More on Sherman and Front Loading
by BradfordMike Gene has written a piece pertinent to the blog entry Pondering Evolution titled A Hibernation Mechanism? Mike makes these observations:
As I see it, chapter 6 from The Design Matrix once again becomes important if one desires to extrapolate Sherman’s model/predictions to a case for teleological evolution/design. With such an extrapolation, the predictions that Sherman makes are more relevant to the realm of epistemological evidence than ontological evidence. Put simply, these predictions, if verified, would be revolutionary and might indeed convince hardcore skeptics of teleological evolution, but I don’t think that we should expect such things from the hypothesis of design through front-loading itself.
As I just explained, genetic information in lower taxons that would be functionally useless but become useful in higher taxons would amount to a pool of pseudogenes in lower taxons that would quickly be erased by mutations. And since developmental programs amount to genes and their regulatory units, these too would decay if they are not useful.
There is no reason to believe that useless genes would be preserved through natural selection. Unless…
Of course, a possible solution exists. Perhaps one could design a mechanism that would protect these useless genes from mutation until needed. Let us call this the Hibernation Mechanism. Yet there are serious problems with any genetic Hibernation Mechanism, designed to protect nucleotide sequences from changing over vast spans of deep time.
You can read Mike's thoughts about the "Hibernation Mechanism" at the linked site. Several commenters have discussed Mike's view of front loading. The potential evidentiary value of Sherman's model was alluded to as well. Mike concludes:
All in all, while Sherman’s model of front-loading leads to crisp predictions with potentially powerful evidence, the question to ask is whether such a model would be expected from a rational designer trying to front-load evolution. As I explain in The Design Matrix, a better strategy than fighting against mutations and the blind watchmaker is to exploit these processes and guide them so they serve your own ends. Or to put it another way, instead of looking for something that fatally wounds the Duck, look for ways that help the Rabbit hop.
[I should mention that if Sherman’s predictions are ever validated, I will acknowledge I was wrong with much celebration.]
August 24th, 2008 at 12:44 am
What a great article. wouldn’t it be great to see a dialogue between these two models of Frontloading? I’ll say it again, real contrast is the best way to flesh out predictions. This is what has been lacking so far with ID verses MET
A Hibernation mechanism could have selection advantages in the medium/ short term as well as the long term. It could also shield genetic information useful only in a seasonal drought or flood for example. In fact given the variability of the environment on earth it’s possible that long term survivability of any population depends on such a mechanism.
It could simply act as a switch the same way that genes are switched on an off as needed in the development of any organism.
The environment could be the trigger the same way that environmental stress triggers increased mutations in bacteria.
If the Hibernation Mechanism is useful in the medium/ short term any selection advantage would be wiped out at the first flood or drought.
I smell another prediction coming on and a possible way to identify the predicted mechanism.
Peace
Comment by fifth monarchy man — August 24, 2008 @ 12:44 am
August 24th, 2008 at 1:05 am
This might explain why ultraconserved DNA can be knocked out with no detrimental effects in the lab. Labs have no floods or droughts
Peace
Comment by fifth monarchy man — August 24, 2008 @ 1:05 am
August 24th, 2008 at 10:05 am
FMM,
Do you have a reference for this? I am not aware of the evidence that ultraconserved DNA can be deleted with no effect.
Comment by David E Levin — August 24, 2008 @ 10:05 am
August 24th, 2008 at 10:13 am
David:
That explains a lot
here
Comment by fifth monarchy man — August 24, 2008 @ 10:13 am
August 24th, 2008 at 11:09 am
FMM,
That's an interesting finding, certainly worthy of investigation. I will be interested see how the issue is resolved, particularly in light of the observation that these sequences appear to be under a high level of negative selection pressure. The two results do not seem to be in concordance.
Thanks.
Comment by David E Levin — August 24, 2008 @ 11:09 am
August 24th, 2008 at 11:53 am
David L.:
We talked about this research on the Open Thread last week. Deal is, the assumption that ultraconserved elements are "ultraselected" is NOT an observation at all. It's a prediction from NDS that has been falsified for at least four ultraconserved elements of more than 200 bp.
Best quote from press about this experiment:
Comment by Joy — August 24, 2008 @ 11:53 am
August 24th, 2008 at 12:15 pm
Joy,
It's not an assumption. Analysis of the population distribution of mutations that do arise in these sequences reveals that they ARE under strong selection. This was the conclusion of the study, which was, in my view, supported by the data. Did you not read the paper?
It certainly has not been falsified for even those elements. For the moment, functions have not been identified. This does not mean that functions will not be revealed for them in the future. We have seen this happen many times. Just because the mice appear to be normal in a laboratory setting, doesn't mean that they do not have defects.
Seems to me that you have jumped to the conclusion you like, despite the conflicting data. The way I left it, I would like to see how this apparent conflict of evidence is resolved.
Comment by David E Levin — August 24, 2008 @ 12:15 pm
August 24th, 2008 at 12:25 pm
The distribution of of the allele frequency spectrum indicates that the conserved sequences are under negative selection. A mechanism that repairs the genome before it enters the population does not explain why allelic variations exist in the population, but don't become fixed. Any new scientific understanding has to be able to explain *both* findings.
Even if the sequences are not required for viability, that doesn't mean they aren't under selection. The authors of Deletion of Ultraconserved Elements Yields Viable Mice suggest many possible explanations for their results, such as regulatory element redundancy.
Comment by Zachriel — August 24, 2008 @ 12:25 pm
August 24th, 2008 at 12:48 pm
Zachriel,
I agree with you, but functional redundancy cannot explain the conservation. If there is another sequence that confers the same function, this would relieve functional selection on the one being maintained. There has to be another answer.
Comment by David E Levin — August 24, 2008 @ 12:48 pm
August 24th, 2008 at 2:47 pm
David and Zach
It depends on your paradigm in mine the two results are exactly what you’d expect and match prefectly they even sugest where to look for clues to a Hibernation mechanism.
I predict a mechanism that allows normal variations in this DNA but doesn’t allow them to become fixed in a population under normal circumstances.
Such a thing is what you’d expect from a frontloader who wants’ to use evolution to accomplish his goal but also insure a certain outcome in the event of rogue mutations .
It would function like a reset button always bringing you back when you veer off track.
I’ve got mine. Let's hear your prediction. Come on go out on a limb what answer would MET lead you to expect?
Dueling predictions, sound's like science to me.
Peace
Comment by fifth monarchy man — August 24, 2008 @ 2:47 pm
August 24th, 2008 at 3:06 pm
Way to go Fifth Monarchy Man.
Let's do science!
Comment by Thought Provoker — August 24, 2008 @ 3:06 pm
August 24th, 2008 at 3:06 pm
FMM,
The evolutionary prediction is that they are functional, that their functions will evenutally be understood, and that they are under strong negative selection.
Comment by David E Levin — August 24, 2008 @ 3:06 pm
August 24th, 2008 at 3:31 pm
fmm makes predictions:
It's good scientific practice to logically or even mathematically derive predictions from a theory. In contrast, your predictions seem to come out of thin air. Could you please give us some insight in your method of derivation?
It seems you're claiming that the ultraconserved DNA currently has no function, but is "hibernating" until it is "needed" in the future, by the descendants of all creatures (mice and men included) that currently have that DNA. Is that correct? If so, any idea what sort of future event that might be, and/or when exactly it will happen? It's not rapture-DNA by any chance is it?
Comment by Raevmo — August 24, 2008 @ 3:31 pm
August 24th, 2008 at 3:39 pm
Redundancy can provide more of some trait, so it may be selectable in that way, such as by providing robustness in stressful environments (which might not show in lab mice). Or imperfect redundancy can provide a library of modified functions as a kind of averaging of past environmental conditions (which again, might not show in lab mice).
Perfect redundancy tends to be unstable, so redundant genes tend to evolve asymmetrically. Models suggest that small populations, large genomes, fast mutators, and developmental genes tend to evolve greater redundancy.
Nowak, et al., Evolution of genetic redundancy, Nature, July 1997, pg. 167.
I'm not trying to salvage the original claim, just pointing out that redundancy may not always be invisible to selection, especially in developmental genes.
Comment by Zachriel — August 24, 2008 @ 3:39 pm
August 24th, 2008 at 3:53 pm
So you're saying that someone would be born with a variant of the conserved gene, then their children would be restored to the original version. Is that correct? Have you attempted to verify this somehow? And if so, then where does the information come from?
Comment by Zachriel — August 24, 2008 @ 3:53 pm
August 24th, 2008 at 4:45 pm
Well, it seems like Mike Gene's attempted clarification has been completely lost on FMM who has latched on to this whole hibernation idea despite all the obvious problems that Mike pointed out. Mike's post is an excellent example of why I respect him even though we disagree. He doesn't tend to blatantly ignore the evidence or dismiss reason like many (most?) ID supporters.
Comment by Todd Berkebile — August 24, 2008 @ 4:45 pm
August 24th, 2008 at 4:49 pm
I agree with Zachriel. It is hardly ever the case that duplicated or redundant genes perform exactly the same function (in quantitative terms). I can imagine a scenario in which redundant genes are co-expressed with different expression levels and that conditions (selection) enforce a mean or norm expression level involving (the sum-product of) both genes. I.e., selection does not somehow “set aside” one gene and target the other, but targets the coordinate expression of both.
Simply saying that “ultra-conserved” elements are “ultra-selected” does not suffice for scientific purposes. One must identify the function and the conditions under which it is selected. If that can’t be done in the lab, then its off into the wild and wide open outdoors for the researchers. (Actually those pasty-faced geeks never have to see the sun and could stay in the lab and simply vary conditions until they actually find a selection regime that maintains the genes.)
True redundancy seems to be a strategy hardly ever implemented in life forms. IDers should have a theory for that, because that’s so unlike designed systems.
Comment by Rock — August 24, 2008 @ 4:49 pm
August 24th, 2008 at 6:19 pm
Raevmo
Sure. I put myself in the place of the frontloader. How would I ensure that my desired outcome came to pass despite the random nature of evolution? A hibernation mechanism to protect vital information seems like an obvious choice.
Zach:
Not exactly I am I saying that someone born with the variant will be less likely to pass this variant to his children due to the mechanism.
Of course not I’m just shooting from the hip.
I think you could select a variant population and segregate it for a period of time to see what would happen.
Then you could introduce individuals with the original gene into the variant population to see what happens.
Then finally I would introduce the variant into to the general population and see if the original gene eventually replaces the variant.
lots of research could be done here.
What information are you referring to? If you mean the original gene it would come from the LCA and eventually the designer.
Todd:
Not at all, I merely pointed out a way that I believe hibernation mechanism could work despite the problems Mike pointed out.
If the mechanism is beneficial in short/medium time scales this seems to insure it will carry forward to long time scales when it will become vital.
Peace
Comment by fifth monarchy man — August 24, 2008 @ 6:19 pm
August 24th, 2008 at 6:30 pm
Raevmo:
No I’m saying that ultraconcerved DNA can give us clues to the mechanism that was used to frontload information in the past. It would be a sort of left over scaffolding in the genome.
I don’t expect any future event but there is always that possibility I guess. It depends on the desired outcome the designer wanted to produce.
Comment by fifth monarchy man — August 24, 2008 @ 6:30 pm
August 24th, 2008 at 6:38 pm
Oh, I understand now. You're proposing removal of variants by natural selection. What a novel idea.
Comment by David E Levin — August 24, 2008 @ 6:38 pm
August 24th, 2008 at 6:52 pm
David:
Not at all. In order for something to be selected “naturally” it must have function. A hibernation mechanism has no such requirement.
The ultraconserved stuff we are talking about appears to have no function at all yet it looks as if it is ultraslected.
That is exactly what a mechanism like I predict would do.
Peace
Comment by fifth monarchy man — August 24, 2008 @ 6:52 pm
August 24th, 2008 at 7:23 pm
FMM,
Call it what you will, but what you are describing is a decrease in reproductive fitness, something that the mouse mutants were tested for. They had no difficulty in transmitting the mutated allele (in this case a deletion) to their progeny. Care to revise your prediction?
One more thing–are you now arguing that this hybernation mechanism operates outside of nature? I mean, if the wild-type allele is not being selected for naturally, is it being selected for supernaturally (i.e. God intervenes when there is a mutation to make sure it doesn't get through to the next generation)? If the envisioned mechanism works through natural means, then why would this not be a form of natural selection?
You see, this making predictions business is not as easy as you might think. As Raevmo indicated, scientific predictions must have some logical and mechanistic underpinnings, otherwise they are just unsupported shots from the hip.
Comment by David E Levin — August 24, 2008 @ 7:23 pm
August 24th, 2008 at 9:04 pm
David:
I'm not saying the variant parent is less likely to reproduce than its normal counterpart I’m saying that it is less likely to pass on its variant genes in the same way that a cancer sufferer does not pass on his mutations to his children.
Of course they did if they didn’t it would be evidence of selection.
A hibernation mechanism on the other hand would look like the gene was selected with no evidence it happened. That by the way is what is exactly what we see.
Natural selection has a precise definition it does not mean anything that happens in nature. Give me a break.
It’s no wonder MET has gotten so complex with one epicycle after another if it simply means anything that happens in nature.
Peace
Comment by fifth monarchy man — August 24, 2008 @ 9:04 pm
August 24th, 2008 at 9:51 pm
Are you saying that descendents always have exact copies of the ultraconserved regions?
Comment by Zachriel — August 24, 2008 @ 9:51 pm
August 25th, 2008 at 2:31 am
Slightly off-topic question to all (Raevmo and Zach especially, I presume):
I was reading Mike's responses to questions about frontloading and had this question come to mind regarding his belief that gene duplication results in increased genetic information.
Since his withdrawal from this blog I presume he doesn't want to engage in this process now, so I bring my question to everyone here.
As you know I'm not a biologist so I presume that there is a simple answer to assuage my curiosity. This off-topic question, I would guess, will be laid to rest with a reply or two.
My starting presumption:
Protein production is regulated, its folding is supervised and chaperoned, misfolded proteins are destroyed, etc. There is an entire network built-in from regulating the gene that results in the protein to the protein's manufacture and form, to its use, timing and placement within or outside of the cell. There are trucks and tracks arranged for transport and labels attached for destinations as well as schedules to keep.
If a gene is kicking out copies of a protein already regulated for wouldn't that gene be, in all probability, selected against as it is wasting cell energy creating an excess of proteins which are not needed or that require additional supervision and/or destruction? Since gene regulation is so important wouldn't one or the other of the genes at least be silenced and, therefore, unlikely, even if later altered, to be able to express a new protein? Perhaps this is already included in the duplication hypothesis - is it presumed that the duplicate gene is silenced?
If so, why would it be regulated for, or turned on, at the specific time that its mutation would allow for it to create a novel protein (ie: increase the genome's information content)?
And if the new protein is to be somehow useful to the organism wouldn't it have to come with its own set of regulators, some already-prescribed use and some existing need that needs be filled? If not, wouldn't it just lie in the cell as waste and cause the new gene to be selected against?
Comment by Pez — August 25, 2008 @ 2:31 am
August 25th, 2008 at 5:26 am
No, I'm saying that in the wild (in the presence of the original gene)a variant parent will be less likley to pass his variant gene to his offspring.
Comment by fifth monarchy man — August 25, 2008 @ 5:26 am
August 25th, 2008 at 7:37 am
The clarifications Mike made over at his blog were very helpful. The idea of Front-Loaded Evolution is becoming more and more clearer to me reading his three 'Misunderstandings about front loading' posts.
Its good that Mike differentiated his version of FLE from Sherman's hypothesis which could be derived from the idea of Front-Loading. Though Sherman did not explicitly derive it in his paper, his hypothesis is not in my view too far from FLE. If Sherman's hypothesis was somehow experimentally falsified in the future, the critics here are gonna go crazy over it. Then we will be able to calm them down and explain the difference. Sherman's ideas don't flow from Mike's assumptions about the Design of evolving life.
This is one of the wise features in Mike's FLE imo. The 'cytosine-de-amination-leading-to-hydrophobic-amino-acid-substitution' comes to mind as the observation that is resulting from this assumption.
Isn't it refreshing to think about Life in a new way? Personally, it is kinda releasing–from being trapped in to one particular framework. A breath of fresh air indeed.
Thinking about the amount of "distinguishing" ideas this Front-Loading could generate, one better take this concept of Front-Loading seriously.
Comment by samsen — August 25, 2008 @ 7:37 am
August 25th, 2008 at 7:48 am
FMM
So, the mutation doesn't get into the germ line, it's just somatic? Perhaps the germline DNA of those individuals in which mutations were detected (presumably from blood cells) should be tested for their mutations.
No, they didn't. Read the mouse knock-out paper. They had no difficulty passing the mutation on to their progeny. They construct the homozygous mutants by mating heterozygotes. They would see a decreased frequency of mutant allele transmission in the progeny if it was there.
Comment by David E Levin — August 25, 2008 @ 7:48 am
August 25th, 2008 at 4:58 pm
David:
Are you deliberately misunderstanding me or am I not making my self clear? what part of
"No, I'm saying that in the wild (in the presence of the original gene)a variant parent will be less likley to pass his variant gene to his offspring."
Is unclear to you. I can imagine lots of way a mechanism could accomplish this
Perhaps when a sperm and egg fuse the original gene is given a privileged position over a variant one
Perhaps organisms with the original gene release a pheromone that renders those with the variant slightly less attractive to the opposite sex.
Perhaps mutations that bring the gene back toward the original pattern are given added weight in the variant population.
This stuff is not hard but you have to come at it with an open mind and not just try to defend your paradigm at all costs.
Are you even listening?
I predict the mechanism that works in the presence of organisms with the original gene. That is the best way to protect against rogue mutations in a population IMHO. The study you mentioned is anything but random. It did not even place the variant side by side with the originals as would happen in the wild?
If a mechanism worked when a variant population was segregated its effects would be indistinguishable from selection and a hibernation mechanism would not be needed.
Peace
Comment by fifth monarchy man — August 25, 2008 @ 4:58 pm
August 25th, 2008 at 8:22 pm
Your statement reads like a definition of Natural Selection. You even provide a darwinian mechanism.
Comment by Zachriel — August 25, 2008 @ 8:22 pm
August 25th, 2008 at 9:14 pm
Sometimes genes, chromosomes or entire genomes are duplicated. This incurs a cost, of course, but can also provide a benefit because extra RNA or protein is provided. Analysis of various genomes indicates that 20-50% of genomes in everything from bacteria to humans are duplicated genes. The proportion is probably higher because some sequences have diverged so much as to be unrecognizable.
If a new gene is functionally redundant and has no selective benefit, then it has only a small probability of fixation (1/2N, N=population size), and may take a long time to reach fixation (4N generations). Even when fixed, as David E Levin mentioned, redundant genes may not provide any selectable advantage, so (typically within a few million years) they often mutate to become pseudogenes, and may eventually be deleted from the genome.
One way redundant genes become stable in a population is if each gene specializes to different aspects of the same function (which can lead to irreducibly complex systems). Sometimes, a related function may evolve. And rarely, a duplicate gene may evolve an entirely new function.
Comment by Zachriel — August 25, 2008 @ 9:14 pm
August 26th, 2008 at 12:04 am
David,
Thanks for continuing to humour this line of questioning.
On your 9:15 answer:
I'm sorry to display my ignorance so transparently, but jargon is just empty to me. Is this functional specialization of the regulatory process? In what way does the function of the gene expression regulation specialize?
For this we await a random copying error, I presume. The mutation above is not controlled with any purpose of either silencing the gene or of increasing adaptability. By that I mean that there was no built-in response programmed and that the cell would have continued expressing this gene redundantly until the gene naturally mutated to either non or new function. This against normal error-correction and the normal mutation rate.
Do we actually have many examples of genomes in which we have two identical genes, both redundantly expressed? If not, is it not evident that the redundancy is actually negatively adaptive? Either its mutation to new function or to a pseudogene status must be more adaptive if we don't have a preponderance of such examples.
Does this sentence refer to the previous one in which you said the gene itself either mutates to non or new function?
In other words, is this part of a mechanism coordinating the mutation rate of that gene?
Or is this mutation to the promoter a serendipitous mutation in the regulation system of the original gene that also allows for regulation of the duplicate?
Or none of the above?
In response to my question to Steve you replied:
Is that obvious reason the mere propensity for copying errors? So redundancy will be maintained until normal mutation vs. error correction results in a silenced gene. This regardless of the preexisting regulation of the original gene.
Again, does this not evidence the negative selectability of the duplicated gene?
Also, you offer the fact that the gene will mutate to a new and additional function. This, of course, is the cause of the increase of information in the genome. As far as this new function goes, this returns us to my other question. How does this new function find expression when its product is a novel protein This protein needs to be regulated, folded, and transported to some place where it will have a valid function. At the same time the cell, which normally would tend to destroy misfolded proteins must recognize this as functional and let it persist. Do these features not all have to come to be at the same time as well? Or can this new protein be folded by the already-present protein-folding machine?
Comment by Pez — August 26, 2008 @ 12:04 am
August 26th, 2008 at 12:06 am
Hi Steve,
My thanks again.
I've got no problem with the fact that duplication happens and that it is not immediately deleterious, but what I'm wondering is why? With the importance of regulation and correction it seems theoretically odd to me that it not be.
Also, as I noteed above to David, doesn't the prevalence of pseudogenes argue for the negative pressure on sheer duplication?
I can buy that. The mere wastefulness is not necessarily enough to cause negative selective pressures - but what about the fact that it would seem to be a defeat the purpose of expression controls?
If we accept, as you certainly make the case for, that expressed duplicates are not negatively impacting the organism we can presume that there is no need to shut off the duplicated gene.
Now, the question becomes, what happens when it mutates? Presuming, as we ought, that a peptide chain can be built in translation, how do we account for its folding, it surviving correction, its transport and utility? How can the cell, or be expected to have a way to put this new protein to use or to know what to do with it in the event it has to ship it off to some other tissue where it might have some novel use?
Comment by Pez — August 26, 2008 @ 12:06 am
August 26th, 2008 at 12:09 am
Here's what remains of my original question, as far as I can tell…
The answer is "no".
We are aware that duplication happens and that duplicated genes do not seem to be necessarily selected against merely for the fact that they are producing an excess of proteins.
Is this not theoretically odd?
No, it is not.
So that brings me to the last question in that comment:
Speaking of redundancy, I've repeated this portion in my last two responses.
Comment by Pez — August 26, 2008 @ 12:09 am
August 26th, 2008 at 12:17 am
Hi Zach,
Thanks for your input.
You are saying here that for a long time they are functioning as duplicates. Does this answer my questions above in the affirmative - we do, in fact, have lots of examples of two identical genes perfoming the exact same function?
Does this then require each gene to mutate simulataneoulsy and beneficially to take on only a portion of the job each could previously handle alone? Is this not wasteful and also highly improbable?
And here how does theory account for the remaining portion of my question in the above comment?
Comment by Pez — August 26, 2008 @ 12:17 am
August 26th, 2008 at 12:20 am
I'm not sure what I will do next here.
I think I ought to leave the issue lie and go away happy with my new information.
I guess that depends upon what responses come next, but as I say, I don't want to harp on this too long.
Thanks to all for your expertise.
Comment by Pez — August 26, 2008 @ 12:20 am
August 26th, 2008 at 1:01 am
Hi Pez. I suspect some of the questions you have in mind have not been answered. You've opened a can of worms. The responses you've gotten have been informative but there is more. If gene y is alleged to have resulted from a duplication of gene x some questions arise: Some of them are:
* What does a comparative analysis of enhancers and enhancer sites show?
* What is the nature of the signaling pathways that determine whether the genes are transcribed?
* Identify relevant transcription factors.
Differences between genes x and y and the elements involved in transcription regulation should give one an idea as to what changes would have taken place. The historic transition details are important.
Comment by Bradford — August 26, 2008 @ 1:01 am
August 26th, 2008 at 3:42 am
Hi Bradford,
I don't know where this comment went the first time - hopefully it doesn't return now.
They haven't been. But that's ok. Maybe I can't formulate them properly.
I certainly suspect so. But we've all discussed things with people who issue challenge after challenge without regard for the answers they're receiving, so I am trying to hear the answers before reiterating the things I don't think are quite being answered. It's especially irritating when those questions are coming from someone who, like me, doesn't really know what he's talking about.
I just see a problem with the logic of this problem but I may have gotten all I can get out this line already … we'll see.
Comment by Pez — August 26, 2008 @ 3:42 am
August 26th, 2008 at 7:19 am
Zach:
Except the ultraconserved gene in question has no function!!!
You can't select junk DNA. The hibernation mechanism would act like selection (in this case ultra selection) yet it would protect genes that are useless at present.
With a Hibernation Mechanism a frontloader would get all the conserving benefits of natural selection with out having to insure that every gene is expressed and functional all the time.
Peace
Comment by fifth monarchy man — August 26, 2008 @ 7:19 am
August 26th, 2008 at 8:18 am
Redundancy is very common in nature, including on the genetic level.
It's incremental. One daughter happens to step left (let's say increasing its binding capability). The other 'discovers' that a step right is beneficial. Eventually, through repeated rounds of selection, they are at the functional poles. (Perhaps one binds and the other disposes, while the original gene did both.) Then other helpers are added, leading to a cascade that does far more, far more efficiently, than the original gene alone.
The vast majority of mutations are not going to be beneficial, and many won't fold into recognizable protein domains. But some will.
What's an excess? It's just another characteristic subject to selection.
Many duplicates will be expressed. Sometimes entire chromosomes or genomes are duplicated! But if the gene is never expressed, then it will be just another pseudogene destined to melt away. Many duplicates are not going to be beneficial, and even those with some slight benefit may not become fixed in the population.
Comment by Zachriel — August 26, 2008 @ 8:18 am
August 26th, 2008 at 8:27 am
Umm, you just said it might create a sexual pheromone.
Much of biology is about sex and sexual attraction. The birds and the bees.
Comment by Zachriel — August 26, 2008 @ 8:27 am
August 26th, 2008 at 9:10 am
Pez, if you want to look at an actual gene duplication claim try this:
http://genome.cshlp.org/cgi/co...
Comment by Bradford — August 26, 2008 @ 9:10 am
August 26th, 2008 at 10:29 am
Pez, it seems to me that there are two overall themes to the discussion of gene duplication here. One is your proposal that duplicated genes are inherently wasteful, and that their retention in a genome is theoretically odd. The second is centered on fates and functions of duplicated genes in various contexts.
Let me just address the first theme, from my own perspective. I don't look at biology from the point of view of an engineer, judging its excellence or lack thereof from a standpoint of "efficiency" or "elegance" or "design." I'm not saying that none of those things is present in biological systems, or that I'm not impressed by the design in biology, I'm just saying that I don't start with the presumption that a system which exhibits inefficiency or kludginess is unexpected.
I think this is a weakness of strong design-based thinking, and I think it has created confusion on your part. The presence of a second copy of a gene seems "wasteful" to you, and so you are surprised that duplicates are retained in genomes, asking if the retention is "theoretically odd" solely on the basis of your judgment regarding "wastefulness." But the retention of many duplicates is not theoretically odd at all, at least because no useful theory begins with the assumption of extreme fine-tuning of cellular metabolism, and especially not of eukaryotic genomes. I would suggest that you have overemphasized the importance of "efficiency," and I suspect this is because the perspective of ID lends itself to such imbalances in consideration of biological phenomena.
Also, you asked whether a new gene would need to have its own regulators in order to be useful. This depends on how the new copy gets used. Sometimes the usefulness of the extra copy is precisely its boosting of expression of the protein (an example here is human amylase). Sometimes the usefulness of the extra copy is in regulating the activity of the first copy, such that the regulation of the expression of the two different genes will be very similar. An example of this is summarized on my blog; the idea is that the duplication relieved an adaptive road block in evolution.
Finally, I note that while gene duplication does not necessarily lead to selectable "wastefulness," there are lineages where it does appear that control of genome size has been a strong evolutionary influence. Birds are a famous example, and Zachriel mentioned bacteria in a comment on the other thread. On the flip side, some lineages appear not to care at all about excess DNA – plants are content with multiple copies of the entire genome, and some amphibia carry around 40 times as much DNA as humans do. I conclude that "efficiency" on the genomic level can be an evolutionary influence, but that it is a mistake to simplistically overemphasize it.
Comment by SteveMatheson — August 26, 2008 @ 10:29 am
August 26th, 2008 at 8:54 pm
Zach
No it's the hibernation mechanism that might create a sexual pheromone not the gene it conserves. The gene itself is with out function hence the need for a mechanism to conserve it. Is this really unclear?
All the more reason for a frontloader to harness it to protect unexpressed genes.
why is it always like pulling teeth with you?
Peace
Comment by fifth monarchy man — August 26, 2008 @ 8:54 pm
August 27th, 2008 at 1:05 am
I guess its spam.
Hi mods, there's a frontloading link in the filter.
Comment by Pez — August 27, 2008 @ 1:05 am
August 27th, 2008 at 7:07 am
So, the hibernation gene detects changes in the conserved region and generates a selectable change in the organism.
The hibernation/conserved system is therefore subject to natural selection. Meanwhile, the hibernation gene should tend to evolve over time to decouple from the conserved region, or be deleted entirely. Neither it nor the organism requires the conserved region to generate a pheromone.
There's nothing you have suggested that would tend to maintain the conserved region in the face of selection. The evidence is consistent with what we would expect if ultraconserved regions are subjected to ultraselection. Of course, you're admittedly just speculating. If you find any evidence of some alternative mechanism, please let us know.
Comment by Zachriel — August 27, 2008 @ 7:07 am
August 27th, 2008 at 9:15 pm
Zach:
Yes this is Mike’s problem with the prediction. My response would be that this could be overcome if such a mechanism were beneficial in the medium /short term. For example if the hibernation mechanism also conserved genes that were helpful in the event of a climatic event like a unusual flood or drought.
Yes and the mechanism would tend to keep that from happening producing results much like we see in the ultraconserved genes
But if the two are linked like a lock and key any mutation would tend to disable the mechanism long before it evolved the ability to produce a pheromone on its own. And if no pheromone is produced the organism functions normally.
I’ll grant that such a thing is possible . But that does not hold for the other possible mechanisms I mentioned.
Do you agree?
No That prediction has been falsified. In order for there to be ultra selection a gene must perform some vital function. Knocking out the ultraconserved gens had no effect on the mice.
As David said “there must be another answer”.
I have got mine. What’s you prediction?
I will.
In the mean time I’ll be happy just knowing that you won’t be able to say that ID makes no distinguishing predictions anymore.
Peace
Comment by fifth monarchy man — August 27, 2008 @ 9:15 pm
August 28th, 2008 at 7:07 am
Zach:
After further thought I need to retract my previous concession.
A hypothetical pheromone produced by the Hibernation Mechanism in response to variation would make an organism less sexually attractive. So if the HM or the gene itself evolved to produce the pheromone on it’s own it would be selected out.
Therefore it looks like if a HM like I describe was frontloaded into a genome it would tend to remain there and functioning, all things being equal.
Peace
Comment by fifth monarchy man — August 28, 2008 @ 7:07 am
August 28th, 2008 at 10:23 am
This was your original proposal.
Those with the original gene put a "Kick Me" sign on their rivals so they never get the girl. Either way, it's a selectable trait.
On the other hand, if the Hibernation Mechanism produces a pheromone that makes the same organism less attractive if its a variant, it would be a selectable and detrimental trait. This is the conservation mechanism.
But…
The pheromone, the response in potential mates, and the relationship (which would tend to become decoupled) between the conserved region and the Hibernation Mechanism (which being detrimental to the organism would tend to stop working) are all subject to evolutionary change. You have proposed nothing that would tend to maintain the conserved region over geological time.
Perhaps there is some other type of Hibernation Mechanism, something more akin to gene conversion. But this is also inadequate.
It had no discernable effect on the mice. That doesn't mean it didn't lead to a selectable difference in the wild.
No, you have a speculation, not an answer. It's not an answer unless you can provide the data to support your speculation.
That there are selectable difference in the mice that were not apparent in the study. However, I am not wed to that concept and am open to other possibilities.
Comment by Zachriel — August 28, 2008 @ 10:23 am
August 30th, 2008 at 8:36 am
Zach:
Actually it’s those with the variant that have the kick me sign and it's put on them by their own HM.
Perhaps I have not made my self clear enough on this point.
The HM could conserve information that would become vital in a periodic climatic event. Like an Ice age or prolonged drought.
In such a case any advantage an organism had lost due to the HM would be regained over night. Leaving the original gene and it’s conserver dominant once again in a population.
Because the climate of earth is so variable the HM would be constantly needed and tend not to degrade. As to exactly what information this would be I can once again I can only speculate but the possibilities are immense.
I would even say that we could look at organisms that evolve rapidly in response to climate fluctuations only to return to their original state after the crisis for clues.
More possible research flowing from a spur of the moment prediction on an internet blog.
Science is cool
Is there a specific reason you are calling it speculation and a not prediction?
I agree we need to do research that’s what you do in science.
A selectable difference does not lead to ultraselection. For ultra selection you need an ultraselectable difference. Is this your prediction?
Peace
Comment by fifth monarchy man — August 30, 2008 @ 8:36 am
August 30th, 2008 at 9:05 am
That's not quite what you said. I appreciate the clarification. But again, there is nothing to stop a mutation from disabling the mechanism itself so that it has the beneficial function and not the detrimental function.
Now, you are positing that the ultraconserved regions are conserved because they are historically selected.
Yes, evolution can keep various genes that were previously useful around for long periods of time. But eventually they drift and degrade unless selected. We have significant data on this, and it doesn't appear to support such long-term storage without evolutionary drift, or even progressive adaptation.
But it is unstable in so many different ways. That which might be useful in one environment might not be useful in another. And a designer would have no way to predict the long-term environment an organism might find itself in.
I said it was a "speculation" as opposed to an "answer". As to it being a prediction, it's rather vague. If you could tell us what observation to make, then you would have a solid prediction. There's nothing wrong with proposing ideas.
But you really need to look at your concept *critically*. Do you understand how evolution works, and why your idea appears contrary to the evidence? It's important in science that you try to determine and answer objections, and not rehash stale arguments without significant new evidence or insights. Consider it a courtesy.
An ultraselectable difference just has to be something that exerts consistent selective pressure. It doesn't have to mean the organism is not viable. Though as you suggest above, it might mean strong but intermittent selection.
I would suspect that there are selectable difference in the mice that were not apparent in the study, as indicated in other studies. However, I am not wed to that concept and am open to other possibilities.
Comment by Zachriel — August 30, 2008 @ 9:05 am
August 30th, 2008 at 3:37 pm
Hi Fifth Monarchy Man and Zachriel,
I think your discussion has been exemplary of a positive ID discussion.
Underlying the discussion is defining distinctions.
To me, it looks like you both are agreeing to the possibility to subtle selections in deep time.
One view (Zachriel’s?) would suggest it is just more supportive evidence of MET.
However, another view (FMM’s?) if it is too subtle it becomes supportive of front loading.
Allow me to suggest the distinction between the two views is foresight. If the subtle, deep time selection pressures are too subtle and too deep it may suggest being influenced by a future need.
Of course, I suggest a direct link to the future need is via interconnect quantum effects. Slight pressures from the future influence the quantum effects in superposition to “randomly” choose a path that provides consistency in the universe.
Whether the foresight comes indirectly from an intelligence or directly from orchestrated Quantum Mechanics, I suggest the distinction of these views from MET is foresight.
I hope this comment is helpful to your discussion.
Comment by Thought Provoker — August 30, 2008 @ 3:37 pm