Science News: Mutations, viruses, and brains
by MikeGeneWe often think of mutations as "mistakes." But in my opinion, the real Mistake would exist if feedback and adjustment were not made possible:
Intestinal colonization of germ-free mice by E. coli was followed by the very rapid selection of bacteria carrying mutations in a master regulator that controls and coordinates the expression of over 100 target genes. The important selective advantage conferred by the mutations was related with their additive and independent effects on genes regulating bacterial motility and permeability.
These results suggest that global regulators may have evolved to coordinate physiological activities necessary for adaptation to complex environments and that mutations offer a complementary genetic mechanism to adjust the scale of the physiological regulation controlled by these regulators in distinct environments.
-Here
Viruses can seduce us into thinking that life can be simple. But viruses exist and propagate only because cellulary machinery exist:
In some ways, HIV resembles a minimalist painter, using a few basic components to achieve dramatic effects. The virus contains just nine genes encoding 15 proteins, which wreak havoc on the human immune system. But this bare bones approach could have a fatal flaw. Lacking robust machinery, HIV hijacks human proteins to propagate, and these might represent powerful therapeutic targets.
Using a technique called RNA interference to screen thousands of genes, Harvard Medical School researchers have now identified 273 human proteins required for HIV propagation. The vast majority had not been connected to the virus by previous studies.
-Here
Life is multi-tasking. Check out what helps the developing human brain:
Synapse elimination occurs during the normal development of a child's brain, but until now, no one knew how certain synapses were flagged for removal. "We have identified the long-mysterious mechanism by which excess synapses are sculpted away in the developing brain," said the study's senior author, Ben Barres, in the journal Cell.
Barres' team found that the brain-sculpting process was controlled by a component of the immune system known as the classical complement cascade. The complement cascade is one part of the multipronged attack the immune system launches throughout the body when it detects a foreign invader.
-Here
And let's not forget culture. Brains are different.
January 12th, 2008 at 2:31 pm
Suggesting downward causation as the correct lense through which to view function.
Mike Gene:
Reminds me of exchanges over self-correcting genomic mechanisms and arguments that viruses show that correction mechanisms are not a prerequisite to maintaining genomic integrity. The biological machinery enabling viral replication does include correction mechanisms though and much more than that.
Comment by Bradford — January 12, 2008 @ 2:31 pm
January 13th, 2008 at 2:41 pm
Hi Bradford: Which self-correcting genomic mechanisms are actually prerequisite to maintaining genomic integrity? If memory serves me right, one can disrupt numerous error-correction mechanisms in yeast without actually disrupting the viability of yeast.
Comment by hrun — January 13, 2008 @ 2:41 pm
January 13th, 2008 at 3:32 pm
Why not cite the reference? The error correction function is essential. Can it be more efficient in some organisms than in others- yes. Is partial impairment necessarily fatal- perhaps not. Let's take a look at the particulars.
Comment by Bradford — January 13, 2008 @ 3:32 pm
January 13th, 2008 at 5:27 pm
Check out the Saccharomyces Genome Database. You can look at various 'error correction' mutations like rad50. Other examples of 'error correction' mutations for spindle alignment are the mad and bub genes.
Most of the genes can be deleted without causing inviability. Thus, it appears that they are not prerequisite to maintaining genomic integrity. They may be helpful, but not prerequisite.
Comment by hrun — January 13, 2008 @ 5:27 pm
January 13th, 2008 at 5:36 pm
So their reproductive fitness with respect to others of the same species is impacted in what way?
The most efficient organism at eliminating genetic mistakes is unicellular. Variation in effectiveness among living organisms exists. I don't know of any cellular organism lacking an error correction capacity. Do you?
Comment by Bradford — January 13, 2008 @ 5:36 pm
January 13th, 2008 at 6:17 pm
Bradford, you claimed that error correction is a prerequisite for genomic integrity. If I can show you organisms that are viable when error correction mechanisms are disable, would that invalidate your claim?
Did you look at the website? It's all there:
phenotypes of rad50 mutant
Comment by hrun — January 13, 2008 @ 6:17 pm
January 13th, 2008 at 7:49 pm
You did not show that all error correction mechanisms were disabled. Mechanisms are plural.
Comment by Bradford — January 13, 2008 @ 7:49 pm
January 13th, 2008 at 9:07 pm
No, I have not. But I am challenging your blanket statement that 'error correction is a prerequisite for genomic integrity'. I don't think that you are able to actually support that statement.
I absolutely agree that error correction is necessary for efficient growth, for the ability to deal with induced DNA damage, … but this does not mean that it is a prerequisite for genome integrity (or life for that matter).
Comment by hrun — January 13, 2008 @ 9:07 pm
January 13th, 2008 at 9:16 pm
Things are ultimately settled experimentally but there is much more to genomic repair than RAD 50 and recombination. The minimal genome studies I'm familiar with include at least a partial set of genes whose function involves error correction. If you want to argue that the matter is an open question I have no difficulties with that but would not have many doubts about the consequences to an organism of disabling a laundry list I could supply.
Comment by Bradford — January 13, 2008 @ 9:16 pm
January 13th, 2008 at 9:26 pm
Yes, there is certainly more to it. Yet, until it is actually established that error correction is essential for genomic integrity (or life), maybe it should not be claimed that it actually is.
For example, I have seen people argue that error correction/DNA repair is another example of IC. For that to be true, one has to show that maintaining a functional genome without error correction/DNA repair is ACTUALLY essential.
Comment by hrun — January 13, 2008 @ 9:26 pm
January 13th, 2008 at 9:34 pm
Your confusing a claim for essentiality of function with IC which has to do with function involving multiple components. While all components of a given system may not be essential (sub-optimal function being possible) biological systems do entail interacting parts- at least some of which are needed for known and identifiable functions.
Comment by Bradford — January 13, 2008 @ 9:34 pm
January 13th, 2008 at 10:48 pm
(there was of course a typo in my previous post and 'essential' should have been 'impossible')
I'm not confusing things at all, the claim goes like this:
1) DNA is required to encode DNA repair proteins and the error correction machinery
2) DNA repair proteins/error correction machinery are essential for the maintenance of a genome
Take 1) and 2) together and you have an IC system.
Comment by hrun — January 13, 2008 @ 10:48 pm
January 14th, 2008 at 12:49 am
Agreed.
Comment by Bradford — January 14, 2008 @ 12:49 am
January 14th, 2008 at 8:53 am
So then you agree that as of now that conclusion is actually not valid: We do not know if DNA repair proteins or an active error correction machinery are essential to the maintenance of a genome.
It is entirely possible that a genome can be maintained, albeit less efficiently, without error correction. And at some later point, the error correction machinery could have made this process more efficient.
Comment by hrun — January 14, 2008 @ 8:53 am
January 14th, 2008 at 3:27 pm
We have very good evidence that correction and error repair are essential. We know the causes of genomic corruption and are aware of the specifcity required for many essential biological functions. We also have an abundance of evidence indicating what occurs when error correction functions are disabled. There are labels attached to the diseases associated with this.
Maintenance entails some form of error correction or control of oxygen radicals and other metabolic by-products that incur damage. The list of damaging causes includes natural and inevitable natural processes like deamination, spontaneous losses and radiation including UV. There is no escaping causes of genetic damage. The problem with the developing at a later point scenario is it assumes new information would be generated faster than genetic information is lost through damaging agents. Testing genomes, stripped of their error correction mechanisms, would offer a means of putting your belief on an empirical basis.
Comment by Bradford — January 14, 2008 @ 3:27 pm
January 14th, 2008 at 3:36 pm
Like what?
Yes, certain mutations in error correction are associated with human disease. But numerous mutations in error correction show no obvious phenotype at all– and definitely not inviability– in yeast. So this does not lead to the conclusion that error correction is necessary for life as we know it.
That's why I say you should check out the phenotypes associated with such mutants. It turns out that numerous of such mutants are viable. I'm just trying to figure out how you are getting to your conclusion that an error correction mechanism is prerequisite for genomic maintenance (and presumably for life in general).
Comment by hrun — January 14, 2008 @ 3:36 pm
January 14th, 2008 at 3:50 pm
But the transcription & translation machinery would be a product of this genome which lacks error correction procedures. They would be comprised of proteins & rna that would be less efficient because they too would have no way of correcting the deleterious mutations that would occur in genetic segments that specified those constituent proteins and rna.
Then the proteins and rna that those transcriptional & translational apparatuses would yield would in turn be less efficient. It seems like the problem just snowballs until a mutational crash would lead to an organism that is corrupted beyond repair.
Comment by Doug — January 14, 2008 @ 3:50 pm
January 14th, 2008 at 3:52 pm
You sourced a reference indicating corruption of the recombination mechanism; hardly a disablement of error correction.
Genetic damage is a natural phenomenon. Continued unchecked damage has only one eventual outcome- genomic meltdown.
Comment by Bradford — January 14, 2008 @ 3:52 pm
January 14th, 2008 at 4:03 pm
I don't see how this helps your position. If it shows no effect at the level of the phenotype than natural selection is unable to even weed out the deleterious mutations. If you would have said that their effects are noticeable at the phenotypic level than at least NS would be able to play some role in limiting these deleterious mutations from getting tossed back into the next reproductive cycle (since NS works at the level of the organism and not the level of a single nucleotide).
This points back to the problem that Bradford is focusing on. With no internal means to correct deleterious mutations & no way for NS to weed them out (since you stated that their immediate effect won't be noticeable at the level of the phenotype) how does this not lead to an inevitable crash?
Comment by Doug — January 14, 2008 @ 4:03 pm
January 14th, 2008 at 4:10 pm
This is not about my position. It is merely about the claim that error correction is a PREREQUISITE (or necessary) for genomic integrity (or life in general). That's the only thing I'm arguing about.
Because the organisms with the error correction a) often grow faster than their mutant counterparts and b) in case they are exposed to larger doses of DNA damaging agents the organisms without error correction may die while the others don't.
But, bear in mind, that neither a) nor b) allows you to conclude that life is impossible without error correction.
Comment by hrun — January 14, 2008 @ 4:10 pm
January 14th, 2008 at 4:15 pm
There is much more evidence for that position than the frequently argued one that genomes self-assembled through an unknown chemical process. ID critics have a way of raising and lowering the evidentiary bar to suit their ideological preferences.
Comment by Bradford — January 14, 2008 @ 4:15 pm
January 14th, 2008 at 4:28 pm
So the support for you position is that there are other scientific positions that have even worse evidentiary support?
And I asked you before: What is that evidence that supposedly supports your assertion? I have just scanned back through our discussion here, and I have not found any evidence to support the assertion that error correction is a prerequisite for life.
Comment by hrun — January 14, 2008 @ 4:28 pm
January 14th, 2008 at 4:36 pm
Doug:
This doesn't make sense. A mutation is deleterious by definition if it induces a phenotype of lower fitness. Hence a mutation without a phenotypic effect cannot be deleterious.
Comment by Raevmo — January 14, 2008 @ 4:36 pm
January 14th, 2008 at 4:36 pm
hrun, if you are ignorant of error repair mechanisms take a good textbook on biochemistry or genetics and study it. If your problem is lack of knowledge as to effects of disabling key enzymes involved in repair pathways then look up related diseases. If you cannot substract unremedied damage to genomes and figure out zero function is the ultimate result noone can help you. To claim there is no evidence is weaker than claiming no evidence for evolution. The claims and counter claims are testable.
Comment by Bradford — January 14, 2008 @ 4:36 pm
January 14th, 2008 at 4:38 pm
But isn't it at least a prerequisite for genomic integrity? I might be confused with the point your making, but how else is genomic integrity to be maintained if not via error correction mechanisms? And if the effects of these deleterious mutations are not having an immediate impact at the level of the phenotype. Unless you and I are understanding 'genomic integrity' differently.
I agree, they will die more rapidly while the others won't - but you've still got accumulating mutations at the genetic level. Initially those effects might be obscured from the level of the whole organism, allowing reproductive success (compared to those with no success). But these mutations are just getting passed on to the next generation. You've got inherited mutations accruing along with new mutations occuring.
Comment by Doug — January 14, 2008 @ 4:38 pm
January 14th, 2008 at 4:43 pm
This paper (Kun et al 2005, Nature Genetics 37) suggests that mutation rates may be small enough to counter Eigen's paradox:
Comment by Raevmo — January 14, 2008 @ 4:43 pm
January 14th, 2008 at 4:49 pm
Excellent. I asked for evidence to support your position. First we get numerous posts without any evidence. Then you cite in your support that other positions are even less supported. Finally you accuse me of ignorance.
I don't have any lack of knowledge about repair pathways. That's why I knew that you can disable numerous players in the repairs pathways of yeast without actually inducing lethality in yeast.
Finally, your argument about 'zero function being the ultimate result' is also not self evident. In order to show that, you have to show that the rate of deleterious mutations is high in comparison to the reproduction rate of the organism in question. For example, if an organism accumulates only a single critically deleterious mutation in the time it takes it to divide, then the population level will be static (all other sources of mortality disregarded). If the reproduction rate exceeds the rate of critically deleterious mutations, then the organisms will persists without any error correction machinery.
Now, again, do you have any evidence to support your position, other than that certain positions have less support or that I am ignorant?
Comment by hrun — January 14, 2008 @ 4:49 pm
January 14th, 2008 at 4:52 pm
Doug, see my previous post to Bradford. In order to be necessary for genomic integrity, the frequence of critically deleterious mutations has to be higher than the reproduction rate of the organism. Simple as that.
Comment by hrun — January 14, 2008 @ 4:52 pm
January 14th, 2008 at 5:19 pm
Then maybe you are ignorant of the significance of citing mutations affecting a function -recombination- and thinking you have made a point about error repair that extends beyond recombination.
Your analysis is flawed. The unchecked mutations would affect every organism of the species. The reproductive fitness of that species would steadily decline. Extinction is the logical prediction.
Comment by Bradford — January 14, 2008 @ 5:19 pm
January 14th, 2008 at 5:30 pm
Bradford, again, I ask, what is your evidence?
Huh? The unchecked mutations would only affect the progeny of the organism that got the mutation in the first place, not the entire species. And what is your evidence that the reproductive fitness of that species would steadily decline? You have none.
Let's do a simple thought experiment, ok? Let's assume that organism X accumulates a single point mutation every day and let's assume that this organism can divide every hour. Let's finally assume that organism X does not have any DNA repair or error correction machinery.
What do you think? Is the genome of said organism stable or will extinction be the logical prediction?
Granted there is no such organism X. In fact the mutation rate will probably be higher, but how much higher? So, extinction is only the logical prediction if the mutation rate is sufficiently high in relation to the reproductive rate of the organism.
Comment by hrun — January 14, 2008 @ 5:30 pm
January 14th, 2008 at 5:32 pm
Bradford:
Bullshit. Have you ever examined a mathematical model for the error threshold? Whether mutational meltdown occurs or not depends on mutation rate, selection differentials and genome size. It's not a done deal without error correction.
Comment by Raevmo — January 14, 2008 @ 5:32 pm
January 14th, 2008 at 5:38 pm
During the time it takes you to read this comment dozens of genetic errors will be repaired in your body. It is a constant process that never stops. If the corrections were not made death would result and would not take long for that matter. What do you think the cause of death is from radiation poisoning- the type found at Hiroshima or Nagasaki or Chernobyl? It is the complete and unremedied destruction of DNA. The errors overwhelm repair capacity. The only diffference between these catastrophic scenarios and the stripped genomes is one of time. It would take longer to cripple genomes stripped of repair capacities.
Comment by Bradford — January 14, 2008 @ 5:38 pm
January 14th, 2008 at 5:41 pm
Mutations rates are over the top when no repair mechanisms exist. The mutation ranges we see are limited by error repair. Take error repair away and the rate skyrockets.
Comment by Bradford — January 14, 2008 @ 5:41 pm
January 14th, 2008 at 5:47 pm
Since we have about 10 trillion cells in our body, the number of mutations that are being repaired in my body during this time is undoubtedly MUCH higher than just a few dozen.
And again you are refusing to engage into the actual point of contention. I do not claim that no mutations occur in the human body. I do not claim that error correction in the human body is unimportant. I do not claim that an organism without error correction functions equally good as an organism with error correction.
I solely challenge you to support the notion that DNA repair or error correction is prerequisite or necessary for life. And you have again failed to give any support to your assertion.
Really, and you know that how? If mutation rates skyrocket if you take away error repair, why are so many yeast mutations in the error repair pathways viable?
Comment by hrun — January 14, 2008 @ 5:47 pm
January 14th, 2008 at 5:57 pm
I've repeatedly said that my claims as well as your counter claims are testable. Researchers involved in minimal genome studies have indirectly tested the claim. Such genomes include at least a partial set of genes whose function entails error correction.
Comment by Bradford — January 14, 2008 @ 5:57 pm
January 14th, 2008 at 5:58 pm
How many error mechanisms other than recombination can you cite?
Comment by Bradford — January 14, 2008 @ 5:58 pm
January 14th, 2008 at 6:02 pm
Bradford:
You make quantitative claims without showing any calculations or references. Why should we believe you? I have shown you a paper that shows that mutation rates in some ribozymes are sufficiently small to overcome the error threshold (which can be calculated according to ancient models by Eigen and Schuster), meaning that without error correction genomes of substantial sizes can persist in mutation-selection balance. What is your quantitative counter-argument?
Comment by Raevmo — January 14, 2008 @ 6:02 pm
January 14th, 2008 at 6:04 pm
What is synthesized as a consequence of the expression of this "ribozyme genome?"
Comment by Bradford — January 14, 2008 @ 6:04 pm
January 14th, 2008 at 6:13 pm
Why should I bother? You don't seem willing to engage in an actual argument anyway… Otherwise you would have know that I already cited two additional error correction mechanisms, namely the spindle assembly checkpoint and the spindle orientation checkpoint.
But we can look at others as well if you like. Double strand break repair, excision repair, UV repair…
So, did I cite a sufficient number for you? Now can you actually engage in the argument? I ask this again: what evidence do you have to support your assertion (other than simply saying 'Take error repair away and the rate skyrockets')?
Comment by hrun — January 14, 2008 @ 6:13 pm
January 14th, 2008 at 6:20 pm
Since you are repeating yourself I'll return the favor. The claims and counter claims are testable. The results are distinguishable from an argument. One thing you have amply illustrated with your arguments is the scientific nature of ID arguments. They can be tested.
Comment by Bradford — January 14, 2008 @ 6:20 pm
January 14th, 2008 at 6:23 pm
Bradford:
What difference does that make? I recommend reading the paper. I can email it if you don't have access.
Comment by Raevmo — January 14, 2008 @ 6:23 pm
January 14th, 2008 at 6:27 pm
The difference lies in the difference between an enzyme and a genome. Minimal sizes of the latter include hundreds of genes. Function and dynamics differ. A minimally functional genome is a better testing candidate.
Comment by Bradford — January 14, 2008 @ 6:27 pm
January 14th, 2008 at 6:43 pm
Mike Gene is the premier front loading advocate, not me. However, it appears that error correction function claims amount to a claim related to front loading of specific mechanisms. The claim can be linked to specific stages of development. At what point does error correction become a necessity? The outset? Later? When?
Comment by Bradford — January 14, 2008 @ 6:43 pm
January 14th, 2008 at 6:45 pm
?
First: I have asked numerous times for support for your assertion. I am keenly aware that your assertion is testable. You just don't provide any support for your assertion. I really don't exactly know why you fail to do so. One can only suspect that you actually do not have such evidence to present.
Second: I know that results are distinguishable from arguments. That's why I actually link to sites that show that mutants in error correction are viable and that contrary to your assertion, the mere inactivation of certain error correction pathways does not lead to inviability (or skyrocketing mutation rates). I have also shown by a simple thought experiment that your assertion 'Extinction is the logical prediction.' is simply and plainly false.
Third: I have not shown that 'ID arguments' are scientific. I have shown that you can not support your assertion with evidence. Your assertion, btw. is not necessarily an 'ID argument'. Even if organisms are designed, error correction does not have to be a prerequisite for life. So every single organism on this planet could be directly designed by a biochemist or by a deity, yet, your assertion still could just as well be false.
Fourth: I knew I shouldn't have bothered with citing more error repair mechanisms. You asked how many I could cite, I cite more, you ignore it. Why do you even ask?
Comment by hrun — January 14, 2008 @ 6:45 pm
January 14th, 2008 at 6:49 pm
Bradford:
Please. Here's once again from the abstract I quoted:
Your argument has been shredded. You do not respond to the quantitative challenges to your unfounded claims. Game over.
Comment by Raevmo — January 14, 2008 @ 6:49 pm
January 14th, 2008 at 6:51 pm
Genomic integrity is compromised whenever a deleterious mutation occurs; the issue would be the reach and extent of that compromise.
But focusing primarily on 'critically deleterious mutations' doesn't seem right. Because if those mutations go unchecked and if no mechanism is in place to prevent their accumulation, it doesn't appear to be an issue of the occurrence of one mutation being criticially deleterious but that of a net accumulation which would be critically deleterious.
But what is the frequency of mutations? If you look at a human genome the average offspring has about 280 mutational variations from his/her parents - and that's with all of the corrective machinery. Also, that number could be on the low end of actual mutational variations - the numbers reflect mathematical scenarios not necessarily the actual mutation rate.
Comment by Doug — January 14, 2008 @ 6:51 pm
January 14th, 2008 at 6:52 pm
I'd ignore the taunt, Bradford…. probably just those monkey genes kicking in again.
Comment by Doug — January 14, 2008 @ 6:52 pm
January 14th, 2008 at 6:57 pm
It would not matter what evidence I presented. You have a closed mind that is threatened by the issue. Continued and unmitigated destruction to genomes has only one end result. The reproduction argument is a fallacy. All members of a species would experience degradation and continual decline of fitness. Selection would amount to the selection of the less and less fit as time goes on. Spell extinction
You're misrepresenting my claim. I never claimed a partial impairment of genomic integrity is necessarily fatal. My claims are global with respect to repair functions. Knock them all out and watch the steady and continous decline in reproductive fitness.
If genomic function fails in the absence of error correction then you can argue for a natural miracle in lieu of design.:wink:
Comment by Bradford — January 14, 2008 @ 6:57 pm
January 14th, 2008 at 6:58 pm
Raevmo, don't you teach how to distinguish an enzyme from a genome?
Comment by Bradford — January 14, 2008 @ 6:58 pm
January 14th, 2008 at 7:02 pm
Doug:
I noticed you didn't respond to my devastating critique of your previous post. If you don't know what you're talking about, it's better to shut up, to paraphrase a famous Austrian philosopher.
Biology has become a very quantitative science as of lately. Gut-feeling arguments just don't cut it anymore.
Comment by Raevmo — January 14, 2008 @ 7:02 pm
January 14th, 2008 at 7:04 pm
Ignore the taunt if you like, but please, finally engage in the arguments made.
Exactly. You are actually engaging in the argument that this is a quantitative problem. The deleterious mutations have to occur at a high enough rate as to overwhelm reproduction.
Again, you are actually engaging in the argument. You are absolutely right in this case. So now one has to show (in order to prove that error correction is necessary for life) that this deleterious mutations accumulate at a sufficiently high rate.
Ok. So humans have a genome size of about 3×10E9 while E. coli has a genome size of only 5×10E6. So 280 mutational variations in every 3×10E9 basepairs would amount to only numerous completely flawless E. coli copies.
So mathematically, the mutation rate can actually be higher and still there would be plenty of flawless copies of E. coli around.
Comment by hrun — January 14, 2008 @ 7:04 pm
January 14th, 2008 at 7:10 pm
Sigh. As I thought. You ignored the taunt AND completely failed to engage in the argument. Again.
But hey, now we have a new line of argument from you. Let's list it chronologically:
1) numerous posts without a single argument
2) other positions have even less support than your assertion
3) I am ignorant
4) I have a closed mind
I wonder what would be next.
Comment by hrun — January 14, 2008 @ 7:10 pm
January 14th, 2008 at 7:17 pm
Bradford:
No I don't. All I teach freshmen biologists is statistics. More advanced biology and modeling for older students. But I hope you do realize that the earliest genomes might very well have been self-replicating enzymes. Oh, and you still didn't respond to my quantitative challenges to your claims about repair mechanisms.
Comment by Raevmo — January 14, 2008 @ 7:17 pm
January 14th, 2008 at 7:38 pm
Hrun, why have you ignored repeated references to minimal genome research that includes, within the minimal number, error repair genes. If error correction was superfluous then one would expect minimal genomes to exclude the related genes.
Comment by Bradford — January 14, 2008 @ 7:38 pm
January 14th, 2008 at 7:43 pm
Comment by Bradford — January 14, 2008 @ 7:43 pm
January 14th, 2008 at 7:53 pm
Doug:
The mutation rate is a small fraction of what it would be without error correction. Global affliction increases the fatality rate for the species. Also increasing would be the amount of mildly deleterious mutations that can be tolerated only to a point. I don't know of anyone other than hrun who thinks error correction is a superfluous function.
Comment by Bradford — January 14, 2008 @ 7:53 pm
January 14th, 2008 at 8:01 pm
Bradford:
You have not produced a shred of evidence for your position, which has been demolished by quantitative counter-arguments which you refuse to answer. Now you mock my suggestion for an explicit model, that which you have so far refused to posit. What's wrong with you, man?
Comment by Raevmo — January 14, 2008 @ 8:01 pm
January 14th, 2008 at 8:05 pm
Bradford:
What is your evidence for these quantitative claims? I think you're just making this up. Are you full of shit or can you back up your claims?
Comment by Raevmo — January 14, 2008 @ 8:05 pm
January 14th, 2008 at 8:22 pm
Why have you ignored repeated references to minimal genome research that includes, within the minimal number, error repair genes. If error correction was superfluous then one would expect minimal genomes to exclude the related genes.
Comment by Bradford — January 14, 2008 @ 8:22 pm
January 14th, 2008 at 8:33 pm
Bradford:
For the third time, I quote from the Nature Genetics article:
Meaning that minimal genomes can do without error repair, contra your claim.
Unless you have scientific evidence to the contrary, can we now finally agree that you might have been wrong in claiming that error correction is essential?
Comment by Raevmo — January 14, 2008 @ 8:33 pm
January 14th, 2008 at 8:44 pm
There is a distinction between essential and minimal. There is also the matter of how long an organism can remain viable. Loss of total error correction capacity would result in a steady degeneration of a genome would you agree?
Comment by Bradford — January 14, 2008 @ 8:44 pm
January 14th, 2008 at 8:59 pm
Bradford:
No I would not. Nor would you, had you looked up the mathematical models that calculate the probability that this would happen. It's very simple: if the mutation rate is small enough, and if the negative fitness effect of a mutation is small enough, then a population can persist without error correction. That's a mathematical fact. The question is if these conditions would hold for actual populations of simple genomes. Well, the paper I cited suggests that they do. If you want to object to these findings, you had better come up with some solid quantitative arguments. Otherwise, we're entitled to conclude that you're just bluffing.
Comment by Raevmo — January 14, 2008 @ 8:59 pm
January 14th, 2008 at 9:18 pm
Your belief that the mutation rate of a correctionless genome would be small enough is based on what?
I have a problem taking seriously the suggestion that self-replicators of an RNA world constitute a genome. The distinguishing feature of genomes is not their link to nucleic acids, it is the function of codons within them. They are symbolic. Enzymatic NRs are not. It makes no sense to ruminate about genomic integry when the "genome" lacks even a convention by which the order of its nucleotides can be understood. If a change occurs in a ribozyme and I ask on what basis is selection evaluated what is your answer? Will the change be retained assuming a continuous reaction and be evident on future SRMs? How does that get us closer to a cell?
Comment by Bradford — January 14, 2008 @ 9:18 pm
January 14th, 2008 at 9:34 pm
Bradford:
For the fourth time I refer to the Nature Genetics paper I cited. Would you please read it?
No, you have a problem taking seriously actual evidence. Your bizarre definition of a genome is completely irrelevant. Well, I am going to sleep now. I hope you come to your senses.
Comment by Raevmo — January 14, 2008 @ 9:34 pm
January 14th, 2008 at 9:52 pm
Raevmo, I asked:
You refer me to a paper the first sentence of which states:
Again I ask you what is the fittest genotype? How is one ribozyme more fit than another in an extracellular world? Why would nature favor the continuity of a reaction of self-replicating molecules? It is your conception of how life must have arisen that is driving your analysis of selection.
Comment by Bradford — January 14, 2008 @ 9:52 pm
January 14th, 2008 at 11:41 pm
A few random comments…. First I will add to the debate and just say it wouldn't hurt if bradford gave a refrence to shut them up! I think they are worthy apponents to spend 10 minutes to find an article that supports your claims (I do agree with what you have been saying btw bradford). Anyhow, I just wanted to say thanks to Mike Gene, I just got the book today and in the midst of doing homework (on the first day back) it was a pleasure to read what I wanted to be the first page of the introduction, which turned into reading all five pages. Mike is a great writer! Ok, now I have to get back to homework, keep up the good debate guys!
Comment by gore — January 14, 2008 @ 11:41 pm
January 15th, 2008 at 4:48 am
Bradford,
Why won't you engage the arguments presented by hrun and Raevmo? It's almost as if you think that evasion is tantamount to victory. It's not. We can all see you running away from the questions.
You're doing the same thing on the Dembski thread, where I've raised a set of issues that you refuse to address.
Here's a way of looking at things that might reduce your confusion:
Every population fights a constant battle against mutations. Mutations are inexorable, and if the deleterious ones are not removed from the population they will accumulate, eventually rendering the population extinct.
How are mutations removed from a population? One way, as you keep telling us, is via error correction mechanisms. If an error is successfully corrected, it is as if the error had never happened in the first place.
But here's the crucial point: error correction mechanisms are not the only way deleterious mutations are removed from a population. Natural selection also does the job. Organisms having a deleterious mutation tend to be outcompeted by those that do not, and they die off, taking the mutation with them.
For a population to remain viable, it must prevent deleterious mutations from accumulating. But it can eliminate them either via error correction or natural selection. As long as the total rate of removal balances the rate of introduction, the population remains viable.
Note that this remains true even if rate of error correction is zero (i.e. no error correction at all). As long as natural selection by itself can keep up with the rate at which deleterious mutations are introduced, the population will survive.
The lesson: error correction is not inherently essential, only contingently so. If the mutation rate and the size of the genome are low enough, natural selection by itself will maintain a viable population.
If you claim otherwise, it is up to you to provide a quantitative argument. Raevmos has repeatedly requested this from you, but this appears to be a request you'd rather evade.
Comment by valerie — January 15, 2008 @ 4:48 am
January 15th, 2008 at 6:58 am
gore writes:
Raevmo had written:
And Valerie parrots:
(From Molecular Biology of the Cell, (Alberts, Johnson, Lewis, Raff, Roberts, Walter) The rate of mutation is about the same for most organisms including those as different as bacteria and humans. Each time DNA is replicated this is about a change in one nucleotide per 10^9 nucleotides (page238). On the preceeding page (237) there is a heading entitled Low Mutation rates Are Necessary for Life as We Know It. There is this highly interesting first paragraph:
Talk about front loading! Our biological enablement (in terms of DNA repair and the number of essential proteins) was evident already in unicellular genomes. But there is more. Is it tenfold?
Each of us was born with at least 350 new mutations that make our DNA different from that of our parents.
Our model bacterium is Esherichia coli the common, and mostly benign, intestinal bacterium. The entire genome was sequenced in 1997 (Blattner et al., 1997) and its size is 4,200,000 base pairs (4.2 × 106 bp). Every time a bacterium divides this amount of DNA has to be replicated; that's 8,400,000 nucleotides (8.4 × 106).
The most common source of mutation is due to mistakes made during DNA replication when an incorrect nucleotide is incorporated into newly synthesized DNA. The mutation rate due to errors made by the DNA polymerase III replisome is one error for every one hundred million bases (nucleotides) that are incorporated into DNA. This is an error rate of 1/100,000,000, commonly written as 10-8 in exponential notation. Technically, these aren't mutations; they count as DNA damage until the problem with mismatched bases in the double-stranded DNA has been resolved. The DNA repair mechanism fixes 99% of this damage but 1% escapes repair and becomes a mutation. The error rate of repair is 10-2 so the overall error rate during DNA replication is 10-10 nucleotides per replication (10-8 × 10-2) (Tago et al., 2005).
Underscoring the front loading point there is this:
GENOME stability is continually challenged by a diverse array of mutagenic forces that include errors during DNA replication, environmental factors such as UV radiation, and endogenous mutagens such as oxygen free radicals generated during oxidative metabolism (LINDAHL 1993). Multiple DNA repair pathways have evolved to minimize the mutagenic consequences of DNA damage and erroneous DNA replication. Most of the major DNA repair pathways have been detected in all three domains of life, suggesting ancient origins.
Then there is this caveat.
And finally if you want to see a ribozyme click here.
Comment by Bradford — January 15, 2008 @ 6:58 am
January 15th, 2008 at 7:40 am
Bradford, quoting from good old Alberts et al:
I don't know how Alberts et al. arrive at these numbers (my girlfriend stole my Alberts so I can't check), but let's suppose they are correct. Then by extending the argument even further, a mutation frequency hundredfold higher would limit an organism to about 600 essential genes. Since DNA error repair removes 99% of replication errors, an organism without error repair could have at most 600 genes. The smallest known genome of extant organisms is smaller than that. Hence, life can be sustained without error correction. QED.
Comment by Raevmo — January 15, 2008 @ 7:40 am
January 15th, 2008 at 10:16 am
good job bradford, I knew you had it in you
Comment by gore — January 15, 2008 @ 10:16 am
January 15th, 2008 at 10:21 am
My apologies, for the most part I ignore what you say now. There are more able, interesting critics to read and listen to.
But where is that devastating critique?
Comment by Doug — January 15, 2008 @ 10:21 am
January 15th, 2008 at 10:38 am
Doug:
And I apologize to you, for being rude. My utterly and completely devastating critique concerned your oxymoronic postulation of deleterious mutations without phenotypic effects.
Comment by Raevmo — January 15, 2008 @ 10:38 am
January 15th, 2008 at 12:21 pm
Contrary to your approach I'm interested in actually learning something. If there is something that shows that my earlier contention is indeed incorrect, great! I'll read it, consider it, and possibly adopt it.
But where is the critique?
Is it this?
Natural selection can only act upon fitness, but how high is the heritability of 'fitness traits'?
So, 'fitness traits' should be under strong directional selection, right? And therefore should display lower levels of genetic variation, right?
Therefore 'fitness traits' should have high heritability, right?
Nearly-neutral mutations can occur and not have an immediate effect at the level of the phenotype. Can those mutations accrue and eventually have a negative impact at the phenotypic level? Certainly, that's my point. Natural selection can't work at the level of the individual nucleotide, only at the level of the whole organism. Some mutations that occur will not have an immediate effect at the level of the organism, so they can't be directly selected again - allowing them to be passed into the next generation.
Reproductive success isn't not an indicator of true-absolute value of genetic fitness, it's only relative to other organisms competing to reproduce. Would you say that this is a reflection of the integrity of the organisms genome? Isn't it quite possible that over numerous generations, as the mutations are accruing, that class of organisms could be headed towards extinction?
Comment by Doug — January 15, 2008 @ 12:21 pm
January 15th, 2008 at 1:40 pm
Doug:
You need to learn better mind-reading skills, as you incorrectly conclude that I'm not interested in learning something. I am and I do.
That's it. If you disagree I'd like to hear your definition of deleterious.
Heritability (in the narrow sense) is additive genetic variance divided by phenotypic variance (h^2=VA/VP). If selection depletes additive genetic variation in fitness, the result will therefore be lower heritability, not higher. Indeed, fitness-related traits usually have lower heritabilities than, say, morphological traits. Nevertheless, considerable natural variation in fitness persists, for numerous possible reasons such as fluctuating selection pressures, sexually antagonistic selection, etc.
I'm not sure what you mean by "immediate effect", but nearly-neutral is not neutral, hence there must a (small) phenotypic effect.
It can work at multiple levels, including below the individual level. There are genes (segregation distorters) that can cause the demise of sperm that do not carry them, thereby becoming over-represented among sperm.
Agreed, that's possible. Like when you X-ray your balls a little too long.
I'm not sure what you mean by "integrity of the organisms genome".
Perhaps, yes. Some conservationists are worried this might happen to small populations ("genetic erosion").
And indeed, it is probably more likely to happen in the absence of error repair. But the point is that error repair has not been shown to be a necessary condition for life. Early, relatively simple life with small genomes might have flourished without. Error repair could have evolved later.
Comment by Raevmo — January 15, 2008 @ 1:40 pm
January 15th, 2008 at 2:15 pm
See, now there's a thoughtful and considerate post…. left out all insults, good job!
I'm going to read it over more thoroughly and get back to you.
PEACE
Comment by Doug — January 15, 2008 @ 2:15 pm
January 15th, 2008 at 5:17 pm
Raevmo:
As I've been saying the claims and counter claims are testable. It should be noted though that a genome with 600 genes is a very small one and would be devoid of redundancy and a degree of robustness that would be found in larger genomes.
Comment by Bradford — January 15, 2008 @ 5:17 pm
January 15th, 2008 at 5:46 pm
Bradford:
Fair enough. As long as we can agree that error repair is not necessarily a prerequisite for life.
Comment by Raevmo — January 15, 2008 @ 5:46 pm
January 15th, 2008 at 6:07 pm
Bradford wrote:
No, you've been saying that error correction is essential in principle for life, and that any genome will degrade without it.
Why is it so difficult to admit that you were wrong?
Comment by valerie — January 15, 2008 @ 6:07 pm
January 15th, 2008 at 6:22 pm
Valerie,
Are you claiming that a genome will not degrade if there is no error correction machinery in place?
Why do you phrase it like this? You assume that he is definitely wrong, he knows it, and the only thing in his way is his inability to admit it.
It just seems like an odd way to discuss these things - but it also seems typical of how you handle yourself here.
Comment by Doug — January 15, 2008 @ 6:22 pm
January 15th, 2008 at 6:43 pm
Doug:
That is possible. A genome will degrade only if mutations are added to the population at a higher rate than they are removed. Error correction is one way to remove mutations, selection is another. Without error correction, if the mutation rate is small enough, there will be no meltdown, because selection removes enough mutations. It's as simple as that.
Comment by Raevmo — January 15, 2008 @ 6:43 pm
January 15th, 2008 at 6:45 pm
I understand the point you're making. But that's not what I directed at Valerie, I simply asked if the genome would not degrade.
Comment by Doug — January 15, 2008 @ 6:45 pm
January 15th, 2008 at 6:52 pm
Doug:
It will not under the right circumstances. Smaller genome and smaller mutation rate making those circumstances more likely.
Comment by Raevmo — January 15, 2008 @ 6:52 pm
January 15th, 2008 at 6:55 pm
Selection doesn't work at the level of individual nucleotide mutation, only at the level of the whole organism. So every mutation that occurs will not be removed by selection, allowing them to accrue.
Comment by Doug — January 15, 2008 @ 6:55 pm
January 15th, 2008 at 7:10 pm
Doug:
OK, I should have said non-neutral mutations. Neutral mutations are irrelevant for the argument anyway. Do you get it now?
Comment by Raevmo — January 15, 2008 @ 7:10 pm
January 15th, 2008 at 7:14 pm
Valerie:
I said error correction is essential. Not in principle. But because that is my view based on the data I'm familiar with. My view is subject to testing. I could be wrong but could be right as well.
You have no empirical evidence showing that I'm wrong. Why is it so difficult for you to maintain an open mind?
Comment by Bradford — January 15, 2008 @ 7:14 pm
January 15th, 2008 at 7:23 pm
Raevmo:
Disagree. Minimal function requires a certain size level in my view. Organisms with extremely small genomes tend to be obligate parasites being dependent on other organisms for needs their own genomes are unable to meet. There is scant evidence that cellular life would be sustainable when the gene count falls to a couple hundred genes or so.
Comment by Bradford — January 15, 2008 @ 7:23 pm
January 15th, 2008 at 8:28 pm