There's an oversimplification going on here, I think (in the article I mean--the experiment is simplified for a good reason). The "one place" that all of these strains reach is that they are more fit for the environment than the starting point of each. This is not a new idea. This was the thesis of "On the Origin of Species".
As for the level of fitness of each strain... after sufficient generations, basic statistics would tell tell us to expect this result. The combination of random mutations influenced by one constant factor (natural selection in a fixed environment) over a large number of generations implies this result.
But the article appears to assert that this means that evolution's course is predictable. But what's not addressed is how each strain is succeeding. They know that the variations in the genes from strain to strain vary quite a bit. Yes, the result measured from one statistic (growth rate) is the same, but future evolution and other traits of the yeast will be influenced by the particular paths taken.
Finally, once you put this mechanism in an environment filled with millions of other competing gene pools and a varying environment not bound by natural selection which means there's an actual threat of extinction, then that's where the true variability of the process emerges.
The "one place" that they reach is the same level of fitness. This is surprising because they expected some strains to take different evolutionary paths and evolve different traits, and therefore have different levels of fitness in the end. Some would be better than others just by chance. The probability of the same mutations happening in all the samples is very small.
Instead they all seem to converge on the same fitness. Even though they have different mutations.
In Per Bak's "How Nature Works" [1] he describes an simulation he did in which he tried to replicate the series of evolutionary discontinuities and mass extinctions we see in Earth's history. The reason this article brought that to mind was that Bak was not able to get the complex behavior by selecting only the fittest to survive. Instead, he ended up having to kill off the least fit of the species and leave the rest to proliferate and develop interdependencies.
His result doing that was some very interesting behavior in which extinctions of all sizes happened in the population of species in a way evocative of natural history on Earth. His idea was that in order for an evolutionary system to reach the self-organized critical state in which big changes happen, the species had to develop interdependencies so that the success or failure of one could affect that of the others.
It would probably be a much more difficult experimental procedure because he would have to keep the yeast strains in a community and somehow eliminate rather than select a variety, but it would be really interesting to see if the result of a common convergence point would survive with the model of "death of the least-fit" rather than survival of the fittest.
> ...not able to get the complex behavior by selecting only the fittest to survive. Instead, he ended up having to kill off the least fit of the species and leave the rest to proliferate and develop interdependencies.
How is "selecting only the fittest to survive" aka killing the least fit different from "kill off the least fit of the species" aka selecting only the fittest to survive they sound like exactly the same thing thing.
I probably could have stated it more clearly like this: "selecting only the fittest ONE to survive," and "killing off the least fit ONE."
The difference was in what the composition of the surviving population was - in the extreme case where only the fittest one survives, there's only one species that makes it to the next round. In the case where the least fit one is killed off, there are n-1 species still in the mix with their relationships intact. You can of course extend it to keeping more than one of the fittest, or killing more than one of the least fit.
I was just re-reading that section of the book to try to give you a better answer and I think I must have mis-remembered his original experiment a bit (sorry about that). The original evolution model that failed to exhibit the behavior was more complex than just a survival of the fittest (he was playing with mutations and connections between species in a kind of complicated way). The part about killing off the least-fit species really was the key to getting complex behavior though, so at least that part of my original comment was right.
The eventual solution was to simplify the model so that the least-fit single species would go extinct and be replaced by a new randomly generated one each round. The fitness of a species was related to its connections, so if one species was to go extinct it could trigger others to cross the fitness threshold and go extinct as well. This led to cascades of extinctions with a power law distribution (like the pattern observed in real extinctions on Earth). The mass extinctions would eventually stabilize at a point where all species were above the fitness threshold, but the "gene pool" had been seeded with lots of new randomly generated species at that time, so there were periods of rapid mutation punctuating long stable periods where not much changed.
His book is pretty interesting, although it's sometimes a little over the top with self-congratulation. I'm reading another one now called "Self-Organized Criticality"[1] which is a much more rigorous treatment of the idea that I think will be a bit less sensational and more applicable to real problems. Bak's book is excellent if you want to get excited about making simple models do complex things though.
Very well written article that allowed a non-biologist but engineer like me to follow. Systems Biology is quickly becoming one of the most interesting fields, IMO.
The observations about organisms catching up and slowing down is consistent with (the regression to the mean)[http://en.wikipedia.org/wiki/Regression_toward_the_mean]. Although it's something I only recently read about in Daniel Kahnemnan's Thinking Fast and Slow, it's applicability is widespread; so much so that I was not surprised seeing it in action here.
One thing that I do not fully agree with (or maybe that I do not fully understand) is the final evolutionary endpoint was measured simply by it's ability to grow in specific lab conditions. While this is a necessary condition for equality of endpoints, it is in no way a sufficient one. It is very possible, and even highly likely that while that goal of fitness is optimized for, evolution changes the final organism's way of reacting to other phenomena. I'm not sure if there is a better way to quantify equality (or some notion of edit distance) though.
I'm a fan of the view that evolution should be viewed as an information transfer process. It transfers information about the fitness landscape into the genome.
Similar environments will give rise then to similar information. There will be variance, of course, but it's going to be a lot narrower than the random walk interpretation of evolution would predict.
There is evidence for this from several places:
* Convergent evolution on Earth at many scales
* Digital evolution experiments showing similar endpoints in similar environments and against similar fitness landscapes
* ... and now this series of experiments showing convergent evolution in the lab even down to the genetic level
There's a lot of interesting implications here. For example: can we really say that another planet very similar to Earth that is of similar age might not harbor another bipedal warm blooded sentient species physically similar to ourselves? I'd be really surprised if we found Star Trek levels of similarity (e.g. biological and sexual compatibility), but some level of similarity in areas like basic anatomical structure would not shock me. Perhaps the anthropomorphic alien depictions in sci-fi (Giger's alien, the "greys," etc.) are not terribly implausible.
Of course a radically different world harboring life might yield something radically different for similar reasons.
All of life on earth operates using mostly the same biochemistry. While there are countless differences and exceptions (eg. different codons, different genome architecture, metabolic and gene product differences), they are vastly outweighed by the vast magnitude of commonality. This is mostly observable at the small scale (nucleic acid, protein, lipid chemistries) but also somewhat true at a larger scale (certain transport processes, etc).
Bacteria have an incredible amount of biochemistry in common with us. And so do all other living things. Bacteria will of course be radically different when we contrast ourselves to other eukaryotes, but we can still see our biochemistry in them.
We all evolved from the same lineage, so all of the "core architectural" decisions were made long ago--before prokaryotes even came into being. The earliest choices made in exploring the chemical-physical fitness landscape turned out to have the greatest impact; all of our biology is expressed in terms of the primitives that were first established. For example, the universal adoption of the amino acid L-isomers. It's a seemingly arbitrary choice of little importance, but once that choice was made, earth biology forever had to live with it.
If there were different paths taken early on, metabolism would be different, polymerization would be different--cellular replication itself might also be different (one can only imagine). I imagine there would be cells. At higher levels, the choice of "building blocks" has an impact on bioinorganic chemistries (think heme groups). How you chelate depends on structure.
I assume that the heme group chemistry only evolved because of a highly oxygenated environment that life could exploit for energy production. Had our aqueous environment (and later atmosphere) been of a different composition, oxygen may not have played the same role.
Further tangent: certain elements are extremely important. Organic (carbon) chemistry is incredible due to the valence. Evolving on a world without oxygen might really suck, because oxygen chemistry is just so perfect for redox. And what of all the other important chemistries? Do they have a bioavailable source?
Anyhow, it's suffice to say that any change in early evolutionary choices would have far-reaching impact on the essential reactions (energetics, thermo, kinetics, dynamics...) and might yield results that are altogether alien to us from an Earth biology standpoint. Once you have your building blocks, you can't exactly backtrack.
I think a lot of it boils down to what the evolutionary pressures (defn: [1]) on another planet might look like. If we discover life on another planet, it wouldn't surprise me if some of the organisms we find resemble organisms in some branches of the "tree of life" on Earth, but I think it would be unlikely that there would be so many similar evolutionary pressures that we would see something particularly similar to ourselves.
For example, life has been around for at least 3.5 billion years on this planet[2]. At this point in time, there is no evidence that any other species on this planet has ever had a technological civilization that would be of particular interest to us in the way that is often envisioned for hypothetical alien visitors. This isn't for lack of different species[3] or for insufficient genomic complexity[4] in other species. In other words, there's nothing about the human genome that represents a "pinnacle" of evolution, as much as we tend to hold ourselves in pretty high regard among other species.
As another example, let's walk the tree of life (http://tolweb.org/Life_on_Earth/1) to get a ballpark idea of how many different features we run into before encountering homo sapiens:
5. Chordates (animals possessing a notochord, a hollow dorsal nerve cord, pharyngeal slits, an endostyle, and a post-anal tail for at least some period of their life cycles) http://tolweb.org/Chordata/2499
9. Sarcopterygii (lobe-finned fish & terrestrial vertebrates. I am not an evolutionary biologist but it seems like this roughly corresponds to the evolution of limb-like structures) http://tolweb.org/Sarcopterygii/14922
There are several more branches before getting to homo sapiens.
I think the big remaining questions are:
1. What features of an organism are necessary for that species to develop a technological spacefaring civilization?
2. Are evolutionary pressures that support the development of those features likely to appear eventually on life-bearing planets similar to Earth? (I think this assumes that abiogenesis on other planets would be likely to produce simple cellular structures with similar capacities for evolution.)
3. Are we in a simulation created by another species in a "real" universe to answer similar questions? :)
I'd guess...
1. Limbs (and the supporting underlying physical structures) with significant capacity for manipulating objects, terrestrial-dwelling (as opposed to living in the sea), a advanced neocortex, and way too much free time, along with very difficult-to-impress mating prospects.
2. Depends on how similar to earth (a waterworld would be unlikely to have any species developing advanced technology, I think... maybe I'm not being imaginative enough). Also depends on how many of the evolutionary pressures that got us to this point were more or less inevitable as opposed to random. It seems that even if it's relatively likely that a species similar to humans will evolve on planets similar to Earth, there are still numerous additional hurdles in the development of different kinds of societies that could make or break the whole spacefaring thing. Even now, humans are not a spacefaring species, and the odds that we nuke ourselves out of existence or create a genetically engineered superbug that wipes us all out are relatively high considering we have absolutely no backup plan as a species for either of those events (or any number of other catastrophic global events). So maybe my answer to #1 is insufficient because we're not really a spacefaring species yet.
3. If they have that kind of computing power, I would've expected something more interesting. Maybe they're trying to predict the odds of encountering other species like themselves during a galactic collision? :) https://en.wikipedia.org/wiki/Andromeda%E2%80%93Milky_Way_co...
The article's "one place" claim will confuse those who don't understand evolution. It's like saying that all algebraic equations are the same because they all have equals signs. It's true, but it misses the point that the equations surrounding the equals signs are all different.
The "one place" being described is "fitness", but to any number of environments in constant change. It's like saying that all women are alike -- they aren't men. It's a meaningless claim masquerading as meaningful.
This is an interesting, readable report on an automated procedure for experimenting on evolution of microorganisms that will be good to try out with microorganisms other than yeast. Figuring out the different gene interactions over many generations of random mutations will eventually develop a deeper understanding of how these one-celled organisms work. And that, as an expert interviewed for the article suggested, will help deepen understanding of evolution as a process. "'I think many people think about one gene for one trait, a deterministic way of evolution solving problems,' said David Reznick, a biologist at the University of California, Riverside. 'This says that’s not true; you can evolve to be better suited to the environment in many ways.'"
Scaling up this approach to study multicellular organisms with longer generation times than yeast will of course take a lot of time and effort. Sometimes (as here) experiments on simple model organisms produce surprising findings. There will be even more surprising findings, I think, as evolution is studied in more detail in multicellular organisms at the molecular genetic level with strict experimental controls of developmental environments.
EDIT TO REPLY TO ANOTHER COMMENT:
Another participant asked,
I feel like a CS education should prepare one for this kind of field. What's a good way to get started learning more about evolution?
It will indeed take a lot of hard work by experienced computer scientists to solve some problems in evolutionary theory. A quite good way to start learning about evolution is to read the book Why Evolution Is True, which talks not only about evidence for evolution but also about implications of the theory. Then read the book's website[1] frequently for the latest news.
Other good things to read are the books by biologist Sean Carroll (not to be confused with physicist Sean Carroll) such as Endless Forms Most Beautiful.[2]
The public perception of evolution misses out on the depth of scientific work being done these days. Two more books that make modern evolutionary research accessible to the public:
1. Nick Lane's Power, Sex, Suicide: Mitochondria and the Meaning of Life - Covers the evolution of the eukaryotic cell, why it's such insanely chance event, why multicellular life depends on it, almost as an aside making a great argument that complex multicellular life is extremely rare in the universe. Also has a couple of fascinating chapters on how the interaction of the mitochondrial genome with the nuclear genome influenced development of sexes and sex.
2. Nick Lane's Life Ascending: The Ten Great Inventions of Evolution - Covers very different material and again all fucking fascinating. It's amazing to see how things like sight have evolved over and over and over because they're so fucking useful, and the genetic analysis that allows us to pick out each separate emergence of a feature.
This is pretty cool! One thing to think about (and its raised in the last two paragraphs of the article) is how this might change in a dynamic environment, and with the addition of other simultaneously evolving agents.
I think its pretty clear that given some static environment, there will be some maxima of fitness - everything is limited by physics and thermodynamics at some point. Aside from the question of 'how close can we get to this maxima', another question is, if we have a bunch of different sets of mutations that together generate an organism that is at or near the current environmental maxima, what happens as the environment changes?
Biological evolution isn't the only kind of evolution that can be dreamed up. "Self-arised change over time" can be applied to all sorts of domains, it would be interesting to have a mathematical definition of evolution that can handle all the things. This appears to be a step in that direction.
It would be interesting to know if inventing an internet is a natural outcome of evolving life on a somewhat-stable planet. Regardless of the aesthetic differences among life forms developing from different initial conditions, will a planetary ecosystem arrive at similar milestones along the way? Mobility, eyesight, language, society, etc.
I think this research is very exciting because it actually is narrowing in on the limitations of evolution, or the "search-space" of possible life forms. This points to the idea that life on other planets may be similar to us, if not only in milestones, but possibly aesthetics as well. Maybe we can even get to the point where if we know the current and past environmental conditions of a planet, we can make educated guesses as to what life would look like on its surface.
Specifically human intelligence of the kind that builds internets and spaceships has evolved exactly once on all of Earth's history, in contrast to things like eyes and wings, which have evolved dozens of times. And it evolved very recently, and appears to be due to run-away sexual selection of the kind that produced the peacock's tail.
All of this suggests that specifically human intelligence--not dolphin or bird or any other kind of intelligence that does not build internets and spaceships--has an unusually narrow bottleneck to get through compared to most other capabilities, despite the ridiculous benefits available once a species gets through it.
As such, I'm betting that we'll find life pretty much everywhere in the universe, and intelligence almost no-where. When intelligence does arise, it will likely be wildly different from us due to the essentially random nature of the sexual selection process that takes it through the bottleneck. The precursor species will almost certainly be tool-using and social, but tool use and social behaviour are both extremely common, so we could find anything from intelligent birds to octopuses.
With regard to the "surprise" that evolution can produce convergent morphology, this isn't really that surprising: genetic studies have shown that such things occur in nature. For example there are two species of coastal lizard in the Yucatan both of which have evolved an extra vertebra in their neck which were thought to come from a common ancestral population, but thanks to genetic analysis in the '90's were found to be due to convergent adaptations to coastal conditions by different inland species. Furthermore, while gene selection does occur, the basic unit of selection is the whole organism, which in most cases either reproduces, or does not.
> despite the ridiculous benefits available once a species gets through it
From an evolutionary perspective, the benefits are far from proven. On an evolutionary timescale, civilization hasn't been around that long and has had a disproportionate number of existential risks over its timescale. In the nearer term, there is also evidence for large scale dysgenetic fertility in many countries with respect to genotypic IQ. http://en.wikipedia.org/wiki/Fertility_and_intelligence
While it's true humans are the only species inventing the internet, science is also finding out other species have greater problem-solving abilities than previously understood.
The course of evolution is predictable only to the extent that the environment and its attributes are known, and that the adaptation to that environment is possible for the organism.
We might postulate evolutionary trajectory is toward adapting well enough to the environment, or only as good as it has to be for the purpose of survival and reproduction.
There's also a factor of limits of adaptation, or maxima for given evolutionary paths. A bacterium will not evolve to be an astronaut in a billion bacterial lifetimes. If it's good enough to live in my gut, or another gut, it's evolved sufficiently.
We know that environments have not been static, creating evolutionary pressure, and novel mutations then have a chance of producing increased adaptation in the changed environment. We can only assume humans evolved the brain necessary to adapt to the existing environment.
The question arising out of these ideas is our brain evolved sufficiently that no further evolution is necessary. If we are able to adapt to changing environments using the brain we have, it wouldn't need to evolve.
In a way global warming and all that it implies are a kind of natural experiment of the hypothesis. We must figure out a way to repair the damage to the planet, or adapt to it, otherwise we would predict disaster for current lifeforms including humans, and evolutionary processes will shape the successor species of the future.
>Specifically human intelligence of the kind that builds internets and spaceships has evolved exactly once on all of Earth's history, in contrast to things like eyes and wings, which have evolved dozens of times.
>All of this suggests that specifically human intelligence--not dolphin or bird or any other kind of intelligence that does not build internets and spaceships--has an unusually narrow bottleneck to get through compared to most other capabilities, despite the ridiculous benefits available once a species gets through it.
do you place Neanderthals in the same bucket of internet building intelligence or not? If not, then is it because you believe that Neanderthals weren't capable in principle to get to Internet age? If yes, then where do you make a cut-off as it seems that we have continuous range of intelligence down through ancient human species to primates to racoons...
Life on other planets will be extremely different from us. Consider the following thought experiment. Imagine a planet like earth, where evolution follows exactly the same path, with the same species, the only difference being that it is 2% older. The earth is 4.5 Billion years old, so 2% is 90 million years.
According to Wikipedia [1] 90 million years ago is about the time placental mammals appeared. So in that time frame evolution pulled mammals in directions as diverse as bats, whales, and elephants. So the human decedents on the other planet will not look at all like us...
Stephen Jay Gould published a lot of books about evolution. "Evolution and Extinction" is a nice light read to get started (just to pick a random book of his). "The Structure of Evolutionary Theory" is a more formal book where he pulls together a lot of the ideas from the other books and from his research.
He develops a lot of ideas in "The Structure of Evolutionary Theory"... basically:
* Organisms are constrained by their evolutionary history... it's "easier" for them to develop in some directions than others. For example, the vertebrate body plans is extremely ancient and there are a lot of genes that do exactly the same job in mice and men, or even in more distantly related ancestors. Embryology and the field of Evolutionary development ("evodevo") has made a lot of this more concrete recently. Similarly, there is evidence that organisms draw on their evolutionary history as a "library" of possible responses to new stimuli.
* Punctuated equilibrium: the fossil record shows millions of years of little to no change, followed by sudden mass extinctions that set the stage for another equilibrium period.
* Sexual selection: if females are looking for some feature, males evolve to have that feature, even if it has no actual use or is harmful to the species. And vice versa. The peacock's tail is the classic example.
* More controversially, Gould argues in favor of hierarchical selection, the idea that natural selection can act on groups or species as well as individuals. This is something that Dawkins vigorously denies, preferring to view the individual gene as the sole unit of evolution.
Great book, yes, but don't stop there. There is a lot more to evolution than Dawkins suggests; selection, no matter the mechanism, is, after all, determined by phenotype (the "realised organism", if you will) rather than genotype. And there are a lot of models of selection (and of modification) that make accurate predictions but don't have the Dawkins seal of approval. You'd be missing a lot if you allow yourself to become stuck in the notions of 1976, no matter what their merits at the time.
As for the level of fitness of each strain... after sufficient generations, basic statistics would tell tell us to expect this result. The combination of random mutations influenced by one constant factor (natural selection in a fixed environment) over a large number of generations implies this result.
But the article appears to assert that this means that evolution's course is predictable. But what's not addressed is how each strain is succeeding. They know that the variations in the genes from strain to strain vary quite a bit. Yes, the result measured from one statistic (growth rate) is the same, but future evolution and other traits of the yeast will be influenced by the particular paths taken.
Finally, once you put this mechanism in an environment filled with millions of other competing gene pools and a varying environment not bound by natural selection which means there's an actual threat of extinction, then that's where the true variability of the process emerges.