Please do not hijack this thread - I am an engineer by profession and a firm believer in science. Please argue on the merits of your argument and do not try to dismiss my post as supporting creationism or any such belief.
Reading around I found two good articles on Adaptive Mutations - here and here
From the Pub Med site : "Adaptive mutation is defined as a process that, during nonlethal selections, produces mutations that relieve the selective pressure whether or not other, nonselected mutations are also produced. Examples of adaptive mutation or related phenomena have been reported in bacteria and yeast but not yet outside of microorganisms. " (Emphasis mine)
In part, I feel vindicated that my arguments in the previous post are somewhat validated despite being dismissed by traditional genetists.
So, if proven true - for the flatfish post on the board, we will have flatfish with eyes on one side much faster if adaptive mutations are present - right ? Is there a scientific prediction as to how much time it will take for a population of flatfish to have eyes move to one side given random mutations only ?
Is scientific understanding of mutations mature or ironically still evolving ?
As a non-biologist but someone that follows/reads this stuff daily, these would be my observations:
1 - Things are always more complex than we think they are today
2 - The wiki article explains problems with the bacteria study
2.1 - They didn’t test for mutation rate if lactose was present and compare the two
2.2 - They didn’t test for whether it’s just due to an overall increase in random mutations due to environmental stress
3 - Lets say the lactose result was valid, then consider all that is going on with the chemical processes at that level, it could be that there is a specific chemical pathway causing that mutation in those conditions. But if you move up a level to something more abstract, like adding neurons to be smarter, now you need to mutate many genes to arrive at that state and there isn’t a direct mapping of chemical states from the problem trying to be solved (need to be smarter to compete better) to the changes required to make it happen.
What I have found genetists to be very elusive about is the time it will take for a given mutation to emerge. For example : if you take a culture of bacteria acclimatized to a pH of 6 and lower the pH to 5.5, and say adjusting to the new pH needs a certain mutation or sets of mutations to have occurred, how much time (or how many generations) will it take for 99% of the bacteria population to have developed this mutation ?
If mutations are totally random - you can statistically calculate that say in N generations x% (+/-z%) of the population would have developed that trait. Then you can experimentally verify this. There could be other experiments to test randomness - I pointed some out in my previous post.
I have not seen a single experimental proof demonstrating that mutations are random.
You would have to know all of the possible mutations to solve that problem in advance (of which there could be many), and calculate based on how many steps each of those methods take.
Even these articles do not suggest as far as I could see, that the mutations are non-random (i.e. targetted somehow to specific results). Simply, if you try to grow bacteria on a medium of only lactose nutrients, then the cell(s) that develop lactose processing mutation will of course dominat the population.
Keep in mind the scale of evolution - in the course of a million years, with organisms that have hundreds or thousands of potential offspring, a generation or many every year - there’s a hell of a lot of random happening and a hell of a lot of selection possible to build on.
it doesn’t take a lot of time… how many generations to get Clydesdales, daschunds? The muscular hefty types of the Samoans or Maori have only been selecting for 1000 years or less, and the culling and breeding selection a little less explicit… or the natives adapted to the Andes altitudes over less than 10,000 years? Even a 0.1% advantage can add up over a thousand generations.
There are many other mechanisms/explanations - see here. Bottomline, to me is that duress was influencing an organism’s future mutations (some explanations say it was only influencing the number of mutations and not the particular kinds of mutations)
The meaning of non-random is NOT RANDOM - it does not imply targeted. If I had 10 coins and tossed them 1000 times, and plot the frequency of heads, I will have the highest frequency of having 5 heads. If I got 8 heads with the highest frequency - that’s non-random.
S
[QUOTE=md2000]
Keep in mind the scale of evolution - in the course of a million years, with organisms that have hundreds or thousands of potential offspring, a generation or many every year - there’s a hell of a lot of random happening and a hell of a lot of selection possible to build on…
[/QUOTE]
All good points - but all points not germane to the original question. Why are you linking together mutation and selection ? I agree that natural selection works. I am questioning whether mutation is totally random. Show me a cite where mutations were measured (independent of natural selection) and shown to be random.
Sorry - by non-random vs. random I mean the mutations would not be in some way producing specific (i.e. “targeted” or “directed”) changes.
The link is a bit beyond me, but does seem to say that MMR activity (error repair?) is reduced in nutritionally deprived cells, implying that a lot more errors/mutations happen in cells that also have the LAC+ mutation. This just seems to be science talk and deeper detail for what I quoted earlier, that stress produces more frequent mutations. the stresed cells hav a lot of mutations, and LAC+ happens to be one. It bears out the preliminary observation, that there are proportionally more mutations generally in LAC+ than the normal LAC- population.
Since it’s almost impossible to separate mutation and selection, as both are always happening in tandem - I suppose the only real world example would be the “age test” of genetic drift. The separatedness of a population is usually measured by the difference in their genetic material in the unimportant or “junk” DNA. The assumption is that this changes at a constant rate, i.e. regular rate of mutation.
I suppose finding counterexamples to this “reguar rate of change” assumption, or correlating drift with stress, would be a good test of your question.
Right. And in the case where the mutation is already known and its “fitness” can be evaulated, there’s a field called “population genetics” that covers how to calculate how fast the mutation spreads. In the simplest cases the math is actually pretty simple, but the simplest cases probably aren’t found in nature much. The math works out differently for sexual vs. asexual reproduction. I don’t know how it might take into account exchanging of DNA as bacteria do, I’ve only encountered it briefly in an intro textbook, so only the simplest cases were covered. But I bet you can learn a lot by googling the term, if you’re interested.
One of the odd lessons of population genetics is that what they call “genetic drift” is really more what engineers might call “genetic inertia” – that stable populations tend to homogenize rather than diversify. That’s IIRC, and I might not.
Population geneticists have worked out tons of these calculations that you say you can’t find anywhere. The only thing I’m aware of that supports what you’re talking about is a finding in E. coli. It’s been shown that one of the stress-response genes is a variant DNA polymerase that has reduced proofreading capability. In other words, when things go to shit, E. coli starts reproducing its DNA in a way that increases the mutation rate.
I am unaware of any evidence in any system that suggests directed mutations - that is, deliberate introduction of mutations that should be adaptive. Indeed, even suggesting a thing brings up a host of complex problems that don’t have any clear resolution. Like, how can an organism’s DNA “know” that there’s another state out there that would be better? If such information was there, why wouldn’t the organism simply adopt that state to begin with? DNA doesn’t know what it codes for, or what that thing does in the larger context of the environment.
In recent years I have been hearing more and more about frog species becomming extinct do to environmental factors. I keep wondering if at some point we will start to see a resurgence of some of these frogs do to mutations allowing them to adapt to the new environment. Years ago I did an experiemnt with guppies involving increased levels of chlorine in the water. The experiement was informal and not carefully monitored but it did rather quickly within a few years produce guppies somewhat more resistant than the original batch. I eventually went too far and killed off the entire experiment ending it abruptly.
I thought of another example that happens every day in all of our bodies - our immune systems. As you may know, B and T cells randomize their genes that code for antibodies or T cell receptors respectively. This provides a huge repertoire of immunity. When a B or T cell is activated because they recognize some antigen in the body, they begin to replicate like crazy, and the antibody gene (let’s just focus on B cells for simplicity) undergoes a process called somatic hypermutation. Basically, the cell “knows” that there’s something out there that the antibody can bind to, so it starts to mutate the antibody’s shape, hoping that it will find a new, similar shape that will bind the antigen better. So you end up with a clonal population of cells, each making a similar but slightly differently shaped antibody. Each cell’s replication rate is controlled in part by how strongly the new version of the antibody binds to the antigen. This produces a mini evolutionary process within your body, as mutations are introduced and selected very quickly with each cellular generation.
Now, if adaptive mutations were a thing that happened, this would be an absolutely ideal place for it to happen. The vast majority of the cells undergoing somatic hypermutation will die off, because their mutations aren’t helpful. If the mutations could be directed and adapted, this could be an incredibly efficient process. But that’s not what we see. Instead we see mutations that are more common, but still totally random. And that’s what we see everywhere in biology.
Of course, in a real organism, mutation is not completely random, but – aside from the one finding in E. coli– there’s no reason to think it’s non-random in any interesting way.
For instance, there are certainly places on a real DNA strand where, due to the specific shapes of the DNA-protein complex and the various copying enzymes, substitution mistakes are going to be more common and/or particular substitutions are going to be more common. Certainly the same is true for other kinds of base-pair copying errors – certain base-pair sequences are going to be more likely to be omitted or have an instertion or whatever. But again, only because of the shape and dynamics of the particular base-pair sequence, completely unrelated to whatever (if any) function the base-pair sequence happens to have.
Or, if you’re looking just at protein production (only a part of DNA’s function, but an important part), because of the genetic code, some base-pair substitutions don’t change the amino acid being encoded. So even if base-pair substitutions were completely random, there would be a bias in the resulting actual amino acid substitutions. And of course, if you’re looking at the phenotype, many amino acid substitutions don’t affect the function of the protein, so you’re not going to see completely random changes to protein functions, either.
And then of course, many mutations are going to be so quickly fatal that you won’t even see them at all – in mammals, for instance, there might be a miscarriage so early that the fertilized egg was never detected.
So sure mutations aren’t completely random, but what’s your point? Biologists would all admit that on these uninteresting levels, mutations aren’t random. But that doesn’t have any real implications for evolutionary theory.
Just curious then…
Do most other mammals have as high a risk of miscarriages?
It’s my impression that when it’s “breeding season” that various animals - sheep, cows, whatever wild animals, etc. - it seems the hit vs. miss situation for becoming pregnant is a lot more successful than it is for humans.
If basically these animals don’t have the miscarriage rate (assumption?), breding season is apparently mostly successful, then how come early miscarriage is always quoted as a way to dispose of unsuccessful mutations when we talk about DNA mutations?
That’s interesting, but not necessarily due to new mutations in your population of guppies. If your population had genetic diversity, your experiment would have selected for genes already present at a low level in the population. You’d be able to get to a certain level of chlorine resistance fairly easily, but then getting beyond that would be much harder. You’d need to wait for a lucky mutation to give more resistance.