...but luckily, I'm set straight over at Effect Measure, where Revere completely refutes my silly notion of mutations in H5N1 by citing this excellent guest commentary in the Greely Tribune (where their top story today is about a hot dog from 1952). The commentary is titled "Bird flu a lame claim to evolution theory" and written by one Mike Martin, former editor of Ag Weekly Magazine. He certainly demolishes my silly science-y notation of just what "mutations" (such as those discussed in the Nature article I cited use for analysis) are all about:
But now, "mutation" is being redefined to include any change. If changes can be found in the population of a species they are proclaimed by those who believe in evolution as proof of the theory. The same logic is used in telling us that mosquitoes have "mutated" to become resistant to DDT. But the fact is, before the use of DDT, a small group in the total mosquito population was already resistant. When the insecticide was applied, the resistant insects survived while the unresistant died. The resistant then reproduced forming a strain (a group with common ancestry) of resistant mosquitoes. This an example of population shift, not of mutation in the Darwinian sense. The same fundamentals are used intentionally in cattle breeding to create a population shift that changes a herd of Herefords into black Angus.
Y'see, the "mutations" are already there! It was just a "population shift"! Like, when you have a pre-existing population, and something comes in that challenges it, and only some--let's call them the "fittest"--survive. Yeah, kind of like a "survival of the fittest" scenario. Just like that. That's not evolution, silly scientist.
Let's not stop and think, though, where that initial variation in the population came from. That's just a step too far removed from Mr. Martin's thinking, even though he mentions common ancestry. Surely there wasn't a population--let's say, the common ancestor of the DDT-resistant and DDT-susceptible mosquitoes--that then happened to diverge via the accumulation of those wacky, non-existent things called "mutations"? And when DDT came into widespread use, it killed off much of that susceptible population, leaving the resistant population--voilà, population shift!
Nah, couldn't be. Like Revere said, "Clearly this mutation business is a trojan horse to get 'evolution' into biological science." Duh.
(Aargh.)
I know creationists don't accept this as actual evolution--I mean, no mosquitoes turned into elephants or anything that would really signify a change--but to deny that mutations had anything to do with it because they occurred in the previous step is like saying you can't cause a car accident because all you did was press down on a little pedal. One must understand the history of events before drawing a conclusion about what is or is not relevant, and the reason the mosquitoes are differentially resistant to DDT--or that H5N1 viruses have different abilities to infect humans--is because of the accumulation of mutations, which may be selected for or against in the future. How hard is it, really, to remember RM + NS?
[NOTE: for those of you who may be unfamiliar with Revere or Effect Measure, please note that he, too, is being sarcastic!]
Comments
Argh.
Although, to be fair, if they "remembered" RM + NS, they'd have to accept it, and that wouldn't be possible. Their whole world would come crashing down.
Posted by: The Ridger | July 10, 2006 11:07 AM
It is obvious that mosquitos carry the gene for DDT resistance, and the gene for resistance for all past, current and future insecticides too, but just a tiny number of mosquitoes carry each one of those genes, and they're just waiting the right moment for a "population shift".
See ? No mutations, no "natural selection", no nothing ! No need of godless Darwinist theories !
Posted by: _Arthur | July 10, 2006 11:12 AM
An experiment with antibiotic resistance in bacteria can be carried out with bacteria descended from a single cell. That quashes the "pre-existing variation" argument pretty quickly.
Too bad Mike Martin didn't cover that material when he was getting his degree in technical journalism.
Posted by: wamba | July 10, 2006 11:27 AM
That is scary, he complely missed a HUGE gap in logic there. May he be struck by lightning to put some sense back into him. A simple counter-proof of his arguement is to take 1 bacteria, plate it (as all of its decendants would be identical in his world) and take half grow them in a vat of antibiotics at x concentration, and watch them die. And in the other slowly tone up the antibiotic concentration to x from 0.
Posted by: Kazimieras | July 10, 2006 11:29 AM
Isn't there a bacteria that digests Nylon?
Nylon didn't exist until 1938 (or there abouts).
So, by this argument even a YEC would have to believe that there was a bacteria that survived 5,000 years with nothing to eat just waiting for Nylon to be invented.
Eh?????
Posted by: Jim Ramsey | July 10, 2006 11:55 AM
Sure is.
Posted by: Tara C. Smith | July 10, 2006 12:07 PM
It's easy enough to poke fun at popular articles about Creationism, but part of that is from the fact that popular articles are just that -- popular. It's easy to criticize anything from a popular perspective simply because they have to write for a wider audience, and usually they come from a non-technical background themselves. While I wish this weren't the case, it certainly isn't unique to Creationists.
I am not that familiar with avian flu myself, but if you wanted to pick on a Creationist presentation of avian flu, perhaps it would be better to pick on the technical descriptions given by Creationists, not just the ones for the wider population.
It's not that I necessarily disagree with your commentary, I just think that (a) you should pick a more worthy opponent, and (b) the issue is not with Creationism specifically but with all such things in the popular press.
Posted by: Jonathan Bartlett | July 10, 2006 12:33 PM
<TIC>See, so the mosquitoes COULD turn into elephants, because some of them have the ability, they just haven't cross bred enough to let them elephant genes express themselves.
See, we don't need that Darwin fella, all we need to know is that any population can shift to become some other population because all it needs is time and some kind of "population shift pressure."
And of course they aren't mutations; if they had mutations, they could control the weather, or walk through walls, or shoot beams of energy out of their eyes.</TIC>
Posted by: No One of Consequence | July 10, 2006 12:37 PM
Jonathan, I have indeed "picked on" avian flu posts written by creationists (such as this particularly bad one by Casey Luskin of the Discovery Institute, which covers much of the same ground as the one you linked to from AiG). They both hinge on what I mentioned above: the virus is still a virus, and not an elephant, which is a complete strawman characterization of evolutionary theory. I don't particularly care whether such a mischaracterization comes from a layman or a PhD working for an "approved" creationist organization--it's still wrong.
Posted by: Tara C. Smith | July 10, 2006 12:47 PM
It's not that I necessarily disagree with your commentary, I just think that (a) you should pick a more worthy opponent, and (b) the issue is not with Creationism specifically but with all such things in the popular press.
Wha-wha-what? Scientists shouldn't point out egregious errors in the popular press unless they correct all the errors in the popular press at the same time??
perhaps it would be better to pick on the technical descriptions given by Creationists...
OK. (from AiG:)
*points and laughs* Ha, ha, look at the strawman.
Posted by: windy | July 10, 2006 12:54 PM
Tara --
I think the issue is that Creationists have no problem with evolution meaning change. They have a problem with evolution meaning that everything has a common ancestor and all differentiations are simply the result of natural selection of happenstance differences.
The bait-and-switch, as I see it, is using the same term for both the small-scale change processes as the large-scale ones, and assuming that evidence for the small scale translates to evidence for the large scale, despite the fact that the differences between them are more fundamental.
I don't know of anyone who disagrees with the general idea that everything is changing, and that these changes can be studied scientifically. I know many people, including one person involved in genome sequencing and analysis secularly, who disagree strongly that (a) this can be extrapolated out to all of life, and/or (b) these changes are not genome-directed.
Why is it a straw man to say that we have not observed the grand transformations positted by Darwinism, and even that there are many reasons to think that they do not occur, or at least do not occur in the non-telic way specified?
Posted by: Jonathan Bartlett | July 10, 2006 1:49 PM
"I mean, no mosquitoes turned into elephants or anything that would really signify a change" ? oh yeah, do you have proof for that. Can you claim that with absolute certainty?
God knows it all, and thx to a little enzyme in my brain the trinityase he even speaks to me now, claiming, sorry that should be a PROclaiming elephants were brough about from paper (the bleached, precission-cut one). The mechanism is not to be questioned or attempted to be revenged (reverse engineered) or so his wrath will lead you straight to hell (meaning a patent-lawsuit will entail). God has become pretty materialistic these days, which when i questioned him as to why that is, was countered as an example of natural evolution and adaptation.
Posted by: lo | July 10, 2006 2:07 PM
Take it up with the evidence, then.
Loaded term.
"More fundamental," how? Because of the increased time it took? At the basic level, the differences aren't "fundamental" at all. It's the same accumulation of mutations, insertions, deletions, transpositions, etc.
I do. I'm not sure how this matters.
Again, so? I don't care to argue opinions; I prefer to discuss the evidence. Do you (or they) have any for this view?
How are you defining "observed," here? We certainly have found much evidence of these transformation, both from an examination of the fossil record and from the examination of genetic sequences from extant species. Or are you actually suggesting that we need to see a mosquito change into an elephant (or a cat into a dog, or a virus into a bacterium; pick your morph) before evolutionary theory should be accepted?
As far as a "non-telic" way, you're going beyond the science here. Is it possible that such mutations were somehow supernaturally directed? Sure. I can't disprove that the Invisible Pink Unicorn (PBUH) is tinkering with the genomes of H5N1 isolates even as we speak. But I don't have any scientific evidence for it, either, and neither do you.
Posted by: Tara C. Smith | July 10, 2006 2:28 PM
Jonathan wrote....
"Why is it a straw man to say that we have not observed the grand transformations positted by Darwinism?"
Taking something to its logical conclusion is bait-and-switch? Only if there is no evidence that those seemingly logical conclusions are untrue. Have we not seen grand transformations in the fossil record? Um, yea, we sure as hades have.
There are reptile-to-mammal intermediates.
http://www.talkorigins.org/faqs/comdesc/section1.html#morphological_intermediates_ex2
How about dinosaurs-to-birds?
http://www.talkorigins.org/faqs/comdesc/section1.html#morphological_intermediates_ex1
Or land mammals to whales.
http://www.talkorigins.org/features/whales/
What about boney fish to amphibians?
http://www.talkorigins.org/faqs/faq-transitional/part1a.html#amph1
I dislike being harsh, but at this late date, to state with a seemingly straight face that there are no known examples of macro-evolution..."grand transformations positted by Darwinism" as you stated...one must be either a fraud, delusional, grossly ignorant of the facts or all of the above.
"I know many people, including one person involved in genome sequencing and analysis secularly, who disagree strongly that (a) this can be extrapolated out to all of life, and/or (b) these changes are not genome-directed."
The ole straw man falicy wasn't enough I guess so you had to throw in an appeal to authority.
Posted by: paleotn | July 10, 2006 2:57 PM
But this is simply not what happens. There is BOTH evidence for smaller scale changes in the short term AND massive evidence of larger scale changes that have happened in the long term. The level of detail is not as fine in the larger scale, but it is simply false to assert that evolutionists merely assume the large scale based on evidence for the small. Whether there are mechanisms at work at the large scale not significant for smaller changes is a matter of debate, but there is no indication that such mechanisms must exist, only that they may. IOW, there is no "fundemental" differences between the two that we know of, other than sheer scale, but we do have actual evidence for the large scale changes.
Because they have been observed, at least in the way that observe things in science in general. And there is no reason to think they do not and have not occurred. That you might find some arguments compelling does not mean that they are.
Posted by: Dave S. | July 10, 2006 3:11 PM
Prior to genome sequencing, this might have been a reasonable point of view. But we now know that at the genetic level, all differences between species, both those that appear to be "small scale" and those that appear to be "large scale," are due to an accumulation of small genetic differences, which happen to correspond exactly to the kind of changes that occur as a result of mutation.
Posted by: tgibbs | July 10, 2006 3:27 PM
"Take it up with the evidence, then."
Okay, try Gene regulatory networks and the evolution of animal body plans for starters. The point is that the evidence is not conclusive by any stretch of the imagination. There are a lot of similarity, but given that there are experimentally-apparent limits to change, it is not clear from the evidence itself that these are the result of common ancestry.
""More fundamental," how? Because of the increased time it took?"
See above paper.
"Again, so? I don't care to argue opinions; I prefer to discuss the evidence. Do you (or they) have any for this view?"
As far as the directed mutations, there are many papers I could point you to. The best compendium, which is a little out-of-date, is "Molecular Strategies in Biological Evolution". A good paper on the general idea is "A Biochemical Mechanism for Nonrandom Mutations and Evolution".
"We certainly have found much evidence of these transformation, both from an examination of the fossil record and from the examination of genetic sequences from extant species."
We have found evidence of multiple species, and a possibility of a time sequence (given the number of lazarus taxa and ghost lineages, it is foolhardy to assume that something's first appearance in the fossil record is when it actually first came to exist). What we do not have is the mechanisms that produced those taxa. Whether you are a Creationist, a front-loaded evolutionist, an RM+NS evolutionist, or Gaia evolutionist (such as Margulis), the fact is that the fossil record doesn't provide any evidence about how a given taxa formed.
"Or are you actually suggesting that we need to see a mosquito change into an elephan"
I think it would at least be nice to see at least the generation of an organ or novel tissue type, and show that it occurred via RM+NS.
"As far as a "non-telic" way, you're going beyond the science here. Is it possible that such mutations were somehow supernaturally directed?"
I think you are misunderstanding ID here. While I think that perhaps Dembski personally holds to a continuous intervention scenario (I'm not sure), most ID'ers and Creationists believe in a front-loaded evolutionary scenario (whether a single front-loading event for front-loaded evolutionists or multiple for Creationists). By "telic" and "non-telic" I am referring not to supernatural intervention, but based on whether or not it is acting on information that already exists in the genome, rather than acting haphazardly. The problem is that even fairly minimal search spaces are nearly impossible to search fruitfully without structuring the search space beforehand (using information about the problem). Ultimately, such a view requires one to ask where the information came from, but you can examine evolution from a telic perspective without invoking God, much the same way that many investigate the Big Bang without invoking God, despite the fact that if the Big Bang is true, it would certainly point many in that direction.
Were the major jumps developed by Symbiogenesis? Special Creation? A time-trigger in a front-loaded evolution? A habitat trigger in a front-loaded evolution? Random Mutation + Natural selection? The only evidence we have are the discontinuities. We don't have any evidence yet on how such discontinuities developed.
The problem with the RM+NS scenario is that I don't think that it is supportable mathematically. There are too many "jumps" that are large that must be crossed. Any single jump over 500 bits is simply impossible by the RM+NS scenario. And Avida showed that even tiny, tiny, tiny jumps are nearly impossible even with computers covering massive numbers of generations in an ideal environment, where regulatory networks aren't even available for modification.
Posted by: Jonathan Bartlett | July 10, 2006 3:47 PM
Tara -
If possible, please publish the text of your letter to the Greeley Trib.
Best,
Jim
Posted by: jre | July 10, 2006 4:37 PM
I fail to see how that paper (which was nicely blogged by PZ when it was new) supports your case. I assume you're focusing on the part where they discuss limits once these kernels are in place, and ignoring the portion where they discuss their evolution in the first place?
The link to the latter is here, for anyone interested. For the first, do you mean this? Again, those don't really support your view. Certainly microbes can increase their mutation rate when exposed to stress, but that's not quite the same as the "directed mutations" you mention.
It would be nice to win a million dollars too. In the meantime, y'all dismiss what are IMO pretty damn incredible examples of RM + NS, such as the aforementioned nylon-eating bug.
There is no singular "ID" to misunderstand. It seems that each IDer has their own, often conflicting, idea of what ID, or creationism, is.
But this is another strawman. We are all shaped by our environment, and there's no reason to assume, using evolutionary theory, that such information as you suggest would be lacking in the genome, or that actions would be necessarily "haphazard." That's the point of the review you linked above, as a matter of fact.
But this is often misapplied, because it assumes a target to begin with. Evolution needs no such target. Additionally, even though this is outside of my area of expertise, smarter folks than I have pointed out that Dembski's applications of such theorems are wrong. This goes to your other information comment as well--check out Mark's blog for other misapplications of math to evolutionary theory.
And it's turtles, all the way down when you go that route.
But again, guess which ones we have evidence for?
Posted by: Tara C. Smith | July 10, 2006 4:50 PM
jre,
Stupidly, I didn't save it. It was one of those submission boxes (like this one) and when you pushed "submit," it was off into the ether.
Posted by: Tara C. Smith | July 10, 2006 5:08 PM
I think Creationists (including ID supporters) should publish more criticism of bad science they see within the Creationist Big Tent. That way we would know they weren't ignorant or malevolent science-phobes with political agendas based on blind allegiance to religious dogma. They could start with issues for which the evidence is clear and the appropriate conclusions have been accepted in the scientific community for a century or more, such as the age of the Earth. Maybe those well-funded folks at the Discovery Institute could help out in publishing a series of essays on such topics.
Posted by: wamba | July 10, 2006 5:15 PM
I encourage all of you to click through Jonathan's link to a baraminology blog for some good entertainment!
Posted by: wamba | July 10, 2006 5:21 PM
Jonathan wrote
And they don't mean what you think they mean. Basically, underseveral environmental conditions, the error checking mechanisism in DNA replication are relaxed, and the mutation rate goes up. The mutations are as random as before. For more details see my post on the non-directednesss of directed mutation
Posted by: Ian Musgrave | July 10, 2006 6:25 PM
Okay, I've never read any postings by Jonathan Bartlett here on Aetiology until now. So I read his comments on this thread, and then Tara's and others' replies, and then looked at his website.
So Bartlett is a computer programmer. I'm not a biologist or a scientist of any type, but given the choice of believing Tara who is a Ph.D. scientist in a biological science, and reading Bartlett who is a computer programmer who appears to run anti-evolution website (despite the descritpion at the top of the home page, that is clearly what it is), I'm going to go with Tara for my information on evolution related topics.
I really have no interest in scientific arguments made by computer programmers. Why Bartlett posts here is a mystery to me - I doubt any intelligent and critical thinking person who reads Aetiology is going to be convinced by any of his arguments.
Posted by: Sandra in Dallas | July 10, 2006 11:00 PM
If it's any comfort, I've done that exact thing with the Greeley Trib, and suspected afterward that my passionate, yet closely reasoned submission went to /dev/null. It is also possible (and maybe more reliable) to send an email to letters@greeleytrib.com. It is my understanding that the Trib prefers letters from local writers. Though I am not myself a UNC alum, I may be able to locate one. And I note that Professor Neil Snow teaches BIO 465 (Evolution) at UNC. If I call this travesty to his attention, and he's not doing some professorial July fun thing, perhaps he might be induced to reply.
Posted by: jre | July 10, 2006 11:01 PM
"ignoring the portion where they discuss their evolution in the first place?"
It was referenced but not supported. It was assumed to be possible, but not empirically or theoretically supported, just stated. The part that I thought was particularly interesting was the following comment: " Current microevolutionary thinking assumes that observed types of genetic change (from single base substitutions to gene duplications) are sufficient to explain all evolutionary events, past and present...But attempting to explain an aspect of animal evolution that depends on one kind of network alteration by adducing evidence from an aspect that depends on another can be fundamentally misleading". Perhaps I am reading him wrong, but this seems to be a big red warning sign against the idea that we can extrapolate observed evolutionary processes to the evolution of animal body plans.
"Again, those don't really support your view. Certainly microbes can increase their mutation rate when exposed to stress, but that's not quite the same as the "directed mutations" you mention."
"Molecular Strategies" only has one chapter related to increasing mutation rate. The rest are on specific changes that may be invokable by the cell (some authors had that view, others did not). In the abstract to the lead-off paper, they essentially say that random mutation should be rejected as the basis for natural selection. For a good example, see Shapiro's paper in that volume. As for the "Biochemical Mechanism" paper, she points out that the cell is in charge of transcription, and can use transcription to induce changes in genes that are needed for metabolism, and that the genes themselves are structured to change in the most likely spots (the loop of the stem-loop structures).
"y'all dismiss what are IMO pretty damn incredible examples of RM + NS, such as the aforementioned nylon-eating bug."
And on what evidence do you have that this change was RM+NS? There is considerable evidence that it wasn't. If the number of changes that occurred in this gene happened genome-wide, it would likely lead to error catastrophe, would it not? The frameshift occurred in a spot that appeared to be designed to undergo frameshifts, included transposable elements (often indicated by supporters of directed mutation as being one of the key components), and had numerous point mutations. Did that kind of mutational activity happen to every gene in the genome? Or was it directed? In addition, there were _two_ genes who changed in coordination. The evidence that this was RM+NS and not directed mutation seems to be waning.
"There is no singular "ID" to misunderstand. It seems that each IDer has their own, often conflicting, idea of what ID, or creationism, is."
That's because ID isn't a theory of origins but rather a theory of causation. Of course, there is also no singular "evolutionary theory", either.
"We are all shaped by our environment, and there's no reason to assume, using evolutionary theory, that such information as you suggest would be lacking in the genome"
So where does it come from? We have no other systems that we know of that produce symbolic coded systems, or create solutions to problems.
"But this is often misapplied, because it assumes a target to begin with. Evolution needs no such target."
If the environment changes, staying alive requires a target or multiple targets.
I'll have to check out your math links later when I have more time.
"But again, guess which ones we have evidence for?"
Please show me an example of a beneficial change for which we have had computational evidence that the change was random and not directed. I've shown plenty of examples which allude to directed change, and just for fun I'll point to one other: "Transposable elements as activators of cryptic genes in E. coli". You have simply assumed that because the nylon change occured that it was because of RM+NS. Where is the computational evidence?
Posted by: Jonathan Bartlett | July 10, 2006 11:34 PM
So, Johnathan:
What's your explanation for the nylon bug? Was it designed to eat nylon by G- , uh, I mean, a prescient designer who knew we would invent it someday? Assuming that was a scientific hypothesis and not religious apologetics, where's your "computational evidence" for that? Or better yet, experimental physical evidence? You know, like an actual scientist would have?
Are you saying all changes are directed and nothing is random? How do you explain biological monsters? Rh 'blue' babies? Co-joined twins? The extinction of 99.9% of all previous species? All part of the Master Plan? Where's the computational evidence?
It might be time for you to turn off the computer and go for a walk out in the real world soon.
Posted by: fnxtr | July 11, 2006 1:10 AM
"What's your explanation for the nylon bug?"
I think that microbes have the innate ability to manufacture new genes for new situations. These are essentially template based, where the cell can take a template, add functional pieces through mobile elements, and make small modifications through single nucleotide changes to finish. All of this done on the right gene (or one of a small set of genes) that it is likely to work on.
"where's your "computational evidence" for that?"
The number of changes that occurred, and the length of time it occurrred in.
"Are you saying all changes are directed and nothing is random?"
No, I think there are two parts of randomness. First of all, I think that randomness is the optimum way to explore a small search space where the outcome is not decidable based on the existing sensing mechanisms. I also think that randomness can be the result of non-telic happenings. I just don't think that the latter set of random changes can produce complex functional adaptations.
"The extinction of 99.9% of all previous species?"
Actually, this is not well known, but the estimates for the number of species that has lived is based on Darwinian evolution (i.e. with innumerable transitions between species). If Darwinism is wrong and those transitions never existed, then extinction has not occurred nearly to the extent that many people assume. In addition, I personally think that the unit of creation was at approximately the family level. Therefore, the disappearance of a created form, while it has happened, has not happened nearly to the extent suggested in the literature. For those ID'ers not believing in special creation, then the unit of creation would essentially be the holistic structures themselves, not necessarily any specific taxa. I am not sure if any of these have ever gone extinct.
Posted by: Jonathan Bartlett | July 11, 2006 1:35 AM
Jonathan wrote:
.Sorry, they all about mutation rate in one form or another (even with horizontal gene transfer, mutation comnes in). You may not recognise where mutation is involed, but it is there.
Jonathan wrote:
.All that happens is that during nutrient stress genes are transcribed faster, so they mutate faster, the mutations are still random with respect to function (and is quite clear in that paper).
Posted by: Ian Musgrave | July 11, 2006 2:25 AM
Luria, Delbruck, and Alfred and Martha Hershey demonstrated that mutations arise spontaneously, even in the absence of selection, in the 1940s....
http://nobelprize.org/nobel_prizes/medicine/laureates/1969/
It is tragic that 60-year-old, Nobel-decorated science that provides the mechanistic foundation for essentially all of modern biotechnology and genetics is still not accepted by a vocal cadre of radical know-nothings.
Posted by: Alexey Merz | July 11, 2006 3:30 AM
Hey, people - don't slam computer programmers. I'm one. However, feel free to slam Jonathan for his cherry-picking, lack of understanding and many other faults.
Posted by: Kristjan Wager | July 11, 2006 5:40 AM
If Darwinism is wrong and those transitions never existed, then extinction has not occurred nearly to the extent that many people assume. In addition, I personally think that the unit of creation was at approximately the family level.
Like the hominid family?
Posted by: windy | July 11, 2006 7:10 AM
Ok, I'm completely new here, and in over my head, but I'll take a stab at addressing Jonathan's need for "computational evidence": It's my understanding that background rates of random mutations are well documented in contemporay organisms, due to studies in cancer research for example. Now, we can also construct well supported phylogenies based on molecular assessments of contemporary organisms. Isn't it the case that when we project these molecular lineages back in time using current random mutation rates that we get dates quite reasonably (within statistically valid ranges) equivalent to the dates of the fossil record? I know that its all way more complex than this, but the general idea holds. If so then this is strong evidence that evolution occurs at the same rate as random mutation, and hence is mostly like produced by that mechanism. Why would "directed" change take the same time as random change? Seems to me that the Creator is trying to look like rm + ns.
Posted by: Sylvilagus | July 11, 2006 8:38 AM
always always the same flaw when creationists bring up resistance. IF there was no pre-existing resistant subgroup then any exposed group would ALL die off because the rescuing mutation isn't going to magically appear in the nick of time, runs the insurmountable argument, ignoring the fact that substances appear at different concentrations in the environment. when the local concentration kills a few percent of bacteria, the population gets by quite happily but a mutation for resistance will spread through the population and permit colonization of more lethal concentrations.
Posted by: snaxalotl | July 11, 2006 10:19 AM
There's a nice Panda's Thumb article by Andrea Bottaro on the possibility but
(so far) apparent nonexistence of directed mutations.
http://www.pandasthumb.org/archives/2004/04/getting_away_wi.html
Posted by: Anton Mates | July 11, 2006 11:54 AM
"I think that microbes have the innate ability to manufacture new genes for new situations."
I think that too. I think its called random mutation/natural selection. I think there are a few papers on it in the research journals.
"These are essentially template based, where the cell can take a template, add functional pieces through mobile elements, and make small modifications through single nucleotide changes to finish. "
I think that too. I think its called transposition. I think there are some papers about this, too, in the research journals.
"All of this done on the right gene (or one of a small set of genes) that it is likely to work on."
This is a bit unclear. If I read you right, there are some very old papers on this by some guy called Lamarck or something. But I think that theory was replaced by work by another guy whose theories actually bore up under evidence. I think his name was Dabbin or Darwin or something like that. I could find that for you if you are interested?
Posted by: Chiefley | July 11, 2006 11:56 AM
"Please show me an example of a beneficial change for which we have had computational evidence that the change was random and not directed."
I think you stumped the panel with this one Jonathan. It would be really difficult to show the computational evidence that the change was random and not directed.
But there is a guy a number of years ago who developed an interesting theory which says that random changes in the traits of organisms might be strongly directed by what he calls "natural selection".
What he means by this is that an organism that acquires a new or modified trait that is better suited to the environment would have a better chance of surviving and thereby propagating its genetical material to its progeny. Its interesting, because that basically places the all the characteristics of the environment into the role of "director".
For example, he would say that rabbits that run slower than foxes would get eaten at a faster rate than rabbits that run faster than foxes. So, over time, the surviving generation of rabbits would on the average get faster. I, personally, find that incredible. Where is the evidence for that?
To confirm a crazy notion like that, you would have to find some biological mechanism that determines the traits of an organism, some mechanism for it to be changed randomly, and then some mechanism for it to be passed on.
I say he is on the wrong track. Clearly this stuff happens "automagically".
I can find that paper for you if you are interested. It was by that Dabbin guy again. Keep up the good work, Jonathan
Posted by: Chiefley | July 11, 2006 12:42 PM
I'm not sure if this was meant as ironically as the rest of your post, Chiefley; just in case it wasn't (or someone else assumes that it wasn't), Luria and Delbruck came up with quite a nice way of doing exactly that. (Which is also covered in the PT article I mentioned above.)
Posted by: Anton Mates | July 11, 2006 12:51 PM
"Please show me an example of a beneficial change for which we have had computational evidence that the change was random and not directed."
As Mr. Mates said.
Try looking in a good genetics textbook such as Griffiths:
http://www.amazon.com/gp/product/0716749394/sr=8-1/qid=1152636721/ref=pd_bbs_1/104-3302810-8486367?ie=UTF8
Look in the Index under "fluctuation test." That is, in large part, the method (developed in the 1940s) that got Luria and Delbruck the 1969 Nobel Prize. Also look up "replica plating," another method developed by Alfred and Martha Hershey, which can answer many of the same questions.
It is not really possible to have a serious discussion about mutations until you understand these pivotal experiments and some of their implications. Otherwise, it's like trying to discuss the merits of the designated hitter rule when you don't know what the bat is for.
Posted by: Alexey Merz | July 11, 2006 1:01 PM
Apologies for the 2x post.
Posted by: Alexey Merz | July 11, 2006 1:02 PM
Jonathan wrote:
You are correct that it would be a mistake to believe that single nucleotide polymorphisms and gene duplications are sufficient to explain the origin of bodyplans. However, this is not to say that we have no idea how they originated. For one thing, there are different kinds of mutations. For example, tandem repeat slippage, polyploidy (aka "whole genome duplication," which will include the Hox complexes), transpositions, and, of central importance to the problem you raised, intrachromosomal and interchromosomal rearrangements.
Such rearrangements are due to the existence of hairpin and cruciform structures which are partly the result of inverse palindromic sequences which can occur in H- and Z-DNA. Such structures result in mechanical stress and may result in breakages. When such breakages occur, there are repair mechanisms, but they don't always get things right. We know that such rearrangements occur in part because roughly one-tenth of one percent of all humans born have such rearrangements. Moreover, half of these cases are not clinically significant. Likewise, such breakages may occur in somatic cells. When this happens, cancer will often be the result, and this is responsible for much of the cancer we find today.
In any case, I realize that this doesn't answer all of the "points" you brought up (frankly I have other priorities), but perhaps this much will be of some value.
Posted by: Tim | July 11, 2006 2:29 PM
Jonathan wrote:
One of the theories which was put forward is that some mutations are directed in the sense that a mechanism had evolved so that when these mutations occur, they will tend to promote the survival of the organisms which have them more often than would be the result of chance.
The null hypothesis of course is that there is no such direction. Some research had suggested that there may be evidence for such a mechanism (much like research had suggested lateral gene transfer between bacteria and multicellular eukaryotes without any viral intermediary), but later research suggests that there is no need to invoke such a hypothesis.
However, what does exist are hypermutations which in some cases result in a far greater rate of mutation when the organism is under stress. We have seen this in e. coli. Likewise, the adaptive immune system of mammals involves such a mechanism. If I remember correctly, two transposons (a type of mobile genetic element) are involved.
Interestingly, mammals also have a transposon which results in the overproduction of neural cells in the brain where those which make the "right" connections are preserved and those which do not undergo apoptosis. This permits greater complexity and consequent plasticity than what is found, for example, in reptiles. It has also been suggested that this transposon may be responsible for the large number of types cells found in the human brain - which may outnumber the types of cells found in all other organs combined.
Posted by: Tim | July 11, 2006 2:56 PM
Kristjan Wager wrote:
Don't worry - I have a great deal of respect for natural selection and genetic drift, definitely two of the best in field - albeit both tinkerers.
Posted by: Tim | July 11, 2006 3:03 PM
Alexey Merz brought up the fluctuation test four posts up - something I hadn't heard of before - and which will probably modify my understanding of the phenomena of hypermutation. I found a webpage which covers gives a fairly introductory level explanation of the test - as well as a more technical article which touches on how the fluctuation test and hypermutation are related.
Thank you for the lead, Alexey!
Luria & Delbruck Fluctuation Test
http://www.biotech.ufl.edu/gabriel/dglLuria&Delbr.htm
Hypermutation and the Preexistence of Antibiotic-Resistant Pseudomonas aeruginosa Mutants: Implications for Susceptibility Testing and Treatment of Chronic Infections
Antonio Oliver, Bruce R. Levin, Carlos Juan, Fernando Baquero, and Jesús Blázquez
Antimicrob Agents Chemother. 2004 November; 48(11): 4226-4233.
http://www.pubmedcentral.gov/articlerender.fcgi?artid=525420
Posted by: Tim | July 11, 2006 3:20 PM
It occured to me a while ago that one of the things that seems to be going wrong in the discussion with creationists is that they just don't understand evolution, and are somehow stuck with a Lamarckian view of it. And that that is why they keep saying things like "Monkeys can't turn into people!!!".
This quote just enforced my idea - they'd happily accept Darwin's idea (even use it as argument to prove the opposite), but it's actually Lamarck they have a problem with.
Posted by: Eva | July 11, 2006 5:31 PM
I wish it were as simple as ignorance. But within the past year, I have debated them on a good number of forums for extended periods of time on more occasions than I can count, and I am fully convinced that shortly after they get started debating (if not before), there is something very deep within them which has become very twisted. And frankly, I hate having to say that about anyone.
The value which I find in debating them largely consists of:
1. have them demonstrate their dishonesty;
2. use their arguments as a foil for explaining what we know;
3. try to answer my morbid sense of curiosity regarding what went wrong.
Who knows? Perhaps there will be a few creationist observers out in the bleachers who will see their "great warriors" for what they are - and will abandon creationism. At least this is my hope.
PS My apologies if this was a little heavy.
Posted by: Tim | July 11, 2006 6:04 PM
Tim --
"We have seen this in e. coli."
Look at the Barabara Wright paper I referenced. Her explanation better explains the e. coli data.
"Likewise, the adaptive immune system of mammals involves such a mechanism."
And in the adaptive immune system we know that, at the very least, the hypermutation is restricted to the specific set of genes in need of change (the V/D/J genes), if not more specific (there are several potential somatic mutational mechanisms proposed which are more specific which may or may not pan out).
What's even more interesting, is that during the recombination, there is evidence that the cell has a "smart" joining mechanism which is able to predict what amino acids are needed, which are not templated, to make a good, functional gene where the templated segments would not themselves have done so.
"Such rearrangements are due...."[followed by a mechanical explanation].
Just to point out, the existance of a mechanical explanation is exactly what I said was there. Obviously, any strategy must be implemented mechanically. What other kind of implementation would there be? However, the interesting fact is that these mechanical systems often point to the most useful areas for change.
"When this happens, cancer will often be the result"
Yes, searches do not always hit the right answer, and genome degradation can cause a search function to go awry.
Alexey --
What Luria Delbruck showed was a method to tell if a given change happened in response to selection or without selection. It does not tell if the change was random with respect to fitness. This may seem counterintuitive, so let me explain.
Luria Delbruck occurred before DNA structure was determined, and well before the code was established. Therefore, they had no way of knowing how large a baseline "random" mutation would occur. For instance, in a genome with 2 billion base-pairs, a given random genetic change (in the absence of selection) involving 2 base pairs would occur less than 1 in 4000000000000000000 combinations. Therefore, while a given change may be partially stochastic, and may change without relationship to selection, if it occurs considerably more often than the base line for randomness, then it is not likely that the change is truly random. It is more likely to be phase-variable or some other similar mechanism. Luria-Delbruck can tell you if a change happens in response to selection or not, but unless you compare the change rate to the expected change rate for the given change size, you don't have any information as to whether or not the change is random. In fact, for a small search space, a random search is the optimum for an unknown number of searchers. For example, if you wanted to keep a population alive in the face of very diverse conditions, a good way to do that would be to have multiple metabolism options which distributed themselves randomly throughout the population. Then, in the face of strong selection, there would likely be a member of the population capable of surviving already.
Eva --
I'm not sure why you made your Lamarck comment. Most creationists are much more sympathetic to Lamarck than to Darwin.
Posted by: Jonathan Bartlett | July 11, 2006 10:11 PM
"What Luria Delbruck showed was a method to tell if a given change happened in response to selection or without selection. It does not tell if the change was random with respect to fitness. This may seem counterintuitive, so let me explain."
The Luria-Delbruck methodology can be used to examine neutral, adaptive, or maladaptive mutations. It's entirely a question of how one LOOKS for mutants, i.e., the assay system. Due to its ease of implementaiton and superb signal-to-noise ratio, Luria and Delbruck used viability as the readout. But there is no reason in principle why they had to do this.
It is also important to note that "fitness" and "adaptivity" can be defined only in relative terms, i.e., with respect to growth compared to some baseline population and in some specific set of environmental conditions.
One must also be very careful when using the word "random." We know, and have known since Symour Benzer's experiments with T4, that there not every nucleotide position is equally mutable. Contemproaneous work (much of it from Hershey) taught us that different mutagens (UV, EMS, acridine orange, etc) cause different mutational spectra. More recent work has elucidated many of the chemical and biochemical bases of these phenomena.
Going back to L-D, with an appropriately tailored assay system, one can use the L-D methodology, replica plating, flow cytometry, or other approches to quantify the rates of *particular* kinds of mutations (frameshifts, transitions, transversions, etc.) in various sequence contexts and in organisms under various environemental conditions. This is not speculative, but rather a broad theme in actual experimentation - resulting in literally thousands of technical papers - that has been ongoing for over half a century. After all, the fluctuation test paper dates to the 1940s and were recognized as Nobel-worthy >35 years ago.
Indeed, there was an active controversy in the mid-1990s (alluded to in this comment thread) over whether the frequency of a mutation at a given site can change if that site is under selection (compared to a site not under selection). Although the weight of evidence at present suggests that such a scenario is uncommon if it occurs at all, it is worth noting that nearly every paper on the topic relied on fluctuation tests the gold standard analytical tool.
Posted by: Alexey Merz | July 11, 2006 10:46 PM
"This is not speculative, but rather a broad theme in actual experimentation - resulting in literally thousands of technical papers - that has been ongoing for over half a century."
Yes, but the question is whether or not mutations are random, not whether or not they are independent of selection.
In fact, what you seem to be indicating yourself is that mutation is a highly biased event. So the question is, are these biases more inline with "sensible" genomic changes than nonsense ones which lead to error catastrophe? If so, how do you think the biasing arose in such a way? If genomic change is non-telic, then we would not expect sensible mutations to have such a predominant place. Why would the biases be biased towards sensible changes and not catastrophic ones?
"it is worth noting that nearly every paper on the topic relied on fluctuation tests the gold standard analytical tool."
However, as I pointed out, fluctuation tests show "random" even in the case of a stochastic cyclic variation. Fluctuation only shows whether a mutation is the result of selection, not whether or not it was random.
Posted by: Jonathan Bartlett | July 11, 2006 11:31 PM
"Yes, but the question is whether or not mutations are random, not whether or not they are independent of selection."
Wrong. The analytical machinery of the fluctuation test can be used employed in some very elegant experimental deigns that *do* address whether, for example, a selection can cause mutation rates to change at different sites in a single genome.
"In fact, what you seem to be indicating yourself is that mutation is a highly biased event."
That's been understood for at least several decades.
"So the question is, are these biases more inline with "sensible" genomic changes than nonsense ones which lead to error catastrophe?"
You're going to need to tell us how, experimentally, we might define "sensible."
I caution you that "nonsense" has a specific technical meaning. It indicates a mutation that causes premature termination of a polypeptide, as opposed to a missense mutation which results in substitution of one amino acid for another, a frameshift mutation, or a silent mutation. So you might want to avoid the term "nonsense."
"If so, how do you think the biasing arose in such a way?"
There are several factors here. First of all, there is the (comparatively) simple chemistry of DNA itself. First, some sequences of bases are intrinsically more susceptible to chemical attack or degradation than others, and consequently will be more mutation-prone. Second, the DNA replication and repair machinery does not fare equally well in replicating all sequences. For example, in long stretches of G's, polymerases can sometimes slip, adding or subtracting a nucleotide.
This is the source of the so-called poly-G repeat diseases such as Huntington's Disease. It is also a mechanism used in many bacterial phase variation systems such as the hemoglobin utilization system of the meningococcus. Note that such systems could quite easily arise through mutation and natural selection, if phase-switching is beneficial to the organism. It's estimated that 10% of the genes in the genome of the gonococcus (a close relative of the meningococcus) are phase-variable due to G-rich or GC-rich repeats found in or near the upstream regions of genes.
Other factors: regions of DNA that are "buried" in double helices, packed on nucleosomes, and generally in a compacted storage form are much more susceptible to damage than regions of DNA that are unwound, such as those being actively transcribed or replicated. In addition, physical location on the chromosome can be an important factor. Finally, there are hotspots for recombination, etc.
In summary, we can say that there are physical, chemical, biochemical, and cell biological mechanisms that lead to different sites having different mutation rates. This is not, as I've already pointed out, a recent discovery.
"If genomic change is non-telic, then we would not expect sensible mutations to have such a predominant place. Why would the biases be biased towards sensible changes and not catastrophic ones?"
We have very little evidence that mutations, taken generally are biased in the way you suggest, and that lack of evidence is not for lack of looking. The most promising example of a system that seems to mutate more rapidly under selection is found with a particular loss of function mutation in the E. coli lac operon, which appears to revert to functional form at an increased rate (relative to other sites) when the cells are carbon-starved. Even in this case - by far the most intensively studied - the mechanisms responsible and the degree to which the mutations are "directed" has been hotly debated for a decade.
I would suggest that a "sensible" approach is to establish WHETHER a phenomenon is happening at all, before we ask WHY it AROSE in the first place, lest we waste our time speculating about faeries. There are a number of good research groups actively working on this question, but it is fair to say, I think, that no one has yet adduced really compelling evidence that mutations are generally biased toward "sensible" changes that reflect specific selection pressures. I am not saying it can't happen, just that there is not convincing evidence that this occurs commonly.
"However, as I pointed out, fluctuation tests show "random" even in the case of a stochastic cyclic variation."
Please explain what you mean here in greater detail. Fluctuation tests do not merely show a signal of "random" or "nonrandom." Rather, they allow one to *measure* the rate at which *new* mutations spontaneously accumulate in a population.
"Fluctuation only shows whether a mutation is the result of selection, not whether or not it was random."
Perhaps you need to define more precisely what you mean by "random" in this context, preferably with detailed examples of what would qualify as random, and what would qualify as nonrandom.
Posted by: Alexey Merz | July 12, 2006 1:54 AM
Jonathan wrote:
What's even more interesting, is that during the recombination, there is evidence that the cell has a "smart" joining mechanism which is able to predict what amino acids are needed, which are not templated, to make a good, functional gene where the templated segments would not themselves have done so.
So the fact that there are self-correction mechanisms and that a system settles down to patterns which are self-stablizing is evidence for "telic causation."
Y