Wikipedia:Reference desk/Archives/Science/2018 February 9

= February 9 =

Can genes de-evolve back to an earlier stage?
The Homo genus first appeared more than 2.5 million years ago. Cooking was probably discovered 0.4 million years later at the earliest, or 1.0 million years later at the latest. The earliest humans known to have cooked were of the Homo Erectus species. Cooking seems to have played a part in the increase of human lifespan to around 70 years (the lifespan of modern hunter-gatherers) by presumably altering genes somehow. This increase probably occurred between 2.1-1.5 million years ago. But if the practice of cooking was put to an end permanently, I was wondering if this would result in humans reverting back to the 30-40 year lifespan (the same lifespan as chimpanzees) they had more than 2.5 million years ago through genes being altered. Basically what I'm asking is: Can genes de-evolve back to an earlier stage? MisterH2005 (talk) 09:33, 9 February 2018 (UTC)


 * Well if the circumstances for an organism change to be like they were in an earlier stage in its evolution it may evolve to become similar to what it was like previously. Dmcq (talk) 10:05, 9 February 2018 (UTC)


 * The first thing to say is that organisms evolve, not genes. In the simplest evolutionary scenario, organisms such as humans, evolve when genes mutate to produce a different effect on the organism that make it better able to survive in its biological niche (gene mutations appear all the time but most of them are not beneficial) and so more of the organisms with the mutated genes survive and the mutated genes become prevalent in that population. This can change as the environment changes so, for instance, during the Industrial revolution the population of Peppered moths in the UK became darker due to pollution but has become lighter again as the pollution has reduced. Also, in the early years of the industrial revolution the health of people working in factories and those living in polluted cites was severely affected and this would have affected the life expectancy of that section of the population. So, certainly a change in the environment can have dramatic changes on lifespan and if suddenly all cooking was stopped it could have an effect on life expectancy. Some of "Darwin's finches" on the Galapagos islands illustrate what one scientist refers to as evolution "running in reverse".  although I think that is a bit of a confusing description - as the comment above says, they are really evolving to fit a new situation.  Richerman  (talk) 10:49, 9 February 2018 (UTC)
 * It's an unfortunate product of pre-20th century thinking that humanity represents a "higher" type of life, and that anything closer to humans are some how "high" in relation to things further from humans. Evolution has no direction except in time, and it has no purpose beyond random changes that occasionally get lucky.  There's no more to it than that.  Organisms don't "de-evolve" to "lower" forms because those forms are not "lower", just "earlier", and one does not "de-evolve" because that's not possible; time never goes backwards.  -- Jayron 32 12:52, 9 February 2018 (UTC)
 * Still, the general trend is to evolve from simpler to more complex life-forms, so evolution back to simpler forms can be called "de-evolution"! 2601:646:8E01:7E0B:0:0:0:64DA (talk) 12:57, 9 February 2018 (UTC)
 * To a point; I suppose, but then again complexity is just another way to measure the progress of time. See Entropy (arrow of time).  -- Jayron 32 13:03, 9 February 2018 (UTC)
 * We have articles Devolution (biology) and Evolution of biological complexity which explain why although it's true there has been an increase in the maximum level of complexity, modern evolutionary biologists do not generally feel that devolution (or de-evolution) or that 'general trend is to evolve from simpler to more complex life-forms' are accurate. Nil Einne (talk) 23:11, 9 February 2018 (UTC)
 * Why assume that cooking somehow had an influence on human genetics? How long someone lives is partly genetic - but also depends on health and nutrition. Cooking improved both health and diet: more food became available, what was available could be digested more easily, the heat killed off a lot of unpleasant parasites. Wymspen (talk) 15:15, 9 February 2018 (UTC)
 * The environment and activity of cooking can also introduce early mortality where ventilation is limited, or where the gathering of fuel is precarious. Hayttom (talk) 16:33, 10 February 2018 (UTC)
 * If everyone dies of parasites in their 30s, then there will be no selection for genes that increase lifespan beyond that. If cooking allows people to live longer, then there can then be selection for longevity (assuming longevity helps to pass on genes). Iapetus (talk) 09:58, 12 February 2018 (UTC)
 * "Devolution" sort of is and sort of isn't possible. There are many features - like height, lifespan, or intelligence - which are to some extent affected by multifactorial inheritance.  This means that there are many individual mutations that may have a small effect one way or another on these traits.  Evolution in one direction (which may come with pleiotropic costs and/or benefits in other aspects of the organism's biology) could be followed by evolution in the other direction.  This does not require finding the exact mutation that caused the first shift and undoing it ... just any mutation with a similar effect.  So if you increase growth hormone level by a mutation in an enhancer in one generation, maybe another mutates the protein to be a little less effective as a ligand.  As a result, depending on the organism's priorities, it might shift one way or another over time.  However, there is not much to demand that a change in one characteristic includes a change in others.  A short Homo sapiens does not become a Homo habilus simply due to the change in size!  So if you find an organism "devolving" to match an ancestor in multiple ways at once (which you usually won't), it would probably mean that it moved back to the ancestral ecological niche and underwent convergent evolution rather than "devolution". Wnt (talk) 00:49, 11 February 2018 (UTC)
 * A “meat-adaptive gene” has been theorized to be the reason that humans have evolved longer lifespans than chimpanzees despite sharing a 99% genetic similarity with them: https://phys.org/news/2010-01-humans-outlive-apes.amp MisterH2005 (talk) 02:49, 11 February 2018 (UTC)
 * Recent research suggests that activity of some genes increases after death, and our bodies may start reverting to their embryo state . 81.147.142.158 (talk) 13:34, 11 February 2018 (UTC)
 * I like phys.org for news, but this sounds like garble, and the article it comes from sounds like garble. The article these all came from, unlinked, appears to be, which is complicated, but basically makes the point that one common allele of apolipoprotein E, ApoE4, really sucks, but they say it is ancestral.  The chimps have an allele like ApoE4 but with two mutations that make it more like ApoE3.  To read about ApoE4, you'd think that its carriers should not be allowed to play sports in high school let alone the NFL, are near doomed to Alzheimer's and heart disease anyway, and might as well be rounded up in camps and sterilized with flamethrowers.  But I think perhaps not all the data is in about ApoE4, since there is precious little to explain just why it is for all their problems its carriers managed to hold their own for hundreds of thousands of years of brutal natural selection.  In any case, I feel unconvinced by such data, especially since that paper starts out by saying that groups of chimps kept under lab conditions in different labs get different cholesterol levels and they don't know why! Wnt (talk) 14:49, 11 February 2018 (UTC)

While the answers above are certainly well thought out, the answer is that nothing would de-evolve, just evolve with new traits. Traits that may be similar to, and resemble those, of an earlier stage of human. Nothing de-volves, it just evolves to something new that survives natural selection for one reason or another, even if only temporarily. Our perception of it as a step back is incorrect in it's assumption. It's just a change that natural factors have favored to continue until such time that they don't. AT that point they either die out, or evolve again.

Question regarding the reversibility of mutations.
This question is obviously inspired by the previous one but is I hope simpler. Besides the OP of the previous question introduced in his preamble several complications I'm not interested in. My question is this: suppose a cell undergoes two successive mutations m0 and m1: is it at all possible that m1 is simply m0 backwards, in other words that m1 simply cancels m0, or is there something about the process of mutation that precludes that? Same question for a chain of mutations m0, ..., mN: could the state after mN be what it was before m0? Thanks. Basemetal 13:23, 9 February 2018 (UTC)
 * Probably not; the chance of a mutation m0 is just 1/N, where N = number of nucleotides in the DNA strand. So the chance of the exact same mutation happening is not better than 1/N * 1/N * 1/3 (the 1/3 factor is because there's only a 1/3 chance it mutates back to the same nucleotide, it could mutate to one of the other two instead).  Since 1/N is a fantastically small number already, then 1/N * 1/N is insanely small.  -- Jayron 32 14:00, 9 February 2018 (UTC)
 * Sure, but the answer to OP's question "is it at all possible...?" is an emphatic yes. As you have illustrated, it is not very likely. NB: mutation rates are not equal at all loci as assumed in your calculation, as described at Mutation_rate). Also there are many other types of mutation in addition to the single-nucleotide switch that you present, and that further changes the odds. But none of that is relevant to the possibility, it is only relevant for the estimating the actual likelihood. A detailed estimate would be very difficult, but "very very small" is probably sufficient for OP's interest. SemanticMantis (talk) 15:56, 9 February 2018 (UTC)
 * No, it isn't. Events that have a probablility of happening less than once over the entire history of the universe do not meet any reasonable definition of "possible" except for obnoxious pedantry.  -- Jayron 32 16:00, 9 February 2018 (UTC)
 * I'm confused. Back-mutations do occur, don't they? - Nunh-huh 20:32, 9 February 2018 (UTC)
 * Yes, and thank you for reminding me of the specific term. Back mutation redirects to a section of the article I already linked, saying "A back mutation or reversion is a point mutation that restores the original sequence and hence the original phenotype." We even have a nice citation for it. That research (dealing with human genetics) concludes in the abstract " Thus, back mutation now becomes, together with intragenic recombination, an important genetic mechanism to consider when explaining examples of a reversion of somatic cells to "normal" in persons with a genetically determined abnormal phenotype."
 * So the answer to OP's question is "yes, this is possible" according to our article, the authors of the cited article, the authors of many biology books, this professional biologist, and in fact many people who have studied genetics. SemanticMantis (talk) 20:53, 9 February 2018 (UTC)


 * OP and I will continue to use the normal definition of "possible", you're free to do whatever you like :)SemanticMantis (talk) 20:26, 9 February 2018 (UTC)
 * Isn't it also possible that a 100 meter wide asteroid will be discovered 12 Feb 2018 UTC and quantum tunnel through many miles of Earth and air on 31 October 13131313 AD GMT (Gregorian) like it wasn't even there? Sagittarian Milky Way (talk) 11:20, 12 February 2018 (UTC)


 * If it happened, it would happen in a different way. For non-sex-chromosomes, each organism carries two alleles of each gene. Even after a mutation, there will continue to be copies of the original version in the population for a long time. Unless the original version has completely disappeared, it is possible for its prevalence in the population to increase, even to the point of eliminating the mutated version. Looie496 (talk) 17:53, 9 February 2018 (UTC)
 * Jayron32's original answer is a little wrong in principle. To calculate the probability that m1 cancels out m0, one does not need to calculate the probability that m0 occurs; that is a given. Rather one needs simply to work out that m1 occurs in the same place as m0 (wherever that was) and in the reverse direction. So that is just (1/N)*(1/3). Since N is large, that is still a small number. But since a lot of mutations happen each generation, it is not so improbable that one of them is reversed at the next "step". Especially because some sites mutate more often than others.
 * However, that is all irrelevant really, because the question is inappropriately framed. One is not really interested in the probability that the very next mutation in the genome is at the same site and in the reverse direction. Rather, one is interested whether the mutation is reversed within a few generations before much further evolution has happened to the same gene. That is much more probable (again, especially because some sites mutate more than others and because some sorts of nucleotide transitions are more likely than others). But I also agree with Looie496 that what is usually more important than back mutations are copies of the original variant remaining in the population. And another point is that a different mutation elsewhere in the same gene or in another may almost exactly reverse the phenotypic effect of the first mutation. Jmchutchinson (talk) 22:12, 9 February 2018 (UTC)

"Cost" of a joule
Hello. A joule is defined as the energy imparted by a force of 1 newton over a distance of 1 metre.

Intuitively, this seems to indicate that the joules are "cheaper" at high speeds than at lower ones. E.g. if an object is moving at 10 m/s, you need to exert 1 newton for a tenth of a second to get a joule, but if that same object is moving at 100 m/s, you only need to exert 1 newton for a hundredth of a second to achieve the same amount of imparted energy.

I guess my intuition is wrong, but what am I missing here?--217.121.5.132 (talk) 16:09, 9 February 2018 (UTC)
 * Work done = force × distance
 * Just doesn't matter how fast you do it. Andy Dingley (talk) 16:36, 9 February 2018 (UTC)


 * It's better to do the dimensional analysis at a more fundemental level here. A joule is a kg m2 s-2, that is a kilogram times meters squared per second squared.  How you "break up" those units to understand work or energy in a particular situation depends on how the situation is constructed.  For example, sometimes it helps to think of energy as work (moving an object over a distance, that is a newton times a meter).  But energy is also pressure times volume; that is if I have a cubic meter of a container of gas that I add 1 pascal of pressure, that also has added 1 joule of energy (I'll leave the reader to do the dimensional analysis on that one).  The reason I bring this up, is to show that your analysis is confusing two measurements: work and power.  Work (energy) is measured in joules, that is kg m2 s-2.  But in your discussion, you're adding an additional term, a "per second" term, which I take it to mean joules per second, which is the watt, or kg m2 s-3.  In your two examples, the energy exerted would be the same, but what would change is the time at which that energy was being exerted, which is the power.  At double the power, you're using energy at twice the rate.  In this case, the so-called "cost of the joule" is simply the inverse of the power.   -- Jayron 32 16:38, 9 February 2018 (UTC)
 * Thanks for your explanation. Then, is my calculation correct that you will spend 0.2 joules if you exert a force of 1 newton on a 1 kg object for exactly 1 second (assuming 100% efficiency)?--217.121.5.132 (talk) 16:47, 9 February 2018 (UTC)
 * No. The mass of the object is implicit in the newton; you exert a newton of force by accelerating a 1 kg object by 1 m s-2.  If you apply that force for a distance of 1 meter, THAT'S a joule of energy.  How long it takes to expend that joule of energy does not enter into this calculation; that would be the power.  If you have an object that is already in motion, it has some kinetic energy equal to its mass times its velocity squared, and it would gain (or lose, depending on the direction of the applied force) additional kinetic energy equal to the force applied times the distance over which that force was applied.  -- Jayron 32 16:54, 9 February 2018 (UTC)
 * Jayron32 and others have already correctly explained the issue; but if I can address the OP's misunderstanding: the OP has correctly identified that adding an extra parcel of velocity, when the velocity is already high, would require less time. This much is true, and is a direct result of basic kinematics.  But the formal misunderstanding is even simpler: just because something happens faster doesn't mean it is any cheaper!  By analogy, suppose you spent $100 to drive from Los Angeles to San Francisco over the span of a week; the next week you flew on an airplane, and spent $100 during the one-hour flight.  The money got spent faster, but it was the same amount spent!
 * By analogy, when a jet engine is throwing out thrust - which is a force - the engine must deliver a different amount of power to expend that thrust at high velocity. This manifests in the fuel-burn-rate per unit time.  If we imagined a super-simplified and thermodynamically-perfect jet engine, the amount of chemical fuel burned per second corresponds to the energy available each second.  At high velocity, we require more fuel per second - even though the jet engine produces the same exact thrust!  This simplification is very profound, and it fundamentally explains why a jet airplane, flying high above us at “nominally” full engine power, eventually stops going faster.  The engine can keep putting more power in, but the force does not increase, and when it exactly equals the force of aerodynamic drag, the airplane acceleration goes to zero.
 * Energy is a lot like money: it is a scalar quantity; its total quantity is often irrelevant, but its relative quantity is quite important; and we can extract many important pieces of wisdom when we study the statistical distribution of this scalar, normalized by units of time or units of space. Perhaps most strikingly, we cannot create it, and we cannot destroy it; we can waste it; we can store it in many forms.
 * If you are a cyclist, or a skiier, or an airplane pilot, or anybody else who regularly has to manage your kinetic energy, you will recognize that it is more difficult to get an extra 1 mph of velocity increase when you are already at your top speed. This is because adding velocity requires adding energy; and your top speed is almost universally defined as the speed you go when you have no more energy left to add!
 * I hope these intuitions help with the concept. Ultimately, the best way you can understand the models is to study the simple equations of kinematics and Newton's three laws, which in themselves represent some of the most important concise statements that relate position and velocity to the more profound, less obvious, entities we call “momentum” and “energy.”
 * Nimur (talk) 18:18, 9 February 2018 (UTC)
 * Regarding the assumed energy-cost-paradoxon also see Oberth effect. --Kharon (talk) 01:56, 10 February 2018 (UTC)