An Increase in Information

[kewltony]

1337 views

[MrKrov]

Nylonase: frameshift mutation. Science: it works, bitches!

[EvoBiologist]

Gene duplication followed by mutation of the duplicated gene. Voila! New information! PLENTY of peer reviewed papers showing that it happens all the time. All genes form gene families, providing evidence that most information in DNA is made this way.

[spark300c]

how ever Dr Standford wrote in his book that gene duplication reduces expression of the gene. Also gene regulation is effected. the rate of gene duplication, as far as I know that beneficial gene duplications extremely rare. Even more rare than non gene duplication beneficial mutation. the size of genome. The section pressure. These topics have to be address.

[EvoBiologist]

When genes are duplicated, they make copies of the protein nearly twice as fast. This means the product of any reaction catalyzed by that protein is made almost twice as quickly. The product then triggers the downregulation of expression of the gene. However, if the copy mutates enough, it will catalyze a different reaction, and thus gene regulation will normalize to it's original one-copy rate.

Examine any gene family and you will see the signals of common descent and new information.

[gamesbok]

'Evolution of Biological Information'

Dr Thomas D. Schneider

Neuclaic Acid Research

2000, Vol 28, n0 14.

[tubewatch59]

From my creationist viewpoint, I can see that the system has evolved towards a desired state under selctive amplification of some kind. You said this was an increase in information. But this information was imposed onto the system, randomness allows the system to be 'mapped' in a kind of preliminary way, and iteratively the system is driven to amplify the desired behaviour. Wouldn't you agree that the information is being imposed from the outside in? Coming 'from the experiment', so to speak?

[Impaler1815]

How could the information come 'from the experiment'? It's not like the scientists were personally substituting each individual nucleotide. The information still came from random mutations.

[tubewatch59]

The information is in the selection criteria. It's the tight link between what is wanted and what is actively selected for. In this way the system is driven in most cases to the desired area. It's evolution, but not quite as natural selection knows it. There the only criteria for selection is relative survival. That can be the result of many many traits (and accidents) all in competition simulataneously to share the one selectable criteria - survival. A different thing than artifical selection.

[Impaler1815]

Nice smoke screen. Whether the mutations were artificially or naturally selected is irrelevant. It's the creationist position that no random mutation can increase information, regardless of what type of selection they're under. This video refutes that position. An increase in catalytic activity would be selected for in nature too, seeing how that's obviously benifical to the organism.

[tubewatch59]

Call me enlightened then. I don't dispute it's theoretically possible. If a functional genetic sequence might also be generated by random mutations - then it can happen. In fact it has been observed to happen.

The relevant questions are: To what extent does it happen, and how long does it take? Could random mutations and selection build biomachines, bacteria, plants and animals in billions of years?

What we observe on the low end of complexity doesn't scale well to the upper end.

[tubewatch59]

You might say I have 'shifted the goalposts'. Fair comment. When goalposts fall over, they need to be shifted. Perhaps we'll have to keep on shifting them until we believe the same things as Ken Miller does: Theistic evolution. But at present I doubt it.

While there's enough time, mutations and population numbers to carry out some evolution of simpler systems, I can't see how it scales to larger more complex systems in the accepted time (4E^9 yrs) available. Evolution too has it's limits.

[Impaler1815]

At 140 odd mutations every generation, and 3.5 billion years of generations, leaves more than enough time to account for the variation and complexity of life on earth.

"What we observe on the low end of complexity doesn't scale well to the upper end."

There is absolutely no reason to assume that is the case with genetics. Information increasing mutations have been observed in all sorts of organisms.

[tubewatch59]

"At 140 odd mutations every generation, and 3.5 billion years of generations, leaves more than enough time to account for the variation and complexity of life on earth."

It's thought that the number of likely functions of a protein (of length N) increases proportional to log(N), whereas the number of combinations increases proportional to 20^N.  You can appreciate the huge divergence that will occur in the ratio of functional proteins : non-functional proteins.

The numbers are staggering!

[Impaler1815]

That would only be a problem if proteins had to evolve in a single step. Evolution works by accumulating mutations that are constantly selected for. The ability for natural selection and evolution to create improbable sequences has been well demonstrated.

[tubewatch59]

"There is absolutely no reason to assume that is the case with genetics. Information increasing mutations have been observed in all sorts of organisms."

The increases of information observed are trivial. And never are seen to form sophisticated multi-protein complexes.

It's because of the huge divergence, the huge numbers of ways in which proteins will not do anything useful, compared to the few ways in which they will, that prevents the evolution of anything but the simplest of systems.

[Impaler1815]

"The increases of information observed are trivial."

They're not trivial, it's just that creationists like to shift the goal posts and apply miracles to explain them away.

"And never are seen to form sophisticated multi-protein complexes."

WRONG! The nylon bug uses mutliple proteins to digest nylon.

[tubewatch59]

There seem to be a few unusual things going on with nylonase. Some speculate that it may be an example of a system that is using something built-in to enable adjustments to the environment. It's too early to say. Until more is known I will concede that it is a very fine example of evolution in action.

[Impaler1815]

There is no evidence for that at all. We have genetic sequences and we know where the mutations took place.

[tubewatch59]

"There is no evidence for that at all. We have genetic sequences and we know where the mutations took place."

There is so much currently unknown that we cannot rule that possibility out (of it being some form of built-in environmental adjustment).

However as you say, there is little evidence for that, so it remains as a mere speculation until it can be demonstrated. See. What I try to smack you with, will sometimes come back and whack me as well.

[Impaler1815]

Thankyou for being honest about it.

Not being able to rule out a possibility doesn't mean it's worthy of consideration.

[tubewatch59]

"The information still came from random mutations."

Just wanted to clarify something. The potential for information comes from the random mutations, but they are just a way to introduce new elements into the system, or to vary existing ones. The information always comes from the selection process which then selects those parts of the initially random elements that conform to the required information...

[tubewatch59]

While that works well for genetic algorithms concentrating their selection criteria on one trait, there is no real applicability to natural selection which must be capable of 'pointing the way to increasing fitness' of all the various subsystems in the organism simultaneously. It can't sort out all of the contradictory signals (including random accidents) that all get lumped into the 'differential surival' control variable.

A bit like trying to fly a jet between cities using just 2 buttons.

[tubewatch59]

Just because there actually are real life scenarios that are compatible with simple selection criteria (higher running speed of a predator is definately an advantage), it doesn't mean that such simple criteria will always suffice.

For example the fastest running predators may be likely to die of starvation first, before their slower (less muscled, more fatty) competitors.

One simple example of why natural selections job is not a straightforward path to higher overall fitness.

[noobapple]

"One simple example of why natural selections job is not a straightforward path to higher overall fitness. "

Natural selection doesn't select for overall fitness.

[tubewatch59]

"Natural selection doesn't select for overall fitness."

I think I agree with you. The only thing it can directly select for is differential survival.

[Impaler1815]

Of course there's always going to be a trade off. However, that trade off may make the organism better off in the right environment. For example, the loss of the ability to breathe underwater might put reptiles at a disadvantage to amphibians overall, but the ability to perform and survive better on land outdoes that in their environments.

[Impaler1815]

Mutations don't work on "all the various subsystems". Each mutation works on one base, which in turn effects one gene. All other things being equal, that's the mutation that will be selected for.

The worst that could happen if multiple different traits are selected for in a population, would be the population splitting.

"A bit like trying to fly a jet between cities using just 2 buttons."

...huh?

[tubewatch59]

"'A bit like trying to fly a jet between cities using just 2 buttons.' ...huh?"

Meaning that natural selection is supposed to guide the development of systems with thousands or more of selectable traits, with very few inputs.

example: Feedbacks needed to optimise:

Eyesight focus, colour, hearing, appetite, speed, stamina, blood sugar tolerance, fat storage vs. burning, bone elasticity, (and thousands of other traits)...

What does evolution do?

Selects on 'survival' every 2 decades!

[tubewatch59]

The far larger problem is that these systems must appear in the first case. And they must be built step by step even though many are irreducibly complex! Which compounds the problem more because then, we can't incrementally improve functions but must change functions around. So each new function must emerge, and must emerge without the help of natural selection because until they have emerged they don't function, and while no function takes place the emerging stage is mutating randomly!

[tubewatch59]

And so it has to find new functions by random blind search - ploughing through the huge 'deserts' of zero functioning protein (and regulatory RNA) combinations until it 'might' hit on a working combination. Which in most cases NS will select it for removal because the rare 'working combination' that was stumbled upon, did something nasty to the already existing functional machinery in place, causing a malfunction. So of course it is selected against, and the blind search begins once again!

[Impaler1815]

There are no irreducibly complex features in nature. So far all the main examples of IC presented by creationists have been shown to have valid pathways. Now you lot are just resorting to a Gish gallop, hoping to present "irreducibly complex" features faster than we can debunk them.

[tubewatch59]

"There are no irreducibly complex features in nature. So far all the main examples of IC ... have been shown to have valid pathways."

A 'valid pathway' might allow it to evolve (theoretically) but the exaptation causes improbability to come back with a vengeance. Each step in the pathway must emerge, and the emergence is a unit. It has to happen as unselected for macromutations. That's why 'emergence' is used to describe it, not 'fine tuning'. It's the price of evolution sidestepping I.C.

[Impaler1815]

"It has to happen as unselected for macromutations."

The whole idea of those pathways is that no step is unaffected by natural selection. Again, improbability is not a problem for evolution as long as there is a series of small steps leading up to a feature's current stage.

[tubewatch59]

"The whole idea of those pathways is that no step is unaffected by natural selection."

That is the idea. But you can't always get what you want.

The problem is: EACH STEP IS NOT A REFINEMENT OF A PREVIOUS FUNCTION as would be the case in a non-irreducibly complex system. Rather: EACH STEP IS A NEW FUNCTION.

Since selection only operates on existing functions, until a new function emerges at each stage, NATURAL SELECTION CAN'T HELP REFINE IT UNTIL AFTER THE NEW FUNCTION HAS EMERGED.

[tubewatch59]

So each new stage must emerge a new function without the benefit of natural selection refining the process. This is the price evolution pays for having to jump the irreducible complexity barrier. Whether you admit to the existence of irreducible complexity or not is irrelevant. Because it's admitted by Dr Miller and others that step by step emergence is how evolution nullifies Behe's irreducible complexity barrier.

The price paid is the loss of natural selection during new function emergence.

[tubewatch59]

I'm not saying that there is no natural selection at all during emergence. Natural selection continues to monitor the overall performance of the system, which is a result of the existing function, and the effects of the new functioning being built up (a series of mutations) before it has fully emerged and begun to operate. This will affect fitness in unpredictable ways, serving to either improve or weaken the existing function of the system depending on the effects of the various mutations.

[tubewatch59]

The point is, that there is no selection process guiding the emergence of the totally new function. That has to be the result of random chance.

Once the new function has emerged as the result of a series of mutations, it begins to operate and be coopted into the system (provided it can be successfully coopted). From that point on it will be either improved (or removed) by natural selection, depending on the overall effect on fitness.

But the new function has to first emerge! RANDOMLY.

[Impaler1815]

Again, these pathways require that ALL stages are functional and selected for. Even the beginnings of the evolution of each part is useful. For example, even the slightest partial clear covering over the pupil will help keep things out of the eye better than no covering at all. The only exception is when parts are co-opted from parts elsewhere in the organism, though they could still evolve step by step in their original function.

[tubewatch59]

"these pathways require that ALL stages are functional and selected for."

An example: cdk007's The Evolution of The Flagellum. @ 1:23 'Association of a second protein makes the pore selective': Lets say the protein comes from an existing one. Now it's clear that initially it doesn't associate. So, it has to (by mutations) get to a point where it is associative. Now those mutations are non selectable. Because until it is associative in a useful way, all that can happen is random trials...

[tubewatch59]

Actually it is selectable in the sense that if any of these mutations cause the second protein to cause some problems in the cell in some way, then that change may be selected against to varying degrees. However, there is no correlation between the mutations occuring to this second protein, and the (perhaps to emerge) function of association. It doesn't know about any possible association, and there's no guidance towards that goal. (There is no goal). If it happens, it's a lucky accident...

[tubewatch59]

See what I mean about emergence being random? There are certain mutational steps, such that even though natural selection is present to monitor overall cell fitness, there is still no correlation between overall cell fitness, and the mutations that must happen to get that second protein to associate. Because the association function doesn't exist - until it does! And we can't assume that the number of mutations that have to occur to cause association are just a few. Maybe so, but probably not.

[tubewatch59]

Now the next thing after a successful association, is a capability to enable selective transport across the pore. Natural selection cannot guide any of the successfully associating proteins to mutate such that they specify the right selection properties. Remember there are no goals, no right proteins, no wrong proteins. But regardless, only a very few trials will turn out to be useful. Very few will specify selective transport that's useful. There is no correlation to selection. It's all random.

[tubewatch59]

Don't take that last sentence out of context. There is selection, that monitors the overall fitness of the cell as it pertains to survivability. However that selection does not correlate to the establishment of not yet existing functions. These emergence events are not smooth or continuous. Without all the mutations in place, there will be no function in existence to select for.

So, do you agree that there are quite a few mutations that must occur - randomly - for evolution to proceed?

[tubewatch59]

Let me make another thing clear. Once the new function has emerged and activated, it will then be under selection guidance. This means that over time it can perhaps be improved upon if any beneficial mutations happen to occur.

However, before the function has emerged, there is no selection guiding the mutational emergence of the new function, except for the overall influence of natural selection which tracks cell survivability. But as I have shown, that cannot correlate to the new function.

[Impaler1815]

If the protein comes from an existing one then there are advantages in having greater specifity in binding with the pore. These advantages would be things like better assembly of the pore and a stronger structure once the protein's in place. Because of this even the first mutation that increases specificity would be selected for, because it slightly increases the chance of the protein binding.

[Impaler1815]

If the second protein is not taken from elsewhere, say if it's a duplication of one of the proteins in the pore, then it's simply a matter of the protein evolving it's selectivity. This is probably the most likely scenario. Either way, each step will have a use.

[tubewatch59]

"If the second protein is not taken from elsewhere, say if it's a duplication of one of the proteins in the pore, then it's simply a matter of the protein evolving it's selectivity."

But then you are taking a protein that normally is part of the pore and trying to jam it up under the pore to associate somehow. There is no way that's going to be any kind of a good fit! A lot of random mutations will be needed to get it to the point where it fits the bottom of the pore well enough to bind.

[Impaler1815]

Specificity can be very vague for it to still work. Citrate and nylon digestion could not have evolved if specificity had to be to as good as you claim for something to bind.

[tubewatch59]

Some protein specificities are fairly low compared to others that are much higher.

Behe discusses how the evolution of antifreeze in fish was able to occur for the reason that it isn't a system that needs high specificity.

Analogy: Sandbags clogging a narrow part of a stream can give rise to a very useful dam of water. But such a simple structure can't be compared in specificity to a windmill driving a pump. Sandbags don't compare to the specificity of moving parts in windmills and pumps.

[tubewatch59]

Like Behe said in his book, evolution is able to achieve a certain amount of transformation. But it has only been seen to do so succesfully in simple to moderately complex systems (such as enzymes) or the malarial resistance to chloroquinone. Increasing further in complexity we meet 'The edge of evolution' (Behe's book :) after which evolution slows down to a crawl. There are just too many required permutations to chug through and not enough time. Successful evolution has it's limitations.

[Impaler1815]

Hold on. You said evolution can tidy up specificity once the original specificity is in place. Now you're saying evolution can only create vague specificity, but cannot progress much from there. Which is it?

There is no evidence for this "edge of evolution". In fact, Behe submitted that hypothesis to a peer reviewed journal, and despite its rejection and refuttation he still published a book on it.

[tubewatch59]

"Hold on. You said evolution can tidy up specificity once the original specificity is in place. Now you're saying evolution can only create vague specificity, but cannot progress much from there. Which is it?"

Theoretically, if a system can exist for long enough (meaning it kind of hangs around without going extinct or otherwise changing too much ...) then by a process of elimination, all the various types of mutations can be tried, found wanting, others tried, etc. until a sophisticated ...

[tubewatch59]

... system emerges. Not neccessarily all at once. More likely in stages.

Why Behe calls it the 'Edge of Evolution' though is because of the extreme time requirements. Sort of like my other argument about the flagellum and the dependancy on many random mutations that can't be taking advantage of natural selection to build the (not yet existing) functionality that will emerge as the end result of those mutations(maybe).

But you asked 'which is it?' It's kind of both I suppose...

[tubewatch59]

If the functionality is already a given (via creation or 'just because it's already there somehow') then natural selection can get rid of malfunctioning versions and adapt a little to changing environmental conditions, etc. Though I'd think such change would be somewhat limited. You need other sorts of change to build new functions, as described before. And in the time available (~ a billion years or so) only simple systems can arise through this random mutational kind of searching.

[tubewatch59]

"There is no evidence for this 'edge of evolution'. In fact, Behe submitted that hypothesis to a peer reviewed journal, and despite its rejection and refuttation he still published a book on it."

I can't see where he would have gone too far wrong.

I also saw a review where it was noted (with great glee) that Behe's example of irreducible complexity squared (IFT) wasn't actually in place in the germ most featured in his book - malaria!

This kind of shocked me for a while...

[tubewatch59]

... until I found out that the reason that some organisms don't have the IFT, is because the cilium maintenance processes can take place in the cytoplasm instead of having to take place outside. Given that case, it's more efficient not to have the IFT! It depends on where the base of the cilium is located. So that particular objection kind of went away!

There may be minor quibbles, but I cannot see how the general idea of the book has been overthrown. His first book wasn't falsified either.

[tubewatch59]

Of course every uni. student and their dogs think that Behe's first book has been falsified! But a bit of detailed investigation shows that the refutations don't hold up very well. As I said, there are theoretical ways to bypass the constraints of I.C. but practically speaking you really can't manage to do it because the price in terms of time required makes it impractical. Behe's reasoning is completely uncontroversial, except that we are dealing with origins, so it's...'handled differently'!

[Impaler1815]

Now you're just going round in circles. You claim IC is valid because of the need for high specificity and function of the proteins in each step before natural selection can take hold. This is directly contradicted by real world examples like the nylon bug and citrate digesting E coli.

[tubewatch59]

"You claim IC is valid because of the need for high specificity and function of the proteins in each step before natural selection can take hold. This is directly contradicted by real world examples like the nylon bug and citrate digesting E coli."

I have said (if not to you perhaps) that evolution is capable of generating SIMPLE SPECIFICITY or uncovering EXISTING SPECIFICITY by a couple of mutational changes. This is quite different from generating NEW COMPLEX SPECIFICITY from scratch.

[tubewatch59]

But you might object that these functions are new, so the specificity generated also must be new.

But I'd counter that it is known in these cases that the relevant mutations were few in number.

If you had random DNA strings that were not associated with any existing functionality, and applied a couple of random mutations, and as a result started generating working enzymes, etc. then that would be evidence that specificity is an illusion. But this doesn't happen. Also, there seems to be...

[tubewatch59]

... seems to be some findings that in already existing organisms there are redundant genes (or genes that can step in and take over) and some kind of mechanism may exist to swich over from one to the other. If this is found to be the case then this would also seem to be evidence of a designed mechanism of built-in redundancy. But that kind of facility isn't evolution as we normally define it.

But I will have to do some more reading on this topic of redundant genes and facilitated evolution.

[Impaler1815]

"If you had random DNA strings that were not associated with any existing functionality, and applied a couple of random mutations, and as a result started generating working enzymes, etc. then that would be evidence that specificity is an illusion. But this doesn't happen."

You mean exactly what the experiment in this video, and many others like it, have done?

[tubewatch59]

"You mean exactly what the experiment in this video, and many others like it, have done?"

No, it's a far cry from what is occuring here. You have an existing ribozyme (which is obviously not random functionless DNA (or RNA)) to which you are removing a portion (still enabling it to function as a ribozyme. Note: IT's ALWAYS FUNCTIONING AS A RIBOZYME! There is never a state of functionlessness in this experiment. Adding the random sequence reduced activity by x 10000, but it still functioned.

[tubewatch59]

I'm also not implying that no information is being generated in this experiment. I already conceded that to be the case. It can do so because here we have an example of ideal Darwinian evolution in action. We selct for a specific result that we want (artificial selection), instead of a very roundabout way that natural selection allows for, and thus we have a very rapid evolution, to converge to a fairly well optimized ribozyme for that particular function.

But that is besides the point.

[tubewatch59]

Why? Because the ribozyme always has a continuity of function situation. This enables mutations to be guided along a continuous pathway of function towards the selected-for goal of the most optimal function attainable.

This is not what is taking place in the scenario I outlined.

In our scenario, we have existing functionality, but we need to go beyond that to generate the next stage. For example: the association of one existing protein with another protein it doesn't yet associate with.

[Impaler1815]

Yes, evolution can create vague specificity from scratch, and from there specificity can increase. There is nothing stopping specificity increasing to the level any protein is at now.

The claims of organisms being created to evolve are nothing more than ways to dismiss observed evolution. We know there have been cases where genes that have previously been switched off have re-evolved functionality, but there's no evidence that that's the case for most examples of observed evolution.

[tubewatch59]

"...evolution can create vague specificity from scratch"

Yes, simple functionality.

"There is nothing stopping specificity increasing to the level any protein is at now."

There is PLENTY stopping that from occuring. If you evolve from scratch, a simple functionality, then you will (provided you can even manage to select for that new simple function) be able to optimise that weakly performing simple function, towards being a high performance simple function. That's what holds you back.

[tubewatch59]

You cannot get complex multi-protein functionality that doesn't yet exist by a low number of non specific mutations.

Actually you can, provided the random sequence is almost specified already AND the correct mutations occur. Which is so improbable that it happens very very very very rarely!

Or you get it by taking a truly random sequence and by repetitive trials you eventually end up with the correctly specified sequence to enable the function, which takes a very very very very long time!

[tubewatch59]

It doesn't really matter to me whether you agree with that or not because it's all pretty uncontroversial.

What is uncontroversial is that you cannot expect to easily get sophisticated functionality in a small number of steps. You can't evolve this kind of thing by Darwinian means unless these functions already exist. Before they exist, they don't exist!! To get them, you need poorly working versions first, to then refine, but we don't have that! Blind search is the only option!

Time!!!

[EvoBiologist]

Actually, there is ample evidence to suggest that gene duplication solves this problem. Genes form gene families where individuals can do very different things.

A gene adapted to do one thing is duplicated. If that duplicated gene gets a mutation that makes it better than the original gene at binding to something new, now matter how weakly, then as you admitted, natural selection is "able to optimise that weakly performing simple function, towards being a high performance simple function."

[tubewatch59]

"The claims of organisms being created to evolve are nothing more than ways to dismiss observed evolution."

In the newish book 'The Plausibility of Life' the authors argue that life, after having evolved the fundamental processes of cellular machinery, and the right kinds of control networks with 'weak linkages' are then all set up to evolve. The problem is: where did these gene control networks and the cellular machinery come from? And does it really allow evolution across basic bodyplans?

[EvoBiologist]

Interestingly enough, basic body plans don't have to change much. Worms are not very different from invertebrate chordates, which have the same basic body plan as all vertebrates.

[tubewatch59]

"If the protein comes from an existing one then there are advantages in having greater specifity in binding with the pore."

Yes, I allowed forthat. But here is a case where random mutations have ALREADY enabled binding to occur. Once that has ALREADY happenned, then NS can tidy-up the specificity. Though anything NS does to improve specifity for binding is STILL NOT CORRELEATED with the type of proteins that eventually must be allowed through. That later stage also needs random mutations.

[Impaler1815]

It's not like each one of those evolves all in one generation.

Yes, over time lots of traits are selected for. All that means is they all add up to effect the organisms survival. All you're saying is evolution's complicated, therefore it can't work.

[tubewatch59]

"Each mutation works on one base, which in turn effects one gene."

From what I have read recently, that may not quite be true. Apparently, most genes overlap each other up to 5 layers deep (on average). It's all the same DNA strand, but typically 5 different genes use the same DNA! Amazing, but apparently true. So mutations might rather have many effects at once. It's probably the high proportional of near neutral mutations that hides these multiple effects most of the time.

[Impaler1815]

I've heard about overlapping genes, but not at that scale. Do you have a source?

[tubewatch59]

No. Sorry, I read that there were multiple parts of genes coming from different regions and confused it with another reference to overlapping. As far as I can tell, it seems the overlapping layers are usually 2 layers deep. Sorry to have misinformed you (and myself).

[tubewatch59]

ALERT: This is probably wrong.

I made an error in attributing multiple (5)serial regions to overlapping in parallel. Actually there are parallel overlaps, but as far as I could find, only 2 deep.

[Gnug215]

Heh heh heh, uh, it said "cleavage", heh heh.

Ahem, sorry. Felt a Beavis moment and I couldn't resist.

Just wanted to say that you have some excellent stuff going here, keep at it! :)

[a01011399]

you provided an example of an ´´artifitial´´ mutation, do we observe something similar in nature?

[Impaler1815]

Artificial mutation? The only thing artificial about it was the mutagen that was added. It's not like the scientists personally substituted the nucleotides themselves. It was still random changes creating order and specificity.

[CHASCHARLTON]

yo impaler why 30 ?

[ProcInc]

I wonder...if, according to creationists a fin changing morphologically to a leg is an increase in information...

What about when reptiles and mammals had their legs change to fins ala cetaceans, ichtyosaurs, pleisiosaurs, turtles etc?

[Impaler1815]

Creationists have stated that would be considered new information, but of course we can't expect to see something like that. However creationists have also stated that direction, rather than scale, is important when it comes to evolution. So any increase in information is evidence for evolution, regardless of whether it turned a fin to a leg, or turned a random sequence self catalyzing.

[ProcInc]

Yes indeed. It's just a thought since I always appear to find that creationists argue "fin to leg" as an increase in information but never vice versa.

Oh, and I forgot to mention this is an excellent video. I was unaware of this experiment

[Finiras]

at the beginning?

[Finiras]

what is the name of the song again?

[Impaler1815]

'Also sprach zarathustra' by Richard Strauss.

[Finiras]

thanks

[tomdschneider]

Although the work does show an increase of information, these authors did not measure it appropriately in bits.

How to do that has been worked out precisely. See my channel for a link to the evj program which you can run on your own computer (It's safe since it's a java applet).

Lots of fun!

[busymartian]

Thanks for clearifying this topic

The Family Guy pic NEVER gets old!

[Erech01]

ive seen the episode with it in:P its only on for a momentxD

[Ender1zero1]

Thank you for the vid and I can't thank you enough for citing your source of information

[Desertphile]

Thank you for the fine video.

It is in fact impossible to get Creationists to say what they mean by "information:" most evade the question, and many tap-dance around it without providing an answer.

It is my contention that DNA, RNA, mDNA, etc., does not contain "information:" the word does not apply to genomes. My critics insist that it is "obvious" that DNA contains "information" but they have not been able to explain why it is "obvious."

[tomdschneider]

Desertphile, let me remind you again of Claude Shannon, information theory and the evj program. Link on my channel.

[Desertphile]

"Desertphile, let me remind you again of Claude Shannon, information theory and the evj program. Link on my channel."

Yes, and as I noted previously I reject the Claude Shannon analogy: organisms are not computers and genomes are not software; GA's are not DNA.

[tomdschneider]

I must have missed your rejection. Please read the Ev paper. It does not claim that organisms are computers. It is a model of what happens in nature. The important thing is that, as predicted, and using natural selection, the information in the binding sites evolves to match that needed to find the sites in the genome. This is what is observed in nature (see my 1986 paper). So Evj is a successful model of what we see in nature.

[Desertphile]

Greetings; the paper in question does not answer the subject of information in the genome--- it mimics genomes; the information in software is only information because CPUs use human brains, and it is human brains that convert the data / images into information. I contend there is no information in DNA, and I have not seen any paper(s) to suggest I am wrong.... yet I freely admit I may be.

[tomdschneider]

The 1986 paper shows exactly how to compute information in the binding sites of proteins on DNA. That IS information in the genome. A majority of the patterns in DNA are caused by interactions with other molecules - and so fall under this rubric. It has nothing to do with computers or humans since the measures are from DNA found in nature. Please read more carefully.

[Desertphile]

"The 1986 paper shows exactly how to compute information in the binding sites of proteins on DNA."

Indeed, and that is my contention and why I reject the notion: humans perform that calculation.

[tomdschneider]

Desertphile, from your videos I know that you have a good grasp of basic science. So I'm surprised by our response. Do you understand Rs evolving towards Rf as shown in the Evj program? These are models of the world. They describe the world. Just because Maxwell's equations were made by humans doesn't mean that light doesn't have the properties ascribe to it by those equations.

[urbanelf]

You just increased the information... in my heart. Favorited.

[jebus6kryst]

Excellent work. This is something creationist can stick in their pipe and smoke it.

[ProcInc]

"This is something creationist can stick in their pipe and smoke it."

The last thing a creationist needs is more stuff to smoke

[ILYIAB]

Bravissimo!

[rozeboosje]

Nice :-)

[DeadMindWorking]

Great work.

You deserve so many more subs

5*

+

I'm now going to advertise this channel

[Gavrillo]

Same same.

Good to see more Aussies in Science.

[vyrme]

great video! really enjoyed it.

however a bit pointless: our creationist "friends" will come up with some idiotic rebuttal in their time.

until then they'll just ignore this piece of evidence - as they managed to ignore the massive amounts of evidence that do not fit their case (that is ALL the evidence).

[Danmill23]

Very interesting