Journal club time: paper by Sanford et al: The Waiting Time Problem in a Model Hominin Population. I’ve pasted the abstract below.
Have at it guys 🙂
Background
Functional information is normally communicated using specific, context-dependent strings of symbolic characters. This is true within the human realm (texts and computer programs), and also within the biological realm (nucleic acids and proteins). In biology, strings of nucleotides encode much of the information within living cells. How do such information-bearing nucleotide strings arise and become established?
Methods
This paper uses comprehensive numerical simulation to understand what types of nucleotide strings can realistically be established via the mutation/selection process, given a reasonable timeframe. The program Mendel’s Accountant realistically simulates the mutation/selection process, and was modified so that a starting string of nucleotides could be specified, and a corresponding target string of nucleotides could be specified. We simulated a classic pre-human hominin population of at least 10,000 individuals, with a generation time of 20 years, and with very strong selection (50 % selective elimination). Random point mutations were generated within the starting string. Whenever an instance of the target string arose, all individuals carrying the target string were assigned a specified reproductive advantage. When natural selection had successfully amplified an instance of the target string to the point of fixation, the experiment was halted, and the waiting time statistics were tabulated. Using this methodology we tested the effect of mutation rate, string length, fitness benefit, and population size on waiting time to fixation.
Results
Biologically realistic numerical simulations revealed that a population of this type required inordinately long waiting times to establish even the shortest nucleotide strings. To establish a string of two nucleotides required on average 84 million years. To establish a string of five nucleotides required on average 2 billion years. We found that waiting times were reduced by higher mutation rates, stronger fitness benefits, and larger population sizes. However, even using the most generous feasible parameters settings, the waiting time required to establish any specific nucleotide string within this type of population was consistently prohibitive.
Conclusion
We show that the waiting time problem is a significant constraint on the macroevolution of the classic hominin population. Routine establishment of specific beneficial strings of two or more nucleotides becomes very problematic.
Well, it’s upon you to do the work that demonstrates that has value. Will you be doing the work?
I think you have confused yourself by considering number of species as a measure of evolutionary change.
Evolution is best considered as what happens to a lineage over time. Speciation is best considered as a series of bifurcations of that lineage. A lineage adaptively evolve a great deal, but not bifurcate. Equally a lineage could bifurcate but the wo offspring branches might no undergo any adaptive evolution.
I think you need to look at adaptive change (and indeed, non-adaptive change) within lineages instead of speciation if you want to understand what evolution can and can’t do.
Forget about species. Although Darwin’s book was about the ORIGIN of species, it isn’t about speciation, but about adaptive evolution – about why different lineages have different features, not about why they diverged.
There is a peer-reviewed paper that shows AVIDA is front-loaded and when the parameters are corrected to reflect reality nothing new evolves.
The effects of low-impact mutations in digital organisms Chase W. Nelson and John C. Sanford, Theoretical Biology and Medical Modelling, 2011, 8:9 | doi:10.1186/1742-4682-8-9
I predict this will be hand waved away.
As an argument, this is barely better than the old creationist canard that Evolution is wrong because nobody has ever observed life spontaneously appearing in a sealed jar of peanut butter.
What the experiment has shown (and shown repeatably,) is that if a dozen petri dish sized populations are exposed to an environment where there is only one niche and no competitive species, one of those populations will mange to fill that niche. In such a tiny sample with no selection beyond the elimination of fatally deleterious mutations, at least one new feature arises. (There may be others whoes effect is not yet apparent:- expose those 12 strains to a different environment and one may already be genetically equiped to dominate it) The real world experiment has ten to twenty orders of magnitude’s more environment to operate in and almost infinitely more opportunity for selection. Should be enought to see significant changes even on a YEC timescale.
That’s not entirely true. What we do know is that if the population survives and there is a mutation within reach, the population will eventually find that mutation.
There is no general rule that challenged populations will fill a niche. If there were such a rule, antibiotics would never work.
OH MY SCIENCE!
That’s twice in the same thread I agree with Petrushka. Studly good insight.
How do you tell bacteria what the target is?
Joe Gallien is trying to post as a sock called “Frankie”, BTW. The idiot posted the same 2011 paper abstract oh his blog. How stupid is that!
I can’t believe you’ve just ignored all the comments on the previous page and posted this instead. How do you live with yourself?
Frankie,
Welcome, Frankie.
This is Joe G, who is banned.
Richardthughes,
Ah, that is quite a “coincidence”. Unless Lizzie wants to be proactive, I’m inclined to give Frankie a bit of rope.
ETA On looking at Joe’s blog,that will be a very short bit of rope!
Agreed, but my understanding is that the mutations behind the adaptions was understood, and required two separate changes. Once the original (neutral) mutation was determined, reviving the ancestral population it which the first mutation became fixed allowed the experiment to be re-run from that time, and similar citrate metabolising ability re-appeared. In the context of stcordova’s original point, this was not a freak event, but an almost inevitable outcome in this case.
Alan Fox,
en
fair enough. Joe’s not smart enough to not use his signature phrases: IDGuy, Jim, JohnPaul and Virgil where all easy enough to spot.
Richardthughes,
I’m on red alert.
No hand-waving required. All that shows is that if you set the parameters so that there is very little selection then you complex traits don’t evolve. In other words, if you supply the conditions that Darwin proposed (some variants with greater probability of reproduction than others) adaptive and cumulative evolution occurs. If you don’t, you don’t.
Elizabeth,
I’d read the paper when it first came out and just reread it quickly to refresh my memory. This struck me as one of the authors’ two major points (the other being that Avida doesn’t always track observed biological behavior):
The fitness score for the nine logic operations tested in Avida varies from a doubling of fitness for simple operations to 32 times fitness for the EQU operator. The paper argues that this range of values is unrealistic and is the source of Avida’s success.
If I have the time and no one else has done so, I’ll fire up Avida and see how often EQU evolves with a narrower range of fitness multipliers.
Yeah but we don’t have time to just sit back and wait it out, if you want a reflection of nearby phenotypical space, you have to actually explore it pretty thoroughly. Normally we’d make a simulation to speed it up, but if the simulation is supposed to be really accurate, we simply don’t have the computer horsepower.
Remember, you asked for estimates similar to those produced my Mendel’s accountant. The average waiting time for adaptations requires X mutations. This requires having thorough knowledge of phenotypical space, something we can’t do on computers. And the real world just isn’t going along very fast. So sorry, but what you want can’t be done.
Actually Lenski didn’t really show this. It was more a case copying the existing digestive function and putting the copy under a different controller so it worked under previously prohibited circumstances.
By the way, in the Lenski experiment over 600 beneficial mutations have been fixed in the population. We don’t know what most of these mutations do that improves fitness, but what we DO know is THAT they improve fitness. So strictly speaking, adaptations requiring 2 or more mutations evolve REMARKABLY fast. If this experiment is anything to go by, we could try to divide the number of generations with the number of adaptive mutations and see if we can get some kind of average of how many generations it takes to evolve an adaptation requiring two mutations.
40.000/600 = 66.7
2*66.7=133 generations.
That would mean that on average, two adaptive mutations are fixed every 133 generations. Now, I don’t think this kind of back-of-the envelope calculation is amenable to extrapolation into large multicellular eukaryote evolution, but it does say something about how frequent “something that works” is in “nearby phenotypical space”. What I do think then on that basis at least, is that “something that works” is pretty often nearby.
How do you know this? We know very little about what all these mutations do, we only have the measured relative fitness of carriers. We know how the citrate function evolved because it is easy to demonstrate that cells can now grow on citrate under aerobic conditions where before they could not, but any one or a few of those 600 fixed mutations could be doing something new we just aren’t aware of what is. Elucidating what the rest of all these mutations actually do has the potential to discover many hitherto not understood “new functions” that we presently only see as an increased reproductive rate as measured by reproductive rate. Doing molecular resolution studies in the same way it has been done for the citrate mutant is difficult work and will require many years of study.
The number of fixed mutations in the human and chimp lineages since their divergence matches pretty well to the neutral mutation rate (which implies most of them are probably neutral). There number of genetic changes that contribute to our differences are probably only a very small subset of the total amount that have happened. I don’t see any particular reason to think the kinds of differences that have evolved are any more unlikely than other combinations would have been.
Why do you think that? What environments are never directly observed but only postulated? It seems to me that natural environments, as in the kinds of environments we find in nature in the wild, or in the human gut where the real E coli lives, are substantially more complex and offer substantially more potential for de novo evolution and adaptation, than the singular unchanging minimal growth-medium at a constant temperature found in Lenski’s flasks.
Why do you think the complexity that evolved in the Lenski experiment evolved at a slow pace? How fast SHOULD it have evolved, do you think, for evolution to be true? Try to do the numbers. Let’s say only 1 new function evolved in the experiment, and 3 previously existing ones were modified. Then we extrapolate to the entire history of life (set it at 3.8 billion years).
3.8 billion divived by 20 years = 190 million new functions, and 570 million modifications of function. And that’s just for a population of roughly 50 billion, passing through a 2 billion bottleneck 1.387 trillion times.
There is estimated to be about 10^30 cells on Earth. So how many 50 billion-size populations are those that could have run simultaneously? 10^30/5×10^9 = 2×10^18
Multiply each of those lineages by 190 million new functions and we get 3.8×10^28 new functions evolved in the history of life. That’s a very very big number. It seems to me the Lenski experiment overwhelmingly supports the evolution of complexity, if we allow one new thing to evolve every 20 years, while three others are modified.
What complexity has evolved in the human and chimp lineage that is vastly beyond what has evolved in the Lenski experiment? Last I checked, we have no new flagellums, or new complex enzymatic functions, or altered membrane transporters or anything. The only thing that seems to have changed are some regulatory elements (which usually only require one or two mutations in a single controller here and there) controlling certain developmental timings, such that limbs and body hair are slightly altered in shape and proportion, and brain development is more pronounced particularly in our frontal lobes. Why is this something that must have taken a long time to evolve? Why can’t it just be “what was nearby in phenotypical space” among many other possibilities? Clearly both we and Chimps evolved from our common ancestor, and there are multiple subspecies of Chimpanzees and humanity is quite diverse too, and we have quite a number of ancestral fossil sister species, some of which might even be our direct ancestors. So clearly, whatever the exact nature of our common ancestor, there were in fact multiple workable solutions nearby in phenotypical space.
Or I suppose you would protest that. Do you think the genetic differences between Chimps and ourselves constitute an insurpassable desert of nonfunctional lethal genes?
What the hell does this even mean? I can only repeat myself: It seems to me that natural environments, as in the kinds of environments we find in nature in the wild, or in the human gut, are substantially more complex and offer substantially more potential for de novo evolution and adaptation, than the singular unchanging growth-medium at a constant temperature found in Lenski’s flasks.
Why not? How does this even relate to what we are discussing? What complexity is found in hominids that is not found in chimps?
Heat is not complexity. The analogy does not seem to apply at all.
I actually don’t think prokaryotes CAN become more complex than they generally are, without endosymbiosis. But that’s another discussion, read Nick Lane’s new book “The vital question”.
Humanity did not spring from bacteria, we evolved from an ape ancestor not very different from us now. And we probably didn’t evolve much new complexity from that ancestor, if anything at all. In the case of our evolution from the other great ape ancestors, I think this has been more a case of modifications of already existing functions than I believe they have involved new complex molecular machines. I don’t think we have anything new in that sense. Merely we have altered quantities of stuff that had already evolved long before we did ourselves.
I think most of the cellular and molecular complexity we see that baffles and impresses us, evolved in our unicellular (but still eukaryotic) ancestors in vast populations probably over a billion years ago and for several hundred million years. And what has been going on ever since has been mostly a case of rewiring and repurposing all that machinery. Evo-devo.
Sal, you write as if you imagine Homo sapiens have vast swathes of new enzymes, complex multi-protein ribosome-scale machines, membrane transport complexes and so on. We don’t. it’s all the same shit you see in our nearest cousins, meaning nothing new was invented to produce us along the way. Again, relatively few mutations in regulatory reagions have just changed some associations and developmental timings, so limb and body hair proportions have been slightly altered, we have smaller teeth and jaws, but bigger frontal lobes. All still made of the same stuff.
You don’t agree with me. You don’t even understand what I am saying.
Since I am bound by the rules of the forum, I will restrict myself to saying that what you say is untrue. Whether you choose to educate yourself or correct yourself is up to your own character.
If you lower the fitness values to realistic numbers, remember to also lower the probability of mutation to realistic numbers too. Set it to three in every 100 million sites, this is not that far from the rate of mutation in humans.
Oh, please do.
One really interesting thing about AVIDA, however, is that every single rewarded function is IC – can only be reached by neutral steps. So while it might be necessary to “over-reward” the rewarded steps to get them to reward, you can’t reward the necessary intermediate ones. The functions are typicaly more brittle (IIRC) than, say, protein sequences.
But not a general rule.
The general rule is that change will occur. Mostly neutral. Adaptive change is the exception. Another general rule is that “highly adapted to niche” populations will become extinct. Nearly all that we know about are already extinct.
IDiots are looking for some evidence that adaptations are planned and foreseen. Such evidence does not exist.
Lenski’s experiment was designed to track neutral change, not to be a bottled WEASEL. The fact that a multi-step adaptation occurred is icing on the cake, but it was not foreseen or planned or required by the experimental design.
Exactly, and anyone of those specific targets would run up against a “waiting time problem”. However if you don’t care what result you get evolution will work just fine as an explanation.
This is an important difference between us. I would be interested to know if the there is only one sequence that could yield human level intelligence. In general I don’t think we should limit what we look for
Peace
The origin of species is what I’m interested in. It’s too bad that Darwin has nothing to say about that.
peace
You haven’t exhibited any interest in learning about origins.
Bacteria don’t have to know the target. The experimenter is the one doing the selecting/designing . Bacteria just do what they do
peace
Well, as there is no target- setter there is no “you” to “care” is there? The “targets” consist of anything that helps the organisms survive and breed. And if there are a great many of those targets, then you may bump into quite a lot of them quite often.
I’m absolutely sure there isn’t “one sequence”. Human intelligence is in any case fundamentally the same as the intelligence of other animals, particularly our closest relatives – but there will be some sequences that contribute to our specifically human capacities. We even know what some of these are – and there are variants of them. An important set of candidates are those involved in the development of cortical surface – we have a much more “wrinkled” brain than other apes, suggesting that some of the important sequences are those governing how long cortical cells keep being being generated during development.
So some of the key changes may have been relatively small changes to sequences that govern growth – like the ones that govern finch-beak size in the Galapagos finches.
So what target do you think that Lenski set, and how did he set out to achieve it?
He has something to say about it all right – it was his observation that exemplars of organisms different islands had subtly different characters that inspired his big idea, but his main theory doesn’t cover the actual process of speciation, it covers the process of adaptation, which includes differential adaptation down related but speciated lineages. There is plenty of work that has been done since on the process of speciation, and of course, isolation is one of them, as Darwin correctly noted.
Elizabeth, Thank you for your response. A few things to consider:
1- Nick Barton and Willem Zuidema, “Evolution: The Erratic Path Towards Complexity,” Current Biology, Vol. 13, R649–R651, August 19, 2003
They explain that AVIDA is stacked and doesn’t simulate biological evolution.
2- Dr Behe:
3- Dr Lenski:
Dr Behe’s IC did not evolve.
And remember, evolution, per se, isn’t what is being discussed as unable to produce IC. IC only pertains to evolution via natural selection and random genetic drift.
Well that is the root of all that is different in our perspectives. You assume there is no target-setter and I presuppose his existence.
Well I’m absolutely sure that humanity is unique in the universe. How do we go about deciding which of us is correct without looking for other possibilities?
He wanted bacteria that did a better job of metabolizing citrate. He achieved this goal by setting up an environment that would select for bacteria that did a better job of metabolizing citrate.
peace
Is IC under discussion. Perhaps some ID advocate will provide an example of a genetic sequence that could not have evolved through drift and selection. That would, of course, mean providing the actual, detailed history of the sequence, complete with the steps that are impossible through drift.
Anything less is simply spewing bullshit.
Are you saying that isolation is a major cause of the origin of species. So that if we isolated a population in the lab a new species would be likely to arise?
peace
You are mistaken, I’m afraid, Frankie. It did.
Behe has two definitions of IC. The first is the “mousetrap” definition – a feature that will not perform its function if any part is missing.
The second is the one you give.
EQU (and, as I understand it, all the rewarded functions) in AVIDA is a “mousetrap” – it does not work if any component is missing.
But more importantly, it can only be reached be a sequence of unrewarded steps – so it is are also IC by the second definition. Between the last rewarded step and EQU there are not only many unrewarded steps (it differs of course, on every run) but it seems from all the runs I have seen that there is a necessary deleterious step that has to precede the step that leads to EQU (in its various forms).
That is why AVIDA simply falsifies Behe. It shows that IC functions (mousetrap definition) can evolve by high-degree IC pathway (pathway definition) quite reliably. Being IC, or being only accessible via a high-degree IC pathway (including quite steeply deleterious steps) turns out NOT to be in, in principle, a barrier to evolution via Darwinian mechanisms.
The most complex function, EQU, does not evolve if there are NO intermediate rewarded steps. But between each rewarded steps there are many necessary unrewarded, and even penalised steps, and, interestingly, EQU rarely (if ever) evolves by the shortest possible path.
How about a specific sequence that would be unlikely to evolve through drift and selection because of the “waiting time problem” ?
peace
Do you have a citation for that? ;P
It’s a major cause of speciation – the separated lineages will diverge simply due to drift, even if there is no selection. Give them long enough, and bring them together again, and they may not be able to interbreed. This is true of ring species too – populations at the extreme ends of a range cannot interbreed with each other even if artificially introduced, but each can interbreed with its near neighbours.
There are many other factors that can lead to speciation, but it is a separate issue from adaptation. You can have adaptation without speciation, and speciation without adaptation.
Bear in mind, though, that “speciation” only applies to sexually reproducing organisms. Bacteria don’t “speciate” – indeed the word “species” although it is used, doesn’t mean the same thing in bacteria as it does in sexually reproducing organisms.
See the OP 😉
Wolves and coyotes can interbreed does that make them the same species?
Apparently species does not mean what you think it means even in sexually reproducing organisms like wolves and coyotes
peace
You could perfectly well posit a deity that did not have a specific target in mind – would have been happy if we’d turned out as intelligent dinosaurs than intelligent apes.
I’m sure we are unique as well. That doesn’t mean that there is “one sequence” that makes us so. And elephants are also unique. And coral. And octopi. And oaks.
In other words, he gave them a challenging environment. That is exactly what Darwin proposed – that given an environment with certain resources and certain hazards, populations will tend to adapt so as to exploit the resources and avoid the hazards.
Would you agree, then, that Lenski’s experiment was a good test of Darwin’s theory?
It means they are closely related. As I said, species is not a binary concept – speciation is a process, and where the lineages have diverged in the relatively recent past, hybridisation remains possible – just increasingly unlikely.
Elizabeth,.,
The second just clarified the first. The two are not separate definitions. Also IC is about the degree of complexity and the mechanism. And AVIDA has nothing to do with biology where reproduction is IC.
It’s a question of degree. There are populations where it is debated whether to call them two “species” or one. That is because speciation is a process not something that happens abruptly and completely.
Elizabeth,
No, Lenski’s experiment isn’t a good test of Darwin’s concept. But it does show that evolution is very limited. The best example of evolution is also a good example for Shapiro’s genetic engineering by the organism.
The E coli already had the ability to digest citrate. The transport protein’s gene was just turned off in the presence of O2. The organism duplicated that gene but put the duplicate under the control of a promoter that was turned on in the presence of O2. Nothing about that resembles what Darwin was talking about.
Well, yes, they are. One is a property of a function. The other is a property of a pathway.
They are related, but they are not the same. They can’t be.
No, not by either of Behe’s definition. He did not specify a degree of complexity for the IC function. He only specified that it must not work if any part is removed.
He did talk about the degree ICness of the pathway – and defined it as the number of necessary unselected steps.
By both definitions, EQU, in AVIDA is IC – EQU has IC properties and can only be reached by a large number of unselected steps.
No, it tells us nothing about OoL. It is about evolution. It falsifies Behe’s claim that something that is IC (by either definition) cannot evolve by Darwinian means.
Reproduction itself is the pre-requisite for Darwinian evolution so reproduction itself clearly cannot have evolved by Darwinian means.
Elizabeth,
In the case of hybrids speciation does happen abruptly and completely.
You could I don’t. I presuppose the Christian God.
I agree and I would assume that the genomic sequences for those organisms are unique as well.
No in other words he gave them an environment with a specific challenge (a target).
I’m not even sure what Darwin’s theory is.
If his theory simply posited that bacteria are adaptable to different environments then I would say Lenski’s experiment was a good test
peace
Not sure what you mean by this. Unless you are using a very non-standard definition of “species”. A hybrid is not a “species” in the usual sense of the term.
So it shows it is limited because a *tiny* sample of total E coli manage to do it. How’d you figure that one out?