The charge may hold true.

 

[Johnny]

It seems like once we have a complete genome of a species, we could compare it to that of a "relative" species and see how we could make slight changes to convert the one to the other.

Indeed, and this is probably being done to some limited extent. I would imagine the problem with doing this in more advanced organisms is primarily analyzing and comparing genetic information. We have the ability to do the required genomic manipulation, but the analysis of the genomic changes--even with a completely mapped genome--would still be tricky. One would have to account for neutral mutations, regulatory changes, changing genetic domains, etc. etc. For example, birds seem to have lost their teeth not because of a mutation or change in the gene that codes for teeth, but instead by a shift in the genetic domain during the early embryological stages. The gene that codes for teeth is still present (at least in chickens)--it's just masked by other genetic changes. Thus, if we were to compare the genome of a dinosaur (assuming we had one) and a chicken, we may just find that the gene that codes for teeth in both organisms is remarkably similar. So it's not as easy as comparing gene-for-gene changes, we'd have to look at the big picture. We're still working on the "big picture". As we move up along the chain of life, the interactions which control gross morphology become horribly tangled. We're still trying to figure why humans are the way they are, much less other species. Maybe by the time you finish your PhD this field will be picking up.

Nonetheless, we are beginning to see research like this emerging on a very simple scale. Last month a fascinating paper was published in the journal Science titled "Evolution of hormone-receptor complexity by molecular exploitation". Basically scientists recreated an ancestral gene (a hormone receptor) based on phylogenetic analysis of modern genes. Then they analyzed the sequences of modern organisms (through the phylogenetic tree) and retracted the mutations that occured. The interesting thing with this paper is that they actually expressed the proteins and showed that the receptor specificity increased as predicted. However, it was extremely simplistic and they only traced two specific mutations. The beauty of this experiment was that it put phylogenetic analysis and directed mutagenesis together to reconstruct historical genes and then compare them to creatures along the phylogenetic tree. It's fascinating, but it also shows just how little we know.

I have to run to class, I haven't had time to reread my post so it's just a collection of thoughts thrown together real quick. I'll try and see if I squeeze something out of pubmed later

 

[Johnny]

It seems like once we have a complete genome of a species, we could compare it to that of a "relative" species and see how we could make slight changes to convert the one to the other.

Indeed, and this is probably being done to some limited extent. I would imagine the problem with doing this in more advanced organisms is primarily analyzing and comparing genetic information. We have the ability to do the required genomic manipulation, but the analysis of the genomic changes--even with a completely mapped genome--would still be tricky. One would have to account for neutral mutations, regulatory changes, changing genetic domains, etc. etc. For example, birds seem to have lost their teeth not because of a mutation or change in the gene that codes for teeth, but instead by a shift in the genetic domain during the early embryological stages. The gene that codes for teeth is still present (at least in chickens)--it's just masked by other genetic changes. Thus, if we were to compare the genome of a dinosaur (assuming we had one) and a chicken, we may just find that the gene that codes for teeth in both organisms is remarkably similar. So it's not as easy as comparing gene-for-gene changes, we'd have to look at the big picture. We're still working on the "big picture". As we move up along the chain of life, the interactions which control gross morphology become horribly tangled. We're still trying to figure why humans are the way they are, much less other species. Maybe by the time you finish your PhD this field will be picking up.

Nonetheless, we are beginning to see research like this emerging on a very simple scale. Last month a fascinating paper was published in the journal Science titled "Evolution of hormone-receptor complexity by molecular exploitation". Basically scientists recreated an ancestral gene (a hormone receptor) based on phylogenetic analysis of modern genes. Then they analyzed the sequences of modern organisms (through the phylogenetic tree) and retracted the mutations that occured. The interesting thing with this paper is that they actually expressed the proteins and showed that the receptor specificity increased as predicted. However, it was extremely simplistic and they only traced two specific mutations. The beauty of this experiment was that it put phylogenetic analysis and directed mutagenesis together to reconstruct historical genes and then compare them to creatures along the phylogenetic tree. It's fascinating, but it also shows that we're still in the very early stages of figuring this stuff out.

I have to run to class, I haven't had time to reread my post so it's just a collection of thoughts thrown together real quick. I'll try and see if I squeeze something out of pubmed later