Uncommon Descent Serving The Intelligent Design Community

Darwinists are Always Surprised

Share
Facebook
Twitter
LinkedIn
Flipboard
Print
arroba Email

Here’s a study on E. Coli. They force the bacteria to “mutate” to process glycerol. After six days, sure enough, a kinase shows up to handle the glycerol. But what is a “surprise” is that RNA polymerase shows up besides. It seems that two simultaneous mutations took place. But, of course, this is ONLY a surprise if you think RM+NS brought it about.

The authors say:

Mutations also appeared in a second, unrelated gene for an enzyme called RNA polymerase. “That was a surprise to almost everybody because RNA polymerase is involved in one of the core processes of any cell,” said Palsson. “You wouldn’t expect that gene to change because a wide variety of cellular process would be affected; it’s like replacing the wiring system in a building when a light bulb burns out. But we repeated the experiment more than 50 times and mutations in the RNA polymerase gene appeared again and again.”

I also enjoy the hesitation you almost hear as the reporter has to backtrack somewhat from RM+NS (listen for the word “presumably”):

All the mutants arose in the experiments presumably as the result of naturally occurring errors in copying DNA into daughter cells during cell division.

We here at UD have a better idea about what’s going on.

Here’s the link to the article.

By the way, some time back, in a heated debate over bacterial mutation and point mutations, I calculated that the amount of bacteria needed to provide TWO simultaneous point mutations simply through random chance would be so large that the known universe couldn’t contain it. I’m sure my calculation was off somewhere, but I think you get the picture: two genes being affected simultaneously is something that cannot happen by chance alone.

Comments
DaveScot Thanks, appreciated the links. devilsadvocate
Wow... that's kinda wild, that's like somewhat coordinated, is it not? I see the argument better now, a light went off... Is this potentially more surprising than the nylonase discovery of rapid plasmid mutations? Michaels7
Or am I missing some greater point here made in the article on PhysOrg? How is this discovery different from nylonase? Ooops... OK, its still an enzyme breaking down glycerol. But, RNA polymerase... ding. I need to read up a bit more. Michaels7
I meant a more general category of stuff that doesn't warrant a design inference yet there is a definite major change. Patrick
Patrick, Nylonase? If you don't mind, what category did it fall into? Plasmids? When I saw this post, first thing that came to mind was nylonase rapid changes in the lab. At the time, it was a surprise to those testing if I remember correctly because it had emerged some 30 years after the introduction of nylon waste into nature. Whereas with stimulation I think they did it under 9 days in the lab culture with bacteria never exposed to nylon. I understand it was accomplished by Plasmids, external to chromosomes, yet self-replicating. When I see repetitive results, it makes my eyes and ears perk up. Are there plasmids on this particular strain? Because plasmids do interact within bacteria cultures, correct? spreading information? Michaels7
Patrick: "Perhaps. Personally I’d advocate caution since I don’t think we have enough data to make a definitive claim but I think that hypothesis is at least worth looking into. " Some people are blessed with intuition. ;-) PaV
Either way, this nearness of time means that the “odds” against this happening stochastically are huge; ergo, this is a “pre-programmed” response.
Perhaps. Personally I'd advocate caution since I don't think we have enough data to make a definitive claim but I think that hypothesis is at least worth looking into. Not to mention, if we did have enough information I'd want to employ the EF first...follow the ID process and not skip ahead. Patrick
HodorH "I think you should explain what you mean by simultaneous. The article certainly doesn’t indicate that the mutations both occur in the same original cell. It is not particularly useful to tell someone to figure it out for themselves when asked for a clarification. " It takes about six days to bring about a mutation, whether for the glycerol or the RNA polymerase, and for culture to be of sufficient size so as to assay it. If "both" appear 6-8 days later, then "both" mutations happened early on. Without specifics, it's harder to say more; but, let's not miss the forest for the trees: they say"The researchers report that mutations in both genes appeared together in several cases after six to eight days in the glycerol-based cultures, and E. coli cells containing both mutated genes grew 150 percent faster than the starting strain of E. coli. " This, surely, is a suggestion that in certain cases the one mutation, and then the other, appeared either "simultaenously", or very near in time to one another. Either way, this nearness of time means that the "odds" against this happening stochastically are huge; ergo, this is a "pre-programmed" response. PaV
Thanks Dave! HodorH: I could be mistaken, but I don’t think E. coli has any quorum sensing mechanism a la Vibrio or Staphylococcus species. You mean we just haven't discovered it yet. But that could be because either we aren't looking or haven't recognized it yet. Joseph
Hodor: The method I described is used in all kinds of tissue culture including bacterial. I happen to know of it because it is used in generating and maintaining pure asexual tissue strains in commercial mushroom cultivation. Even so genetic and epigenetic mutations accumulate after too many generations of reculturing pure tissue strains and senescence results. Storing the originally isolated strain with the desireable commercial characteristics in liquid nitrogen is the only way to keep it going ad infinitum but that's very expensive and beyond the means of small culture labs. They will thus typically then produce sexual variants via spores, isolate colonies produced from individual spores via the method described above, and then search through as many of those monocultures as required to find another commercially suitable pure (clonal) strain. DaveScot
I could be mistaken, but I don’t think E. coli has any quorum sensing mechanism a la Vibrio or Staphylococcus species. http://scholar.google.com/scholar?q=%22quorum+sensing%22+coli&hl=en&lr= At least some strains of e.coli have identical mechanisms and produce the same signal molecule. The common laboratory strain of e.coli has the gene to produce the signal molecule but isn't observed to express it however there are other signaling molecules used in quorum sensing that aren't identical to those in the species you mentioned. DaveScot
DS- I don't have access either. The method you described is common for isolating parasites, but bacteria are usually grown in culture from a single colony, which represents the descendents of a single bacterium. I'm headed out the door, maybe can discuss tonight. HodorH
HodorH: Creating and maintaining pure clonal cultures requires some extra steps when reculturing any given strain. Typically it's done by dilution in sterile water until there are few individuals left then inoculating a culture plate from the dilution then immediately reculturing individual colonies that bloom before they can merge. I don't have access to the paper or the methods used. Were there specific steps taken to insure that each individual culture was indeed started from a single bacterium or is that a presumption on your part? DaveScot
This requires “cellular intelligence” of some sort- that being intelligence to “recognize” it is alone along with the intelligence to do something about it
I could be mistaken, but I don't think E. coli has any quorum sensing mechanism a la Vibrio or Staphylococcus species. HodorH
Chris Hyland: The paper seems to suggest there were several hundred generations between the appearance of the RNA polymerase mutation and the glycerol kinase mutation. Thanks Chris and if that is the case that would end any alleged "simultaneous" implications. Joseph
HodorH: The problem with this is each culture is almost certainly clonal, that is, grown up from a single bacterium. So the information in one cell is the same as in the aggregate, until mutations start occurring. And mutations could start occurring at the first reproduction. This requires "cellular intelligence" of some sort- that being intelligence to "recognize" it is alone along with the intelligence to do something about it- reproduce with variation creating mutations. and even allow for swapping of genetic materials amongst the newly formed population. Joseph
devils http://www.google.com/search?hl=en&q=davescot+epigenetic+lamarck+site%3Auncommondescent.com DaveScot
JGuy The same article (DNA is Not Destiny, Discover Nov 2006, pg 32)came to mind when discussing surprises for Darwinists. It seems epigenetic changes would be an ideal way to allow for rapid adaptation to a changing environment that still allows for stability in the population over the long term as the changes are heritable but yet reversible. I don't know how these findings would apply to prokaryotes due to the differences in transcription but I am certain any return of Lamarckism in any degree is sure to stimulate some interesting research projects. Everything old is new again. devilsadvocate
My thinking is that perhaps the information required to cover all possible contingencies was just too vast to be covered by one genetic algorithm to fit into the comparment space (ie one E. Coli in this case) alotted. However if one splits up the responsibilities among the population, the space requirements are reduced and the goal (continued survival of the population) attained.
The problem with this is each culture is almost certainly clonal, that is, grown up from a single bacterium. So the information in one cell is the same as in the aggregate, until mutations start occurring. HodorH
The paper seems to suggest there were several hundred generations between the appearance of the RNA polymerase mutation and the glycerol kinase mutation. Although it's quite hard to tell at what specific time the mutation occured. Chris Hyland
To HodorH- My thinking is that perhaps the information required to cover all possible contingencies was just too vast to be covered by one genetic algorithm to fit into the comparment space (ie one E. Coli in this case) alotted. However if one splits up the responsibilities among the population, the space requirements are reduced and the goal (continued survival of the population) attained. And that different solutions were found, although many deaths occurred, demonstrates that at least some variety remains, which could also be a goal of the GA. A designer who "sees" the big picture would know this and should plan accordingly. Joseph
“The researchers report that mutations in both genes appeared together in several cases after six to eight days in the glycerol-based cultures, and E. coli cells containing both mutated genes grew 150 percent faster than the starting strain of E. coli.”
My apologies but nothing in that passage leads us to "simulataneous". Nothing in the linked article suggests it either. Does the Nature paper tell us how many generations occurred after the first bene (glycerol kinase gene) appeared until the second one appeared (RNA polymerase)? Is there any data that shows the RNA polymerase appeared first in any organism? Joseph
Sorry to ask such a fundamental question, but if such a gene-engineering toolkit referenced above exists, how would it be explained by RM+NS? It would seem just by logic that three components are required: 1. creation of the toolkit, 2. a mechanism to activate or trigger the toolkit, and possibly de-activate the toolkit at the end of the process, and 3. guidance criteria that would apply the toolkit actions to only situations that are beneficial and improve the chances of natural selection. Until all three are in place, nothing promotes natural selection. Are we back to an irreducible complexity situation? Ekstasis
HodorH: I can’t imagine how the information for directed change would be held in a population but not within the cells. Because you are not looking at this correctly. It isn't "information for directed change", rather it is information that will allow for the population to survive. That information would be spread throughout the population- amongst the individuals. That way all the "eggs" aren't in one basket. Variety/ diversity within the population would be the key to the population's survival. Look at this in a similar light as when evolutionists say that "populations evolve", even though the mutations occur in individuals and NS acts on individuals. And again the key is the population survives not necessarilly every individual within it. HodorH: Could you suggest something in particular? I did- "Evolving Inventions" SciAm Feb 2003. Given the resources and a goal, along with a cleverly written genetic algorithm, and there you have it. In this case the resources lie within the population and the goal is the survival of the population. Joseph
PaV, I think you should explain what you mean by simultaneous. The article certainly doesn't indicate that the mutations both occur in the same original cell. It is not particularly useful to tell someone to figure it out for themselves when asked for a clarification. HodorH
Chris Hyland:
Im curious as to what you mean by simultaneous, because if it means two mutations occuring simultaneously in the same organism I can’t see any evidence for that here.
From the article: "The researchers report that mutations in both genes appeared together in several cases after six to eight days in the glycerol-based cultures, and E. coli cells containing both mutated genes grew 150 percent faster than the starting strain of E. coli." It takes six to eight days for the kinase to appear. I'll let you figure out the rest. PaV
That wouldn’t necessarilly be true. It could be that the population is front-loaded with the necessary information for its survival- see comment 28
I can't imagine how the information for directed change would be held in a population but not within the cells. Could you suggest something in particular? HodorH
From the article:
"Opinion surveys indicate that many people don't believe that evolution occurs," said Herring, "but if the skeptics could witness evolution actually occurring, as we did, I think they'd be more likely to believe that it's not just a theory."
Either this guy doesn't understand the debate or he is blatantly misrepresenting what is being debated. ------------------------------------------------------------ HodorH: Well, then this would represent an opportunity for some real ID/frontloading research. If the cells are frontloaded, then all or most should experience the mutation upon replication. That wouldn't necessarilly be true. It could be that the population is front-loaded with the necessary information for its survival- see comment 28 Joseph
Patrick: I’m also hesitant to claim this change wasn’t due to NDE without knowing all the details involved. As Dr Behe has told us:
Intelligent design is a good explanation for a number of biochemical systems, but I should insert a word of caution. Intelligent design theory has to be seen in context: it does not try to explain everything. We live in a complex world where lots of different things can happen. When deciding how various rocks came to be shaped the way they are a geologist might consider a whole range of factors: rain, wind, the movement of glaciers, the activity of moss and lichens, volcanic action, nuclear explosions, asteroid impact, or the hand of a sculptor. The shape of one rock might have been determined primarily by one mechanism, the shape of another rock by another mechanism. Similarly, evolutionary biologists have recognized that a number of factors might have affected the development of life: common descent, natural selection, migration, population size, founder effects (effects that may be due to the limited number of organisms that begin a new species), genetic drift (spread of "neutral," nonselective mutations), gene flow (the incorporation of genes into a population from a separate population), linkage (occurrence of two genes on the same chromosome), and much more. The fact that some biochemical systems were designed by an intelligent agent does not mean that any of the other factors are not operative, common, or important.
Joseph
RobertC: There are two different genes involved, both found mutated in survivors. In different E. coli survivors, the genes have distinct(different) mutations. My wording was bad. The two genes are different but those two are the same two- always (survivors). That is what I meant. The two different genes are the same two affected but the mutations to those genes are different. OK great and thanks. Multiple but constrained solutions- that being any number of mutations could provide a solution but only two genes hold the key- in this case. It would be interesting to know if all the differences observed in the survivors- the different mutations- had the same effect, ie produced the same molecular product. That would be one way to design built-in solutions. IOW there are multiple ways to get the answer 42. Which again is exemplified by "Evolving Inventions". I remember my first programming class. The prof. didn't like my programs- the style was a bit unorthodox- but the bottom line was he couldn't argue with the results- I wrote them and they worked as specified. Unfortunately the only way to tell the design process I mentioned from RM&NS is by stepping back and seeing the big picture- that being if living organisms did not arise from non-living matter via some blind watchmaker-type process there would be no reason to infer all variation is due solely to those types of processes. Joseph
I'm also hesitant to claim this change wasn't due to NDE without knowing all the details involved. Many here have jumped straight to the assumption that CSI is involved without first using the EF. Changes, however striking, that appear by natural necessity and are therefore highly probable don't end up at the filter's design node. This find may be in the same category as that nylon-eating enzyme discussed a while back. Patrick
Joseph- re: "I don’t have access to the paper so could you please tell me if the the two genes were the same but the mutations to those genes different?" There are two different genes involved, both found mutated in survivors. In different E. coli survivors, the genes have distinct(different) mutations. RobertC
Robert C- I am partial to design solutions exemplified by the Feb 2003 SciAm article "Evolving Inventions". That would explain why the solutions differ and why some don't find the solution in time. However the goal in this experiment would be the survival of the population. And sometimes that occurs at the expense of a few or even many individuals. I don't have access to the paper so could you please tell me if the the two genes were the same but the mutations to those genes different? Joseph
@RobertC: I don't know the specifics of this search problem's combinatorial complexity, so I'm speaking generally. When we see a chance event occuring that shouldn't have enough replicational resources to get "lucky" enough, (the odds are stacked against us enough to not expect random success), and when this occurs over and over again, then we have evidence of non-random processes taking place. In this specific situation, maybe there were enough bacteria to randomly get one error needed to make the first enzyme, but when you also have the second one occuring, it makes the combined probablility out of the expectation of random chance guessing. (According to PaV.) It is then legitimate to look for non-chance processes, such as guided genome engineering. Atom
The article also makes a point about how we can witness change occuring in these bacteria in real time, "evolution" as the scientists interviewed referred to it. (Making a polemical point agianst those who don't accept NDE). Anyway, as another author once brought up, NDE advocates often say that evolution occurs much too slowly to witness directly, yet too fast to be preserved faithfully in the fossil record. (A convienient position, I'm sure.) But why should that be the case? With organisms like fruit flies and bacteria, that reproduce in large numbers quickly, why shouldn't we be able to view large scale change? What first principles of biology would make that a logical impossibility? The fact that we've yet to witness novel cell types, tissue types, organs and body plans develop is telling. If directional, CSI-rich change is really due to DNA copying errors, then the more copies, the more expected change, right? Why then don't we ever witness large-scale changes (cell types, tissue types, organs, body plans)? (Sorry mods if this is off topic, but it is a question that seems related to issues of telic vs. random change.) Atom
I suppose this proposed system would be sort of like the immune system, which has to adequately respond to all sorts of combinations in order to be effective. As such, a "solution set" might include multiple approaches that would end in the same or similar result. Patrick
Atom, Re: "if solutions are found much faster than we’d expect from random guessing" What would you expect from random guessing. How fast is evidence of design? This study lasted 44 days (660 generations), where 10^7 bacteria were passaged daily. From the few 'winners,' the authors conclude 1.8x10^-11 beneficial mutations are fixed per genome per generation. Is this quick? (I'm not being sarcastic, I honestly want a number placed on what would be a detection of non-random sampling. RobertC
@Robert, in regards to #1: Since we basically have a search problem, if solutions are found much faster than we'd expect from random guessing, we know that something non-random is occuring. When searching large spaces and finding very small targets within those spaces, consistently and quickly, we then have even greater suspicion to suspect the use of heuristics, rather than random sampling. And heuristic mechanisms are signs of design. (See Dembski's displacement argument to see why randomly finding a suitable heuristic is harder than simply finding your original target.) Atom
DaveScot, RE: "If the product is substantially different every time then the toolkit hypothesis isn’t supported" From the sequencing of the E. coli that grows fast on glycerol, the authors conclude "We identified 13 different de novo mutations in five different E. coli strains and monitored their fixation over a 44-d period of adaptation." Not all confer the same level of fitness-and most affect different amino acids in the kinase or the polymerase. It also reads that some of the mutations happen first, followed by the second-not simultaneously. Lastly, very few of the E. coli are surviors. I wonder if Eukaryotes are a better place to look for such a toolkit-multi-domain proteins, with recobination between them? RobertC
RobertC I don't have a subscription to Nature. If the product is substantially different every time then the toolkit hypothesis isn't supported. The hypothetical toolkit is something of a holy grail where biologically useful proteins are constructed from a finite subset of building blocks larger than individual amino acids but smaller than typical whole proteins. Since we don't know how to predict protein folding building blocks that produce the same major and minor biologically active sites needn't have remarkably similar sequences but still be components that produce the same topography in the final product. We might not be able to recognize it until we learn how to predict protein folding. Or it may not exist at all. I believe IBM has done more work than anyone else in looking for repeating patterns that could be a set of building blocks for genes. IIRC they haven't found anything significant in that regard. DaveScot
I'm fascinated by the idea of a "enzyme design toolkit," but I have a few questions. 1) If the mechanism of the "toolkit" is mutation at the DNA level, how is it distinguished from random mutation? 2) I read the Nature genetics paper. The commenters here seem to suggest the SAME modification happens each time. Acutally, there are an array of different mutations/modifications that make a 'faster' glycerol kinase, and more polymerase. Is this variety 'random' or a predicted by frontloading? 3) It seems like the solutions are not universal-as the mutant/modified outcompete the others. Why does the 'toolkit' fail to fire in some clones, but succeeds in others? RobertC
"It seems that two simultaneous mutations took place." Im curious as to what you mean by simultaneous, because if it means two mutations occuring simultaneously in the same organism I can't see any evidence for that here. Chris Hyland
Can anyone explain to me how this experiment proves evolution ?... Sladjo
Atom, I'm trying to figure out an ID research plan here, rather than simply making pronunciations of design. Perhaps the next time the Templeton Foundation comes around offering money, ID scientists will have be able to put forth some concrete ideas. We also don't know whether the mutations were the same in the replications of the experiment (though someone with full access to the paper could tell us). We only know that mutations occurred in two different genes, and often in conjunction.
But we repeated the experiment more than 50 times and mutations in the RNA polymerase gene appeared again and again." The researchers report that mutations in both genes appeared together in several cases after six to eight days in the glycerol-based cultures, and E. coli cells containing both mutated genes grew 150 percent faster than the starting strain of E. coli.
HodorH
HodorH, RM is just that: Random. This is statistically NOT random, since the same solution appears with the same side effect mutation. In other words, engineering on the cellular level, not random mutation, is what is responsible. Bacteria apparently have the ability to actively change their genomes to meet pressures, rather for waiting around passively for "random mutations" to occur (or not occur). Darwinists are surprised, because RM+NS (the bedrock of NDE) is not responsible for directed change. Something more telic apparently is. Atom
DS- If the solution is passed to the other cells then it would have to be on some form of mobile DNA. I don't have access to the paper, but it sounds like the changes were chromosomal. The passing on of the solution to others is readily testable, though. Just generate some "improved" bacteria and mark an unrelated gene so that its decendents are traceable, then drop it in a culture with regular bacteria under selection (high glycerol) conditions. Wait a couple days, test a bunch of clones for the traceable marker (as well as the presumably selectable mutations). If the information is passed, then most clones should NOT have the marker, whereas if it is strictly inherited, they will. Aside from that, except for extensive use of telic language, I don't really see how your scenario is distinguishable from normal mutation and selection scenarios. HodorH
Hodor If the cells are frontloaded, then all or most should experience the mutation upon replication. No. The cells are probably front-loaded with some sort of enzyme design toolkit. Under stress they all start working to build a new enzyme and when one of them succeeds it passes the solution along to the others. If the toolkits are all the same and the solution is something that can be built with it then it's just a matter of time before the solution emerges (or they all die before the first one of them has time to discover it). If there's only one solution possible and the order in which trial balloons are constructed is the same then the observed result should be the same enyzme emerging over and over again in approximately the same length of time. DaveScot
rrf The designer isn't intervening. Obviously since the same enzyme keeps "evolving" over and over again in a matter of days (the experiment was repeated 50 times) it isn't appearing de novo but exists in some form where its emergence is inevitable. That's not evolution. Evolution isn't supposed to be repeatable like that. If you didn't know that you really shouldn't be commenting here. DaveScot
It’s a “pre-programmed response” on the part of the organism.
Well, then this would represent an opportunity for some real ID/frontloading research. If the cells are frontloaded, then all or most should experience the mutation upon replication. Conversely, RM/NS would predict that the population of new and improved bacteria arose from a single mutant (or one mutant followed by another, since there are apparently 2 mutations). I think this is a doable experiment -- I'll think about how it would be done. HodorH
Benjy_Compson: "Would you please spell it out? " It's a "pre-programmed response" on the part of the organism. PaV
JFD: No, it seems the RNA polymerase is used to speed up the production of the glycerol kinase. The bacteria would die without the presence of some RNA polymerase. These two mutations worked in tandem to simultaneously produce the kinase and "extra" production of RNA polymerase. That's how I read the resume of the article. PaV
A guestion, When the bacteria was forced to mutate was the second mutation a response by the bacteria so the first mutation would not be destructive? Was the second mutation a way to preserve itself? JFD
When you said that UD has a better explanation for what is going on, I assume you mean that the RNA polymerase mutation was the result of Intelligent Design. Interesting! Do you have any thoughts why the Designer is intervening in what seems a routine experiment on E. Coli? This could be a really fruitful avenue of scientific research for the ID community. rrf
PaV: "I calculated that the amount of bacteria needed to provide TWO simultaneous point mutations simply through random chance would be so large that the known universe couldn’t contain it. " Perhaps you are already aware of this but there is experimental evidence to confirm your suspicions - see the link below and click onto 'What Can Evolution Really Do?' http://www2.uwsuper.edu/rseelke/index.htm Bottom line: under the RM&NS paradigm, evolution can only plausibly plod on one little step at a time. antg
We here at UD have a better idea about what’s going on.
Would you please spell it out? Benjy_Compson
It’s just a matter of time till this is spun and heralded as “proof” of evolution. Everything is proof of evolution – everything but the actual evidence. That’s just an inconvenience though. shaner74
Atom, "Self-genetic engineering." In light of the repeatability of this experiment, that does seem to be Ockham's explanation. bFast
Did anyone read the article on epigenetic effects in mice defied the paradigm of genetic mutation. I think in the latest or last months edition of Discover. It was a very interesting surprise that contradicted a deleterious mutation as the culprit of certain deleterious & heritable trait. JGuy
Given how repeatable this experiment apparently is, i'll be interested to see what is going on when they pick apart what is going on in more detail. jwrennie
Self-genetic engineering. Atom

Leave a Reply