Some mutations sound too good to be true

Note:

I am releasing this thread into the Biological Evolution science forum and restoring Faith's priveleges here. I do so with trepidation.

​

To those who choose to participate on this thread (including Faith): stay focused on the data and on the science. I will keep all posters to a higher standard than usual. I will be watching this thread closely, and will have a "quick hook" on removing posting priveleges to this thread for those who fail to contribute appropriately.

​

Please keep this in mind when posting here. We're here to talk science; nothing more, nothing less. Thanks.

--AdminBen

---

I'd like this to be a question and answer thread, and request that scientists responding keep jargon, math and dependence on external links (as substitutes for their own explanations) to a minimum.This question comes from the thread, Is bacterial resistance really due to mutation?.It's been made clear on that thread why mutation is considered to be the only explanation for the development of resistance: When you isolate a single bacterium and grow a population from it, you are supposedly eliminating built-in mechanisms of variation, and that leaves mutation as the explanation.It was also explained on that thread how to be sure that you are starting with a nonresistant bacterium. So I understand how they arrive at the conclusion.Since this resistance develops rather predictably, however, this still suggests to my mind a built-in process of some sort -- or even something frankly Lamarckian in nature. Mutation, which I've understood to be basically an anomaly in the process, random and unpredictable, and often lethal, hardly seems a reliable method for developing any kind of survival-enhancing capacities in anything.This raises questions about exactly WHAT mutation IS. It's beginning to sound like there is a kind of "mutation" that really IS a normal built-in process, but that the mechanism for this is not yet understood. I mean, why should a random change in a gene so predictably lead to a bacterium's resistance? Shouldn't the probability be astronomically low-to-nonexistent?This message has been edited by Faith, 09-19-2005 02:29 AM

The predictability is a statistical one, in that you can predict what percentage of the bacteria acquire resistance in a given period of time, but not which particular bacteria. You can't pinpoint a certain bacterium and infer from the mechanism that it will or will not acquire resistance. Random mutation and natural selection is exactly the mechanism that fits the bill of this kind of predictability. And indeed it isn't. Many mutations lead to the demise of the organism. Only those mutations that are advantageous remain in the gene pool. And besides, there is no preset goal in evolution. Survival of the fittest just happens, and the fittest is not specified in advance.Buying a lottery ticket isn't a sure fire way of acquiring a lot of money, but some people do win the lottery. No, because the amount of trials is also staggering. The number of bacteria in a culture is enormous. If only one out of a billion acquires resistance, that's the one that's going to survive and be the ancestor of the generations to come.This message has been edited by Parasomnium, 19-Sep-2005 03:43 PM

---

We are all atheists about most of the gods that humanity has ever believed in. Some of us just go one god further. - Richard Dawkins

It sounds like you are thinking of mutation as big changes. I agree, they sometimes are big changes and those changes are over lethal. But mutation can also be tiny changes. As reversal of a single letter in the code is a mutation. It may have no noticable effect.The reason the process is predictable is because we are controlling one of the variables - the force of selection.The experiment in it's simpliest terms is this - Take a billion bacteria, kill all but the strongest ones, breed those, repeat.

Plaease bear in mind that this resistance is highly variable, there are a number of different genetic solutions for resistance to even one type of antibiotic. Mutation is arguably an anomaly in the process of DNA replication, but it is a persistent one. In fact in its many discrete forms it is the most significant method for the introduction of genetic novelty, which is the absoloute pre-requisite for devoloping any new kind of capacity whatever.Mutation is 'inbuilt' in the process of DNA replication, due to DNA replicating systems not being 100% perfect. A simple 100% perfect replicating system would be very efficien at multiplying itself but would never have any capacity for novelty or developmentd.I'm not sure if anyone else pointed this out but inheritance in Bacteria can quite easily be thought of as Lamarckian. 'Characteristics' can be acquired by lateral gene transfer allowing for a much quicker spread of a beneficial gene contained on a plasmid through a bacterial population than for a beneficial gene through a sexually reproducing population such as humans. Any mutation occuring through misrepair rather than an error during mitosis is arguably an acquired characteristic and will be inherited by any descendants of that bacterium.When there is no distinction between somatic and germ line lineages Lamarckian phenomena become quite acceptable.TTFN,WK

In a way, it IS built-in. If populations didn't have a way of changing through time, extinction would have been very quick.Staphylococcus' genome is 85% coding (Source). Any mutations will probably have some affect on the bacteria population...but the interesting thing, which could be explored in this thread is that the antibacterial part of of the genome resides within plasmids, which from what I can see are often only a few thousand base pairs long.So, you're right, these plasmids may indeed be a 'built in' mechanism for assuring that bacteria populations survive. Bacteria and their ancestors have been fighting a war for billions of years and the flexibility of prokaryotic structure can be a massive advantage.Then again IANAMb (I am not a microbiologist), so I could be barking up the wrong tree.

There doesn't seem to be a lot to be said that doesn't incur some risk of violating one or the other of these constraints, but I'll stick my neck out with a few comments. It might help to try taking a closer look at the way you view the process. It can be very difficult (for any of us) to completely avoid imposing upon his view of a process (any process) ideas about how such a process should work; about how it might work if we tweaked this or that factor just a smidge; about how it would work if we had designed it, etc. Performing these sorts of projections is what we do. The ability to do this quickly and intuitively serves us well in our daily lives, and the flexibility and creativity we can bring to bear on such problems is testimony to the awesome power of the platform (the human brain) on which these simulations are run. But if science has taught us nothing else, it is that our most intuitive ideas about things quite often turn out to be wrong, and it has been often pointed out that this is particularly true with regard to our ability to intuitively analyze statistical data or to intuitively make predictions regarding probabilistic outcomes.Teleological assumptions tend to get mixed in with observations in subtle and transparent ways -- and this even when one is making a conscious and deliberate effort to avoid that. Here, you don't even seem to be making the effort, and, not surprisingly, the teleological assuptions therefore appear rather glaring.What I'm saying is that you begin by placing the question within a teleological frame, and everything that happens after that takes place within that frame. Mutation is either a failure of something that ->shouldor the 'flaws' aren't really flaws at all, but explicit design features of the process. But on both sides of that false dichotomy lies the assumption that because the process started somewhere and from there, went somewhere else, that it did that because it somehow wanted to.

That "reliable" in there may be your shibboleth, Faith. When ten billion bacteria die due to an antibiotic and one lone mutant survives due to its resistance, "reliable" may not be the right word. And remember, Mama Clam lays a million eggs, and only two, on the average, make it to adult clamhood. "Reliable" and "efficient" aren't good adjectives for reproduction.

Hi Faith,I'm a Christian and a YECer. I've watched several threads involving you with interest.Here are some pretty decent threads. In How well do we understand DNA?, I propose a new, creation-based idea about "random mutation." The discussion was interesting, imo.I think I first began contemplating this "on purpose random mutation" in Evolution is a basic, biological process. The discussion is interesting (concerns your topic) from Message 287 and onward.In What's not Macro about Chlorella v?, you can see some of my ignorance about what evolutionists mean by random mutation and monocultures get wiped away.--JasonThis message has been edited by TheLiteralist, 09-19-2005 04:33 PMThis message has been edited by TheLiteralist, 09-19-2005 04:35 PMThis message has been edited by TheLiteralist, 09-19-2005 04:42 PM

Faith, I am not exactly sure what you mean here. I think there actually a limited number of ways (I recall reading about only 4 or 5) that bacterial resistance works. Is that what you mean: that it's "amazing" that the "random mutations" within a monoculture (culture started from one cell) always reach these same handful of resistance solutions?[if i'm wrong about the number of "solutions", I shall be quickly corrected]--JasonThis message has been edited by TheLiteralist, 09-19-2005 04:41 PM

Thank you. I'll check them out. This has to be a learning process here if I can absorb the information. But it always helps if you can do a bit of translating of the content of a link before you present it.{Edit: Please participate fully on the thread as you feel inspired to. I can use all the help I can get.}This message has been edited by Faith, 09-19-2005 04:48 PM

Yes, I understand this. I have no problem with natural selection once the resistance is established -- that makes perfect sense. The problem is with the "random" in "random mutation." No matter how many billions of chances are involved, it is hard to understand how there could even be ONE beneficial purely random mutation in the lot -- let alone the fact that this beneficial change is encountered in population after population after population of bacteria exposed to antibiotics.Billions of NO mutations whatever, or even billions of useless mutations, make sense, but even one purely random mutation in one bacterium that can protect it from a specific poison seems just plain impossible to me considering the complicated nature of the change required. The amazing fact that resistance occurs with such alarming frequency keeps suggesting something a lot more predictable than the word "random" implies.I just keep thinking that there has to be a naturally-occurring mechanism, a structure, a process, something in the range of molecular/chemical combinations possible, something, that makes this process not so random.If possible, if it's not beyond me, I'd like to learn from this thread more about the actual science that this involves:What all has to be taken into account to compute the probability accurately? This implies the question, What exactly is the structure of the necessary mutation that accomplishes this protective function? Specifically how is the gene altered? What did it used to do that now it does differently? Etc. (Wouldn't there be such a complicated specificity to the change that that in itself should add some enormous number of zeroes to the probability calculation(if that makes sense)? Yes, I understand, but again the probability involved seems off the charts-to-just plain impossible to me. Not to you? Yes, but in that case you know *somebody* will win it because there is a definite stash to be won. I would agree that it seems to me that something similar MUST be the case with mutations as well, simply because useful ones do occur, but this is not how the idea is presented. There would have to a specific positive genetic outcome that could be defined in advance as PROBABLE, not purely RANDOM.Again, the idea that a RANDOM process could achieve a useful outcome in a situation where the far and away most likely outcome is NONusefulness, appears just plain impossible unless there IS a specific predictable calculable "paydirt" involved somewhere in the system that increases the probability somewhere above the impossible. I understand, but one out of a billion is too many when the adaptation required is a specific change to a specific sequence of chemicals that make up a gene. You see what I mean? Unless I'm completely wrong about what mutation does to a gene, and that is the area I'd like to learn more about.

Well, in this case, the links are to other EvCForum discussions. I didn't think about summarizing it. They're unfolding discussions, which would be difficult to summarize (though not impossible).***NOTE***
I edited Message 8 several times: I couldn't get the links right. So, they may have changed from the time you read it, but they lead to the proper messages and threads now. One thread is immensely long, but I link not only to the thread itself, but also to the message where appropriate and related (to this thread) discussion begins.I'm pretty ignorant (though not completely), so people had to break things down pretty good.--JasonAbE: If you find a particular post in one of the threads particularly difficult, ask about it here...and I'll try to decipher it.This message has been edited by TheLiteralist, 09-19-2005 05:21 PM

I don't know what your level of understanding is when it comes to mutations and what they actually are, so forgive me if this is simplistic.I'll start with the structure of DNA, leaving out the sugar backbone of the molecule. The code of DNA consists of only 4 bases, adenine (A), thymine (T), guanine (G) and cytosine (C). Imagine them sitting on the sugar backbone. The backbone is long and the bases run along it's length. The order that the bases occur in is the genetic code.To read the code you take the bases in groups of three, so, for example, you would read atggctgctatattc as atg gct gct ata ttc. Each group of three (called a codon) codes for an amino acid and long chains of amino acids strung together make proteins. If you change one base in a codon, you can change the amino acid it codes for. Occasionally a single base change results in a group of three bases which still codes for the same amino acid as the original group of three (let's call this a silent mutation.) On other occasions a change of a single base results in a group of three bases not coding for any amino acid at all - it may code for "STOP" which usually tells the cell's machinery that the code for the protein has finished and the following code is for something else (remember it's a long chain with many genes on it).Sometimes an extra base is inserted or one base is removed. This will then shift the entire reading frame. In the example above if you remove the first "t", atg gct gct ata ttc will become agg ctg cta tat tc and this may code for totally different amino acids than the original. Additionally it just so happens that atg is the signal for the start of a protein code. The frameshifted version doesn't have that at the start, so the cell has no way of knowing where the protein code starts or what frame the bases should be read in. If you want, I can provide you with the list of which groups of three bases code for which amino acids. This would also illustrate why the genetic code is called degenerate.Anyway, now that you can see that a single change in base can have important consequences, imagine this scenario. An antibiotic targets a specific protein in a cell. The antibiotic manages to damage the protein by attacking one part of the chain of amino acids. Now imagine what would happen if the antibiotic was confronted with a protein that had different amino acids in the part that the antibiotic usually attacks. Sadly for the bacterium, the altered protein probably won't be as efficient as the normal one. In the absence of the antibiotic these mutants wouldn't be able to compete with normal bacteria and so would keep dying out. Add the antibiotic, however, and suddenly those with the altered protein are at an advantage. All the rest are being killed by the antibiotic and so the altered ones have no big bully boys to compete with, therefore they don't die out. However, the bully boys do die out so only the altered ones remain in the population. The antibiotic is the selection pressure which allows the random mutation to persist and thrive.Does this help and would you like concrete examples? I've tried to get the concept across without relying on specific examples to keep the jargon out.Edited to define "codon" and to add a line breakThis message has been edited by Trixie, 09-19-2005 05:32 PMThis message has been edited by Trixie, 09-19-2005 05:48 PM

I think this is your problem right here, Faith.The mutation is question isn't a big deal, to my understanding. Antibiotics attack specific proteins in the cell walls of bacteria. A very minor mutation could change the structure of those proteins ever so slightly as to be immune to (or at least highly resistant to) antibiotics. There is no "desire" or "motive" to adapt such a defense - it's just that the few who have oddly-formed cell walls will happen to be immune.A similar mutation in more human terms would be weebed toes. Now, granted, they don't provide any disease resistance, but they are an incredibly minor mutation that occurs in a statistivcally predictable portion of humans.Another, better example would be sickle-cell anemia. It's just a mutated deformation of red blood cells, a slight change in the structure of the cells themselves. But it also happens, purely by chance, to make those cells immune to malaria infection. Again, no intent involved, no motive to adapt a defense. It just occurs as an occasional genetic mutation, and in this case the mutation has a benefit (of course, sickle-cell anemia has a darker side too, and causes many other medical issues, but at least you don't die of malaria. This is why it's not common in areas unaffected by malaria, but is nearly universal in areas where malaria runs rampant - those without sickle-cell have all died, and only those who carry the "defect" survive to reproduce).Remember also that we are talking about a population of billions, and many many generations removed from the original progenitor cell. The little tiny changes that occur every generation (observable in every organism right up to us) can occasionally add up over many iterations to a significant change. The only process is mutation, and the fact that it doesn't take a very big change to prodice such an effect. The immunity to antibiotics is not, so far as I understand it, universally the same in all samples. That is to say, the structure difference may not be the same or even similar between different samples - they just all produce the same observed effect.Basically, with so many bacteria in the tested sample, we can be sure (assuming that mutation happens) that at least some of the bacteria should have minor structural irregularities that make them more resistant or even immune to antibiotics (because antibiotics are so targeted at a particular part of the cell, and won't work properly if the cell wall structure changes). In the same way we can statistically predict that over a sufficiently large population of humans, some will be born with a minor birth defect like webbed toes. Since we see that the prediction is confirmed in experimentation (some bacteria do wind up being resistant or even immune, and some people are born with webbed toes, etc), we can conclude that mutation is an observed fact. We further determine that, since most of the bacteria do not become resistant or immune, it is a random process.

---

Every time a fundy breaks the laws of thermodynamics, Schroedinger probably kills his cat.

I feel I should answer this before I have really thought it through just to say thanks, because it is a helpful explanation of the process. I've read quite a bit on DNA and seen some really nice animations of its structure, but always find it hard to keep elements of it in mind, so whatever I've already read about can always bear much repeating. I don't recall ever running across the idea that the code is to be read in groups of threes, however, although I've run across the term "codon" -- apparently without understanding it -- so that is new and useful information, as well as the specific function of a particular group, for starting or stopping a sequence. All very informative, thanks.I've seen the lists of amino acids I think but that too could use repetition.Now I need to think through the example you gave. That will take some work. This whole thread is going to take work.