Poll: Does the Theory of Evolution have practical applications?

[Alan Kleinman]

But the Kishony and Lenski experiments are credible evidence and these experiments show that it takes a billion or more replications for each evolutionary adaptive step.

Well, you are starting with clonal isolates. Most populations already start with significant diversity. Wouldn't replications = population x #number of generations more accurately reflect a general case? Yet you refer only to a list of population sizes in a different post above.

Even if you are starting with a single founder bacterium, the population very rapidly diversifies. For example, in the Kishony experiment, he uses e coli. Assume the genome length is 5e6. When that founder and descendants have done 200 replications, there will have been 1e9 base replications. For a mutation rate of 1e-9, that means on average there will be one variant with a mutation somewhere in the genome and 199 clones of the original founder. As the population grows (more replications) another variant with a mutation somewhere in the genome will occur every 200 replications. With a billion replications, you will have on average 1 member in that population with a mutation at each site in the genome. You will have 5E6 variants and (1E9-5E6) exact clones of the original founder (on average). Lenski's founders were far from being exact clones. And you are correct, you have to sum over generations the number of replications in each generation to get the total number of replications of a given variant in an evolutionary lineage. But you don't sum all variants because a mutation that might be beneficial for one variant in a population might be neutral or detrimental to a different variant in that population.

If a particular population doesn't have sufficient numbers and then is subject to selection pressure, those populations often go extinct.

Your examples also use rather extreme selective agents. it might take many generations for a population to go extinct for more typical pressures, say "it's a bit colder than usual this winter". And indeed, most populations do go extinct.

Kishony's team has to limit the concentration of antibiotics such that only a single beneficial mutation gives improved fitness. If the concentration is too large (requiring more than a single mutation to give improved fitness, his experiment doesn't have the carrying capacity to give the population size needed to give those two or more mutations. His population would run out of nutrients in the drug-free region before the drug-resistant variant ever appears.

As I pointed out in post #354, Lenski published a paper where his founder bacteria showed drug-resistance to numerous antibiotics even though his bacteria were never exposed to antibiotics. Variants occur simply occur because mutations occur. The larger the carrying capacity of the environment, the larger the population size, and the larger the population size, the greater the diversity of the population. Lenski's founders were grown in a much larger carrying capacity environment than used in his experiment. That's why these drug-resistant variants appeared even though these populations were never exposed to antibiotics.

 

[pitabread]

Edit to add: also, the selection pressures here are rather extreme; most of these agents will kill/inhibit 99.9% of their targets. Do you find anything in nature that extreme outside of asteroid impacts?

I was wondering this too. He seems to be assuming fitness effects of mutations / selection pressures are constant. Which is obviously not the case in nature, since selection pressures are environmentally dependent.

Even mutations are not strictly beneficial or deleterious, as it can be entirely dependent on environmental context.

 

[Alan Kleinman]

I was wondering this too. He seems to be assuming fitness effects of mutations / selection pressures are constant. Which is obviously not the case in nature, since selection pressures are environmentally dependent.

Even mutations are not strictly beneficial or deleterious, as it can be entirely dependent on environmental context.

I don't make that assumption. What I'm doing is computing the number of replications necessary for a particular mutation to occur, or for two particular mutations to occur, and so on. Therefore, if a particular selection condition requires a single mutation to give improved fitness, that gives the number of replications required to give a reasonable probability of that single mutation occurring on some member of the population. If the selection conditions require 2 mutations to give improved fitness, the appropriate calculation gives the number of replications necessary for that event to occur and so on. What I think you are having trouble seeing, for a mutation rate of 1e-9, a billion replications of that genome will on average give some member of that population a mutation at every possible site in the genome. Most of those variants will have neutral or detrimental mutations but as illustrated in the Kishony experiment, he will have a variant with a beneficial mutation for Ciprofloxacin and a variant with a beneficial mutation for Trimethoprim. The drug used in the particular experiment will demonstrate which was a beneficial mutation because that variant will be able to grow in the region with that particular drug.

 

[pitabread]

I don't make that assumption.

I was thinking in the context of your claims re: historical evolution.

 

[Alan Kleinman]

I was thinking in the context of your claims re: historical evolution.

The probability of rolling double 6s is the same past, present, and future.

 

[sesquiterpene]

And you are correct, you have to sum over generations the number of replications in each generation to get the total number of replications of a given variant in an evolutionary lineage. But you don't sum all variants because a mutation that might be beneficial for one variant in a population might be neutral or detrimental to a different variant in that population.

Thanks, that was exactly my point, although I didn't express it all that clearly. If you look at a population numbering in the millions, they might sample all of the single point mutation space in a thousand generations - might that actually be enough?

I'd like you to consider the absolute extreme in population size. The population of the bacteria Pelagibacterales is estimated to be 2.4 × 10^^28. That's an amount of almost a million moles! (yes, I'm a retired chemist). I'm too lazy to do the math this late at night, but given some idea of the mutation rate, genome size and replication rate, I'd suspect that this organism explores it's entire single mutation space in a matter of microseconds, the double mutation space in a few minutes, and triple mutation space in days. Well, actually I'd guess I'm off a few orders magnitude in any particular direction.

At any event, many, if not most, of the adaptations we've inherited might well have been the results of evolution occurring in vast populations in the distant past. How can your calculations prove that this isn't true?
Lists of organisms by population - Wikipedia

 

[FrumiousBandersnatch]

Now consider the claim made in the ToE that reptiles evolve into birds. That means a replicator that doesn't produce feathers must accumulate the mutations that would enable it to produce feathers. But it doesn't stop there, birds and reptiles have different respiratory systems, different circulatory systems, different excretory systems, different metabolisms, different musculoskeletal systems (pneumatic bones and flight muscles), and for each evolutionary transitional step from one form of replicator to the other will take a billion replications for each evolutionary transitional step and that is under the best of conditions when only a single selection is acting on the population at a time. If multiple selection conditions are acting simultaneously, the number of replications for each adaptive step goes up exponentially. The ToE simply doesn't have the population sizes necessary for such evolutionary transformation to have a reasonable probability of occurring.

It's worth pointing out that the therapod dinosaurs thought to be the ancestors of birds have been shown to have had many, if not most of the physiological and anatomical developments you mention - see Similarities between Birds and Dinosaurs:

Similarities:

“Pubis (one of the three bones making up the vertebrate pelvis) shifted from an anterior to a more posterior orientation and bearing a small distal “boot”.
Elongated arms and forelimbs and clawed manus (hands).
Large orbits (eye openings in the skull).
Flexible wrist with a semi-lunate carpal (wrist bone).
Hollow, thin-walled bones.
3-fingered opposable grasping manus (hand), 4-toed pes (foot); but supported by 3 main toes.
Reduced, posteriorly stiffened tail.
Elongated metatarsals (bones of the feet between the ankle and toes).
S-shaped curved neck.
Erect, digitgrade (ankle held well off the ground) stance with feet postitioned directly below the body.
Similar eggshell microstructure.
Teeth with a constriction between the root and the crown.
Functional basis for wing power stroke present in arms and pectoral girdle (during motion, the arms were swung down and forward, then up and backwards, describing a “figure-eight” when viewed laterally).
Expanded pneumatic sinuses in the skull.
Five or more vertebrae incorporated into the sacrum (hip).
Straplike scapula (shoulder blade).
Clavicles (collarbone) fused to form a furcula (wishbone).
Hingelike ankle joint, with movement mostly restricted to the fore-aft plane.
Secondary bony palate (nostrils open posteriorly in throat).
Small, possibly feathered dinosaurs were recently found in China. It appears that many coelurosaurs were cloaked in an external fibrous covering that could be called ‘protofeathers’. * ”
(1)* Even more recent studies have proven that certain theropods infact, had feathers and not just protofeathers.

 

[pitabread]

The probability of rolling double 6s is the same past, present, and future.

Which is irrelevant when we're not dealing with dice.

 

[VladTheEmailer]

Which is irrelevant when we're not dealing with dice.

It is important because that is a major flaw in his argument. Given an infinite number of chances, the odds are 100%.

 

[Alan Kleinman]

Thanks, that was exactly my point, although I didn't express it all that clearly. If you look at a population numbering in the millions, they might sample all of the single point mutation space in a thousand generations - might that actually be enough?

I'd like you to consider the absolute extreme in population size. The population of the bacteria Pelagibacterales is estimated to be 2.4 × 10^^28. That's an amount of almost a million moles! (yes, I'm a retired chemist). I'm too lazy to do the math this late at night, but given some idea of the mutation rate, genome size and replication rate, I'd suspect that this organism explores it's entire single mutation space in a matter of microseconds, the double mutation space in a few minutes, and triple mutation space in days. Well, actually I'd guess I'm off a few orders magnitude in any particular direction.

At any event, many, if not most, of the adaptations we've inherited might well have been the results of evolution occurring in vast populations in the distant past. How can your calculations prove that this isn't true?
Lists of organisms by population - Wikipedia

That's right again, a constant population of a million each member replicating every generation will give a billion replications over 1000 generations. That would give enough replications for a single beneficial mutation to a single selection pressure. But we don't have to imagine that situation. The Lenski populations reach a value of 5e8 every day starting from an initial population of 5e6 and that experiment has been going on since Feb. 1988, over 10,000 days and 73,000 generations. He has two selection conditions acting on his population, starvation and a daily random sampling that removes 99% of the population before the next day's growth (bottlenecking). Every possible mutation that his bacteria could get has happened many times over. What has to happen in his populations, the most fit variant at a given state of the experiment drives the less fit variants to extinction and then some member of that most fit variant gets another beneficial mutation making that member the founder of the next most fit variant in that lineage. If you want to see how to do the math for that evolutionary process, check out this paper:
Fixation and Adaptation in the Lenski E. coli Long Term Evolution Experiment
The selection conditions for the Lenski experiment have remained constant all these years. Real environments are not so kind.

As far as Pelagibacterales are concerned, with populations that huge, you will have members of every variant many times over. The marine environment will select for particular variants based on temperature, osmolarity, salinity, and probably many other physical and chemical conditions. But any improvement in fitness for a particular lineage will still be contrained by the probabilities of accumulating a particular set of mutations in a lineage. So, for example, let's say a subset of this population is living in warmer waters and is carried into cooler waters putting those variants under thermal stress. The most fit variant in that subset that survives and replicates will get mutations and some of those mutations will give improved fitness to that cold stress. But let's say the bacteria are moved into a different environment where there is not only a temperature change but also a change in osmolarity and salinity. The multiple simultaneous selection pressures mean that it will take many more replications to adapt to these simultaneous selection pressures. And obviously, if the stress is too severe, that subset will be driven to extinction.

And to your last point. What vast populations are you talking about? And what environment are you talking about that selected for the traits that we inherited? Consider a particular example, the beta-keratin gene. Are you saying that some replicator millions of years ago evolved the beta-keratin gene that was passed on to reptiles to produce scales and passed on to birds to produce feathers and to humans to produce hair and fingernails? It takes a lot more than a beta keratin gene to produce scales, feathers, hair, and nails. It takes the genetic control system to turn these genes on and off and make these proteins grow in the correct position and form. Explain how that happens by the accumulation of a set of mutations and what the selection pressures are that would do a genetic transformation like this.

 

[Alan Kleinman]

It's worth pointing out that the therapod dinosaurs thought to be the ancestors of birds have been shown to have had many, if not most of the physiological and anatomical developments you mention - see Similarities between Birds and Dinosaurs:

​

So, therapods have the same respiratory, circulatory, excretory, metabolism, musculoskeletal system,... as birds. That's a thought. Did it ever enter your mind that therapods might be an extinct species of bird?

 

[Alan Kleinman]

Which is irrelevant when we're not dealing with dice.

So you think that DNA evolution worked differently in the past than it does today? Why don't you explain to us how DNA evolution worked in the past? In fact, why don't you explain to us how DNA evolution works today by giving us your explanation of the Kishony and Lenski experiments.

 

[Alan Kleinman]

It is important because that is a major flaw in his argument. Given an infinite number of chances, the odds are 100%.

Even Pelagibacterales doesn't have an infinite population. And Lenski's experiment has had about 5e12 replications and in that time there have been about 100 adaptive mutations

 

[pitabread]

So you think that DNA evolution worked differently in the past than it does today?

That's not what I'm suggesting. I'm suggesting that your claims of post hoc probabilities are irrelevant when dealing with circumstances that have already unfolded.

If you want to keep beating the drum, have at it. It's not going to go anywhere.

 

[Alan Kleinman]

That's not what I'm suggesting. I'm suggesting that your claims of post hoc probabilities are irrelevant when dealing with circumstances that have already unfolded.

If you want to keep beating the drum, have at it. It's not going to go anywhere.

I published the probabilities that govern the Kishony experiment before the experiment was run. And there is no real, measurable, and repeatable empirical evidence that contradicts this math. But if you can present contradictory evidence, have at it.

 

[pitabread]

I published the probabilities that govern the Kishony experiment before the experiment was run.

Once again, not talking about the Kishony experiment. I'm still referring to your "reptiles evolve into birds or fish evolve into mammals" claim from earlier.

 

[Alan Kleinman]

Once again, not talking about the Kishony experiment. I'm still referring to your "reptiles evolve into birds or fish evolve into mammals" claim from earlier.

So you think that if I had published this math after Kishony performed his experiment, it wouldn't apply.

 

[Ken-1122]

You should be concerned because Lenski is performing one of the best evolutionary experiments that demonstrate the physics and mathematics of evolution.

Again; what's preventing your ideas from being taken seriously by those in the scientific community?

 

[Alan Kleinman]

Again; what's preventing your ideas from being taken seriously by those in the scientific community?

It's only the ToEite community that doesn't take this seriously, there are scientists that understand this math and they publish this work.

 

[pitabread]

So you think that if I had published this math after Kishony performed his experiment, it wouldn't apply.

I haven't referred to the Kishony experiment. I repeatedly told you what I was referring to. If you are having this much trouble following along then this will be a rather short conversation.