New Genetic Information?

[jon1101]

In my circle of friends I often defend evolution as more reasonable than often painted by creationists, not because I am an evolutionist but because the rampant fallacies of the creationist movement unnerve me. After refuting countless arguments regarding second laws, moon dust, paluxy tracks and whatnot, I'm left primarily with the 'no new genetic information argument.' When inquired of the mechanism for the creation of new genetic information I typically respond 'mutations,' which in turn evokes the classic 'there are no beneficial mutations' argument.

Now, I have no data to confirm either position, but I suspect that the ever enigmatic 'beneficial mutations' do in fact exist. Or, I could simply be wrong in my answer. I'm not terribly erudite regarding evolution, so it wouldn't surprise me. If one of you evolutionist type people would enlighten me as to the source(s) of new genetic information, I would appreciate it.

-jon

 

[Morat]

  Mutations. *shrug*. The problem comes with the concept of information. It's a fun word to throw around, it's very slippery, and it's hard to pin down.

   Sort of like "entropy" before people started getting serious about it.

  Information is defineable, and can be calculated.

  So, if they throw that at you, the first thing you need to do is ensure they know what information is.

  Ask them which of the following (if any) would be an increase in information:

1) The appearance of a new allele.

2) Doubling the length of a gene.

3) Doubling the length of gene, then changing a few base pairs in the new section.

4) The appearance of a new biochemical pathway.

 

   By any common (like Shannon) definition of information, all four are "new information" and all four happen.

 

 

[seebs]

CTI: If you'd like to have a serious discussion on what we mean when we talk about "information", I'd be happy to try to discuss this. It's one of the few parts of the evolution debate that is *actually in my fields*.

 

[jon1101]

Morat: are these four examples you gave beneficial to the given organism? Do they increase the organism's complexity? Any documented examples of your answers and the corrisponding link would be greatly appreciated.

-jon

 

[seebs]

The point of his examples is that they "add information" in a technical sense, *whether or not* they are beneficial.

These are two separate questions. Imagine that we are creating sentences. It's pretty clear that a longer sentence probably has more information, whether or not it's a "better" sentence in any way. Similarly, sometimes, a *shorter* sentence may be "better".

Note that "beneficial" is not something you can judge in a vacuum. It's at least partially a function of context and environment. Look at the famous example of moths changing color to match trees; it's not that "black" or "white" is a better color in a universal sense; it's that it's an adaptation to a specific environment.

Arguably, the change from "white with flecks" to "black" counts as an increase in information. (If you want to debate that one, we can, and I'm not sure it'd be a winnable point, but it's a debatable one at least.)

 

[jon1101]

Originally posted by seebs
CTI: If you'd like to have a serious discussion on what we mean when we talk about "information", I'd be happy to try to discuss this. It's one of the few parts of the evolution debate that is *actually in my fields*.

I would, thanks.

-jon

 

[seebs]

Okay, this'll necessarily be a bit brief, 'cuz I'm at work, and I have to get back to work shortly. I'll try to get more time for this tonight.

The question of "what is information" is not as easy as it sounds. Let's imagine that we agree that "red", "black", and "white" are all the same amount of "information" when describing the color of a car. (In fact, this is probably untrue, but I'll get back to that.)

Now, let's say we're discussing the number of cars of different types that have different colors, and we discover that one model of car sells many more red cars than black or white, and another sells mostly black cars.

The *populations* contain new information! We know something about the selection pressures on these cars, even though no one car contains this information. Neat, huh?

Now, let's look at the gradual development of a genetic line over time. I think we can agree that simply having the same genes forever is not a change in the amount of information.

So, let's look at changes. Several kinds of change could happen:

1. We could change one base pair to another.
2. We could omit a base pair.
3. We could add an extra base pair.
3a. It could be a duplicate of some existing material.
3b. It could just be a new chunk.

So, starting from ATCG as the pool of base pairs, imagine an organism which has the gene "AAAA". Changes from this could be
AAAT
AAA
AAAAA
AAAAT

From an information theory perspective, all of these but the second are definitely adding new information.

Now, let's look at a more interesting case.

AATTAATTAATTAATTAATT

Let's say we lose a single pair:
AATTAATTAATAATTAATT

In fact, this sequence is now *MORE COMPLICATED* than the longer one - we have *ADDED* information!

Cool, huh?

So... "added information" is very hard to define. If you want a way to see what it looks like in practice, pick a word, and deal out that many scrabble tiles. Now, take away every tile that wasn't part of the word. Now, deal out tiles to fill those empty slots.

Although, in theory, you're only adding random letters, you will find that you eventually get your word. Where did the information come from? Selection pressure.

 

[jon1101]

The point of his examples is that they "add information" in a technical sense, *whether or not* they are beneficial.

So basically, the aforementioned processes add information which may or may not be beneficial, and then natural selection does its thing to further the evolution of the given organism?

Note that "beneficial" is not something you can judge in a vacuum. It's at least partially a function of context and environment. Look at the famous example of moths changing color to match trees; it's not that "black" or "white" is a better color in a universal sense; it's that it's an adaptation to a specific environment.

That makes sense, but I've always been told that the peppered moth experiment was an evil evolutionist conspiracy that involved gluing moths to trees and other such deceitful practices. Were they wrong?

-jon

 

[seebs]

Originally posted by Cancer To Iniquity

So basically, the aforementioned processes add information which may or may not be beneficial, and then natural selection does its thing to further the evolution of the given organism?

Sort of. The information comes in from the combination of random alterations and selections.

If I shuffle some cards (just one standard "riffle"), deal them out, then pick them up again, I'm probably not adding any information.

If I deal some cards out, then pick them up in a standard order, it's quite likely that I just keep them in the same order forever.

if I shuffle, then deal, then pick the cards up with some preference order (say, lowest cards first), after a while, even with the randomizing, the deck will tend to be in lowest->highest order.

New information!

That makes sense, but I've always been told that the peppered moth experiment was an evil evolutionist conspiracy that involved gluing moths to trees and other such deceitful practices. Were they wrong?

I believe it wasn't originally an experiment, just an observation. It might well have been false in that particular case, but no problem, we've got tons of variants. Take a bacterium that generally thrives around 70 degrees. Now, make two cultures, give them plenty of food, and gradually chill one to 60 or even 50 ("gradually" == "thousands of generations of time", which can be weeks for bacteria) degrees, and gradually heat the other to 80 or even 90.

If you go slowly enough, you will probably end up with one population that thrives at 50, but doesn't do well at 70, and dies at 90, and another that thrives at 90, but doesn't do well at 70, and dies at 50.

New information!

 

[jon1101]

Okay, I think I understand how the new information comes about, but I'm not entirely sure how the complexity of the being is increased. In the bacterium example the new information led to a change in temperature tolerance, but I don't understand how this advances the complexity of the organism. Is a bacterium that thrives in 90 degree water more complex than one thriving in 70 degree water?

The typical creationist response is that this process is incapable of advancing a species. Rather, that the genes that allow for surviving in hot water already existed in some of the bacteria, as created by God, and the experiment just kills off all of those bacteria that cannot survive in the hotter water, leaving only those who can. The remaining bacteria then proceed to reproduce and thus there is a change in the population's heat tolerance. Therefore, there is actually a loss of genetic information - a pruning of the gene pool if you will.

For example, if you have a bacterium with a gene of TtTt that thrives in 70 degree water, and that bacterium stays in such water and reproduces you get some TtTt bacteria, some TTTT, tttt, etc. The latter would perhaps thrive in 50degree water and thus die in the 70 degree water. If, however, the water was chilled to 50 degrees, the tttt bacteria would survive and propagate. Due to the fact that there are very few 'T' genes left in the gene pool, the bacteria as a whole now has different heat tolerance.

Thus, they say that such evolution is actually 'adaptation' and cannot move the species any further than the current genetic information allows.

-jon

 

[Morat]

 See, this is where you've gone awry (actually, into the realms of "No True Scotsman"). You've taken an identified term "information" (it's nebulous to most people, but precise and measureable definitions exist) and added a second term "complexity".

  What is complexity? Well, I know what information density is. Is that complexity? I know quantities of information. Is that complexity?

  What is complexity? It's a word added to "information" to handwave away examples of informational increases. Dembski plays at defining it, but his examples seem to boil down to airy claims and "I know it when I see it".

  To continue this, you have to define complexity in a precise and measureable way.

   Otherwise, claiming something doesn't exhibit a change in complexity is simply a baseless dismissal.

 

 

[jon1101]

Originally posted by Morat
 See, this is where you've gone awry (actually, into the realms of "No True Scotsman"). You've taken an identified term "information" (it's nebulous to most people, but precise and measureable definitions exist) and added a second term "complexity".

  What is complexity? Well, I know what information density is. Is that complexity? I know quantities of information. Is that complexity?

  What is complexity? It's a word added to "information" to handwave away examples of informational increases. Dembski plays at defining it, but his examples seem to boil down to airy claims and "I know it when I see it".

  To continue this, you have to define complexity in a precise and measureable way.

   Otherwise, claiming something doesn't exhibit a change in complexity is simply a baseless dismissal.

 

Okay, so the complexity argument is facile. Am I to then assume that the example I gave also involves the adding of information, and that creationists deny this by using faulty definitions of the word 'information?'

-jon

 

[seebs]

It's really hard to define what we mean by "adding new information". We tend to point to simple examples, because they're easier, but it's in the complicated ones where nothing seems to be happening at first that you really get more information.

As I understand it, a typical way for "new information" to get added is that, first, a replication error produces some extra/unused DNA, and later, some of that DNA changes in a way that makes it useful.

As an example, let's say we're talking about the part of the genetic sequence that's making sure you end up with the right number of vertebrae. If it screws up, you can end up with partial extra ribs (IANMTU; a friend of mine has an extra rib on one side). If it screwed up in a slightly luckier way, you might just have a couple of extra ribs... and your kids might end up with that trait, too. If that's useful, eventually you have people with more ribs. "New information"? Maybe, maybe not; I would say it is.

 

[Morat]

  The example you gave was one of evolution. The frequency of alleles in a population changed. That's all evolution is. Seebs was using this to point out how "beneficial" was something that depended on context (the enviroment).

   The part that you're curious about is the source of new alleles. As we pointed out, mutation handles this easily.

 

[Morat]

  Speaking of ribs (*Warning Off Topic*) the number of ribs in a given species tends to be far more an "average" than anything else. On average, humans have X rib pairs, but sometimes they have more..or less. Which is why the "horse evolution/rib number" argument always cracks me up.