Hacker News new | comments | show | ask | jobs | submit login
Massive genetic study shows how humans are evolving (nature.com)
323 points by gwern 40 days ago | hide | past | web | 141 comments | favorite

Now that we have access to the entire genome directly, instead of relying on twin and adoption studies, we are going to get many more of these insights. And it's not all going to be comfortable. There will be cries of biological essentialism, while the blank slate idea is no longer one that can be held by an educated rational person.

On more controversial lines, see this http://www.biorxiv.org/content/early/2017/09/06/184853.1 study (N=280,000) that correlates specific genes with IQ.

There are two big risks, as I see it: 1) the human tendency to turn statistical averages into absolutes. Imagine genetic testing saying that gene A is associated with a slightly higher than average propensity for coding ability (say, 3%). Then we see people saying "Oh you can't code? Must not have gene A!" Or genetic testing revealing that someone doesn't have gene A, so they're told they can't code, even though they could very well have been great at it. All because of an extra 3% probability.

2) Making it impossible/unbearably expensive for people with genes that are correlated with higher risk of whatever conditions to get health care/insurance.

We need to be really careful with how we talk about genetic insights as we gain more and more understanding, as well as what uses of that information we allow, or we're going to get ourselves into another period of atrocities justified by what boils down to eugenics. Perhaps not on the scale of genocide of entire populations, but still bad enough.

I'd add 3) A Genetic arms race between nation states.

It's only a matter of time before a nation state starts spending a not-so-insignificant amount of money on IVF + DNA-based screening to start picking good genetic offspring for their populace. It's only going to start becoming more and more plausible the cheaper we make this technology.

Very awkward and important questions/decisions await us in the near future.

Genetic arms race, or intentionally crippling your population to make them easier to control? I wonder which the various governments will go for?

Are gene-modded babies going to have genetic compulsions that make them predisposed to propaganda and suggestion? Biological kill-switches?

There's a lot of good and a lot of bad that is going to come out of these technologies.

> Genetic arms race, or intentionally crippling your population to make them easier to control? I wonder which the various governments will go for?

This hypothetical doesn't make sense in the face of robotics. If people are around, they are because they are useful.

The GP's point is, to put it in your terms, whether the people are more useful for their friends, families and humankind or if they are more useful for governments to retain (and grow) their power. The activists today who are fighting their governments and government agencies are of the former kind, and there is a natural conflict between them and the governments in many areas (not saying all areas).

Unless one sees all governments as benevolent agencies that are truly "of the people, by the people, and for the people", this is a distinction to be wary of.

Nation states could already do this, but they've been remarkably uninterested. Yao Ming is the only example I know of.

You don't need DNA tests, or even knowledge of the concept of DNA, to participate in a genetic arms race. DNA tests were not necessary to produce dogs that would rescue drowning humans or cows that would produce incredible amounts of milk. They also aren't necessary to produce "good genetic" humans.

And furthermore, even after producing / selecting those, we struggle to give them a good education in a repeatable way.

China already paid people with high intelligence to ship them DNA for research in 2013, they are already doing embryo gene manipulation. They are years ahead of us due to not caring about ethics. Political correctness will be the death of the west.


The BGI project referenced above was hindered by certain suboptimal methodological assumptions. It does not appear that the Chinese are any further than the West in the field, and the most notable replicable advances have occurred in the supposedly-stymied labs of the West.

It is worth remembering that China is a highly ideological society with blinders and "PC" of its own, and is not entirely uninfluenced by Western priorities.

Different ethics != not caring about ethics. And if not caring about ethics were the key to winning the nation-state game, the ethical response would be to tear it all down ASAP, which is usually quite unethical in practice.

I'm not sure I follow your argument. Tearing it all down is or isn't ethical, therefore not caring about ethics is not the key to winning the nation-state game?

Ethics is stuck between a rock and a hard place if nations collapse because they're not disregarding ethics enough, as the previous comment suggests. (I'm disputing the premise.)

I phrased it a bit harshly, more like less inhibited. The US can't even have a discussion about IQ and genetics due to the pushing of BS like blank slate and political correctness, which makes doing research impossible.

Well, there are potential reasons beyond PC for caution.

Keep in mind that eugenics, the "old school" version of genetic manipulation, is widely considered a failure, and has a nasty association with certain unsavory leaders that loom large in European/American history (the most notorious one being, um, Hitler). One can say that eugenics has that nasty issue of raw tribalism clouding any potential good judgement. Personally, I would consider the sordid history of eugenics to be a warning sign to everyone in regards to genetic engineering.

The other angle to bring up regarding genetic manipulation of humans, is that genetic manipulation of animals and crops, whether through the old school "selective breeding" or more modern techniques, have not been without unexpected big problems. At the very least, specific health problems have popped up in certain purebreds, and certain bred crops are far more vulnerable to being wiped out by single parasites (eg: our "perfect bananas" have this nasty tendency to keep getting wiped out by the Panama disease). Biodiversity is also the reason that in-breeding of humans is widely looked down upon.

The other ties with the former: it is not impossible for some unsavory leader to take some research, bastardize the science through blinders of tribal instinct, say "Ah! Gene X is the key to IQ!", and do some nasty tribal things "justified" by science. And in the end, get it quite wrong in the end due to lack of biodiversity.

If by political correctness you mean the basic human rights of the constitution Of course we could add an exception to the constitution that excludes lab grown children. Well, unfortunately gene manipulation is eugenics unless you are willing to support your "defects" until they die of old age plus their offspring but then china is probably going to win again because they are disposing their defective subjects as soon as they notice the defect.

It doesn't take much imagination to think of this scenario. Creating artifical super children that grow up inside a lab and get killed off when convenient is the most common staple in scifi books and movies involving gene editing.

What's more important? Basic human rights for all humans or competitiveness?

> If by political correctness you mean the basic human rights of the constitution

I think the US constitution is silent on the subject of genetic research.

> What's more important? Basic human rights for all humans or competitiveness?

If a nation is uncompetitive, it won't matter in the long term what it thinks about human rights, as that nation will will be out-competed by other natinos that are more competitive.

I agree. Sometimes cowardice masquerades as ethics. Things will go wrong with genetic experiments. We should do them anyway.

I doubt it will end well.

Point 2 is not a problem per se, it can be solved with a national/mandatory health insurance system like the one we have in most european countries.

In the enlightened USA, The GOP-controlled congress is considering a rollback to the ACA rescission and pre-existing condition regulation that would allow insurance companies to not cover these folks.

Coding is such a nice example to go with. Imagine if gene A showed a 3% higher chance of having some socially extremely taboo sexual attraction? Imagine how a child would be treated, growing up with everyone knowing this about them?

Insurance companies actually have strong incentives to price things in a way that reflects the expected value of the money they'll have to pay out, and they have people who are paid to do those calculations. If some genetic variant makes someone ten cents per month more expensive, on average, then an insurance company would have trouble charging much more than that.

That's not really the threat, though. We're worried about what happens when someone has an 80% chance of having a disease that requires expensive treatment for the sufferer's whole life. At that point, insurance is either too expensive to buy or amounts to giving money away.

Reminds me of GATTACA.

Which reminds me of GATTACA law which congress passed after the movie came out.

> "Oh you can't code? Must not have gene A!"

To which the response could be: "You can't perform logical reasoning? Must not have gene B!"

I wonder how long it will take until some enterprising society will start a new eugenic breeding program, this round with a faster feedback loop. It could start as simple as extra tax breaks for desireable genotypes if they have children, or tax reductions for undesireables if they get long-term birth control. Already in you can reduce prison sentences by voluntary sterilization in Tennessee.

Top down control is kind-of hard. Given increased availability of genetic data, democratised action is easier to envision. You start off with compatibility tests against genetic disorders [1]. Then those that lead to potentially the better offsprings [2]. That opens the door for egg/sperm selection [3]. Then third-party gene introductions [4]. All the way to complete designer babies. These action may even completely cut out evolution from the equation, but evolution can be sneaky.

At every step of the way it becomes slightly more acceptable to push the boundaries just so slightly.

[1] Been the case for many decades now in the case of sickle-cell anaemia, this affects partner selection. In the last decade selection against mutations in BRCA1 and BRCA2 i.e the "breast cancer genes", has lead to searches for egg donors. Also, some people with history of severe depression that runs in the family are self selecting against child bearing (this selection is more common among females).

[2] This could even be pseudo-science at the beginning.

[3] Sperm and egg banks already allow for a crude version of this.

[4] Either natural or synthetic.

It doesn't even need to be breeding, it could be CRISPR technology. How close is that to the mass market? To some parents it could be very defensible as technology to fix a child's genetic defect, which was preventing longer life or peak brain function.

According to this article, it's not yet ready for the mass market: http://www.sciencemag.org/news/2016/05/gene-editor-crispr-wo...

I'm not a biologist, but reading this sounds more like a few years till mass market, not the ominous "30 years" you hear when people really have no clue how long it's going to take :-)

Is CRISPR suitable for doing more than single location fixes? I feel that breeding would work better for traits such as intelligence or longevity that seem to be controlled by many genes, each of which has only a tiny influence.

I am probably being naive but I see nothing wrong with the examples you listed. Isn't this how iceland essentially erradicated downs syndrome? Using information about genetics to inform how we steer the populations' gene pool sounds very good to me.

Some believe that Britain did this in the past; with the combination of the richer in society having many more children, and hanging (gene pool removal) criminals on a regular basis.

IIRC, China is trying to do just that.

Siddhartha Mukherjee had a fantastic point about this on a recent Sam Harris Podcast - "The more specific we make tests, the more likely we'll find genetic reasons".

If we define beautiful as having blue eyes and blonde hair, of course beauty is genetic! That's equally analogous to how we define intelligence and find it similarly heritable. More work needs to be done on the former before we can for sure talk about the latter.

That's a sly attempt at trying to eschew the existence of superiority, make no mistake, it exists

Luckily we already have ample proof that nothing is more important than diversity for survival of populations. Any desire to "optimize" for certain traits is inherently flawed and dangerous. 'Wouldn't it be great if everyone had the gene for a high IQ?' No, it wouldn't. It would make our population a target for parasitic and pathogenic systems which every successful system leads to. And without variation, they could eliminate us no matter how brilliant we might be.

Yeah, bacteria should not be taken lightly, they can evolve drills to chop through hazmat suits any moment.

Bacteria are worrying with their exponential reproduction, but viruses are what worry me. Radically smaller, not alive (so there isn't anything to kill), just a little jumble of parts that if they happen to get bumped into one of your cells, hey, it hijacks everything and forces the cell to create copies of the little jumble of parts until it explodes.

> instead of relying on twin and adoption studies

Twin and adoption studies have never been reliable. They have ranged from dubious methodology to outright fraud such as in the case of Cyril Burt.


Yes, but let's not forget that the variation in IQ within races and genders is larger than between them.

The variation in altitude between the highest and lowest aircraft is greater than the difference between average altitudes of planes and cars.

I could have worded it better but I'm sure you understand what I was trying to say.

With gender, it's easy. There appears to be no solid evidence for any difference in IQ between men and women, in terms of averages.

I found the article confusing, I don't understand the link between the paper and evolution: 1. APOE-Alzheimer-propensity gene is a new gene, emerging in the younger population. (Clearly, prevalence of this gene reduces a person's life-expectancy but not until they are done giving birth). 2. Same with CHRNA3-heavy-smoking-propensity gene. 3. These genes are not evolutionary drivers since they don't effect E[number of offspring] or E[survival rate until given birth to final offspring]. 4. From an evolutionary perspective, prevalence of these genes should be found in equal proportion across all ages of the population. 5. Then why are these genes emerging? Isn't it far more likely that there is some new factor in our lifestyle/diet/habitat etc that is causing the gene to appear in the population? Which has nothing to do with evolution. Where in the paper do they talk about evolution?

The confusing part is that we are used to thinking about evolution as spanning multiple generations: a gene increases fitness, so its prevalence increases over the generations. Or the flip side: it decreases fitness, so the members that have it reproduce less, its prevalence decreases.

But this study takes a snapshot of a population, and looks at gene prevalence as a function of age. So what you see are not effects of reproduction, but rather on survival. It finds the Alzheimer's gene less in 70 year olds, because those with the gene are more likely to have died earlier, and thus not appear in the study.

Thus your statement "APOE-Alzheimer-propensity gene is a new gene" is likely not true. It is likely just as prevalent in the current generation as two generations before; just those individuals from two generations before who had it are more likely to be already dead.

So the study looks at how genes influence survival rates -- something that is very obviously related to evolution if it occurs before reproduction. If survival rates only depend on some genes in higher age, they have only small, second-order effects on reproduction (like a child being more likely to survive if it has healthy grandparents or parents).

(Disclaimer: I'm not a biologist, so take this with a grain of salt).

> It finds the Alzheimer's gene less in 70 year olds, because those with the gene are more likely to have died earlier, and thus not appear in the study.

So wouldn't the study be: "Those carrying genetic disease have a tendency to die early(er)"? What does evolution have to do with it?

BTW, it's perfectly possible for a gene to be wildly successful from the evolutionary point of view, while at the same time killing its owner invariably by the age of 50. How would this study have classified such a gene?

I suppose that does explain the drop off in expression of the gene: Alzheimer's risk of death/year 1-3% [1]. That annual mortality rate is in the right ballpark to explain the frequency of gene expression decrease by about 50% between ages 40 and 80 [2].

Agree the study does look at genes influencing survival rates. I still don't see the link to the headline 'humans are evolving!'. Small second-order effects, okay, sure. So how about, 'Massive Genetic Study shows genetics affect mortality rates, plus maybe there's a small secondary effect on reproduction, who knows?'.

[1]. 83,000 (i) or 3,349 (ii) Alzheimer's deaths per year out of 2mm deaths (iii) => 1-3%

i) http://www.nydailynews.com/life-style/health/alzheimer-kills...

ii) http://www.alz.org/mglc/in_my_community_60862.asp

iii) https://www.cdc.gov/nchs/fastats/deaths.htm

[2] See graphs in paper.

I'm confused. Sure, they identified two alleles with occurrence that decreases with age. And sure, that does suggest that those alleles reduce survival rate.

But to demonstrate evolutionary change, they'd need data on occurrence of these alleles as a function of birth date, stratified by age. It'll take some time to get such data. Probably at least a few decades.

People over 40+ still have kids. So, there would be a direct link if people started to die at 41.

Evolution wants individuals to die soon after giving birth or in the case of mammals, soon after their offspring are independents. Older individuals take resources and keep around an old gene pool.

When considered in a population, mortality is a feature, not a bug. Actually, it seems like aging is an additional feature, not a constraint of our bodies. There are animals that seem to have "accidentally" deactivated it: Galapagos giant turtles, some species of shrimps and, I think, some naked mole rats. These animals seem to have enough causes of mortality to keep evolving, without the need of a timer.

I thought the classical understanding of evolution was that it had no desires...

I consider my interlocutors smart enough to understand that it is an image and that evolution is an emergent behavior.

With regards to your third point, there is a third optimization target you're not considering: E[(other two factors for relatives) • (probability said relative shares the given gene)] - hence humans not doing immediately after ceasing to be fertile, tending to care more about their kinship groups, and possibly also explaining genetic factors in homosexuality (since they may have handy side effects when it comes to capability to help out relatives).

> But if that were the case, there would be plenty of such mutations still kicking around in the genome, the authors argue. That such a large study found only two strongly suggests that evolution is “weeding” them out, says Mostafavi, and that others have probably already been purged from the population by natural selection.

This seems like a lot of speculative leaps to me.

1. assuming that they should necessarily find many of these correlations, absent evolutionary elimination (maybe the effects of individual groups of genes just aren't that strong?)

2. assuming that the lack of them indicates that they existed before and were eliminated (maybe they never existed at all?)

3. more generally, assuming that evolutionary selection is a determining factor in the data they're studying at all (I don't often see noticeable changes in a few generation in highly-contrived genetic algorithms I play with, let alone real-world data which has way, way more noise)

* Analysis of 215,000 people's DNA suggests variants that shorten life are being selected against.

* tested more than 8 million common mutations, and found two that seemed to become less prevalent with age. A variant of the APOE gene, which is strongly linked to Alzheimer’s disease, was rarely found in women over 70. And a mutation in the CHRNA3 gene associated with heavy smoking in men petered out in the population starting in middle age. People without these mutations have a survival edge and are more likely to live longer, the researchers suggest.

* certain groups of genetic mutations, which individually would not have a measurable effect but together accounted for health threats, appeared less often in people who were expected to have long lifespans than in those who weren't. These included predispositions to asthma, high body mass index and high cholesterol.

* Most surprising, however, was the finding that sets of mutations that delay puberty and childbearing are more prevalent in long-lived people.

This puzzled me, as these Alzeheimer's and smoking results do not seem to be evidence of evolution; one would surely need to see a decrease over generations? The article goes on to say 'That such a large study found only two [late-acting genes] strongly suggests that evolution is “weeding” them out, says Mostafavi, and that others have probably already been purged from the population by natural selection.' The article and paper then repeat some hypotheses as to why natural selection would eliminate these hypothesized former late-acting genes from the gene pool, given that they seem to act after most people's reproductive years.

So the article and paper seem to be arguing that a low frequency of late-acting genes indicates ongoing evolution, but acknowledge that this raises questions about why natural selection would act on them.

The relevance of the decrease of the prevalence of these genes with age is apparently that survivorship bias is being used to distinguish causation from mere correlation.

An Attenborough TV show suggested that the reason people live long compared to animals is that grandparents are needed to help raise the young. The grandparents raise the young while the parents work/hunt/farm/whatever.

So having healthy grandparents gives a survival advantage.

This theory is mentioned in the article as well, but it doesn't seem very plausible if we're talking about living into the 80s or 90s.

At that age, the effect should be reversed as people often need care themselves rather than caring for grandchildren (who may be adults by that time anyway)

I suppose the "grandmother theory" originally only explained why women live beyond menopause, not into their 70s, 80s or 90s.

Alzheimer's in particular does not reduce lifespan all that much; rather, it increases the resources that children and grandchildren need to use to care for the parent. So selection against Alzheimer's would presumably (we have very little evidence for any of this) be even stronger than selection against traits that, say, killed at 70.

Is actually caring for an Alzheimer's stricken relative a new phenomenon? Could it be that a hundred years ago, a chronically ill, continually deteriorating relative was just left in a room to waste away?

Professional Alzheimer's care is new;, without good professional work you're going to get sores and muscle atrophy and all kinds of other complications that can be fatal at old age. We've gotten very good at dealing with those, but not at actually curing the underlying disease. (Speaking from family experience.)

Maybe even more importantly, though, 100 years ago other, unrelated diseases of old age were likely to get you well before Alzheimer's really got bad.

Even today, people rarely die from Alzheimer's, because something else gets them first, as Alzheimer's is a very slowly progressing disease.

Increasing longevity to the point of incapacity may be a consequence of improved health care, rather than biological adaptation.

And while average life expectancy has been increasing for a long time (thanks to improved health care), the maximum life span has not. The oldest living people are still "only" about 115-122 years old.

Is this because "improved health care" only makes typical illnesses treatable/prentable whereas maximum lifespan (ie, the deteriation of cells/organs/etc with age) are not being focused on? Or is it because they are very challenging things to extend?

Primitive societies are known to abandon old people to die when they become a burden rather than an asset.

An ugly, indisputable fact. Saddening to see you downvoted.


What's a nomadic tribe to do with someone who can no longer keep up? When there is insufficient food (commonplace in older societies) who is going to get the food?

When resources get scarce, there are a lot of ugly realities.

You might do better to provide citations over argument.

"Traditional nomadic tribes often end up abandoning their elderly during their unrelenting travels. The choice for the healthy and young is to do this or carry the old and infirm on their backs — along with children, weapons and necessities — through perilous territory. Also prone to sacrificing their elderly are societies that suffer periodic famines. Citing a dramatic example, Diamond said Paraguay’s Aché Indians assign certain young men the task of killing old people with an ax or spear, or burying them alive."


Thanks, exactly.

But many animals reach sexual maturity much faster (e.g. cattle at ~40 months) so you would expect many more generations to be alive simultaneously.

I had the same reaction as you at first, but read carefully:

"Mostafavi and his colleagues tested more than 8 million common mutations, and found two that seemed to become less prevalent with age"

Old people are less likely to have this gene than young people. This can have two possible explanations: old people with this gene are more likely to are already dead (and thus not show up in the study), or the gene is becoming more prevalent in the younger generations. The statement "People without these mutations have a survival edge and are more likely to live longer" favors the first explanation.

This is not the effect of natural selection on gene prevalence. (And it would be weird for Alzheimer's, which usually occurs at an age where reproduction is often finished, and maybe even raising the kids is done already).

My apologies for editing my reply to add a third paragraph while you were replying to it, but I think that explains the relevance of the age dependence to this study: the decrease in the prevalence with age suggests that they actually do exacerbate the diseases they are associated with, and so the association is not merely a correlation.

Note that they are not suggesting these genes have been the subject of selection; it is the hypothesized number of such late-acting genes that were not found that are assumed to have been selected against.

> This puzzled me, as these Alzeheimer's and smoking results do not seem to be evidence of evolution; one would surely need to see a decrease over generations?

At least with smoking, we've been experiencing exactly that decrease for generations. Smoking has become vastly unpopular compared to how prevalent and socially accepted it used to be. Social factors had been one of the main driving forces, but that doesn't mean it has to be the only driving force.

The social issues complicate the problem of identifying genetic effects.

Could be that the genes have a deleterious effect in youth, not just old age.

Or it could be that aging is a complex process which a naive experiment like this is unlikely to shed much light on.

That is one of the hypotheses advanced in the article.

That last one may have a great deal to do with it.

As people live longer and safer lives, they can afford to have children later. And child mortality has been seriously reduced, so the children they have are comparable to thise among people from previous centuries who had many children younger.

Having children takes a toll on an organism, and having many children takes a larger toll, presumably. I wouldn't be surprised if there is a correlation between lifespan and number of children, among women.

So - people live longer and have children later. The children grow up richer and bigger and in turn have children later, as well.

I bet there is a correlation between age of having children and lifespan.

I definitely think postponing reproduction would coincide with longevity, but for a completely different/same-but-differently-stated reason. Evolution doesn't seem to care very much about the survival of an organism once they've successfully reproduced an optimal number of offspring, except to the extent that the organism can see to the well-being of offspring.

It stands to reason that postponing reproduction for as long as reasonably possible would contribute to selection for longevity. From evolution's perspective, you have to survive at least that long, or else you're not good for anything. If humanity collectively decides to postpone having children, evolution may begrudgingly allocate more of its resources toward the welfare of individuals, rather than viewing our early deaths as an acceptable loss in a cold utilitarian calculation about long-term gene proliferation.


What about the increase in defects/problems with women having children in their mid-late 30s? And older father's sperm containing more mutations https://www.nature.com/news/fathers-bequeath-more-mutations-...

You are right that if this were attempted on day one by banning pregnancies by everyone except thirty-somethings, it would be a non-starter, because of the many unsolved health issues with pregnancies at that age. It wouldn't work if we put all the eggs in that basket. The strategy would be to gradually expand the range of sustainable, healthy ages to have pregnancies, so that the outer extremes eventually become part of the "heartland" range of sustainable pregnancies.

You could imagine a graph showing the percentage of pregnancies that occur at each age. The strategy would be to gradually move the percentages over time so that some percentage of sixteen year old pregnancies get shifted over into seventeen year old pregnancies, some seventeen year old pregnancies become eighteen year old pregnancies, and so on.

The goal is that, over time, there is escalating pressure to solve health problems that occur at later ages, problems that evolution currently has little incentive to address.

Very interesting! I find this kind of line very exciting.

But dows that mean some sort of Lamarckian evolution is taking place after all?

Sexy son hypothesis is also similar. Is it related?

>I bet there is a correlation between age of having children and lifespan.

There is (at least for women).


> A significant association for later maternal age was found whereby women who had their last child beyond the age of 33 years had twice the odds of survival to the top 5th percentile of survival of their birth cohorts compared to women who had their last child by age 29 (OR=2.08, 95%CI 1.13; 3.92 for age between 33 and 37 years and OR=1.92, 95% CI 1.03; 3.68 for older age).

It's important to know though that delaying childbirth among women is associated with all sorts of other "healthy" factors like high education, high income, stable marriage, etc.

mothers, even mothers who don't carry all the way to birth, get fetal cells, very similar to stem cells, that protect and repair their mothers, but can also trigger autoimmune conditions like lupus, or rheumatoid arthritis.


Apparently I was wrong - according to this


But this study is contradicted by many studies


the radiolab is from 2012, and i'm sure they've learned a lot since then, but they do mention that there are a lot of factors (sex of children, number of children, number of fathers) that correlate with both good and bad effects of the fetal cells.

So which is it?

Do children at 20-30 increase or decrease life span?

Obviously people who live longer may have more children, especially the men. I am talking about the other way.

The simple answer is its complicated. Having a child is stressful on the organism, but there are ultimately 2 types of stress eustress and distress.

Sometimes it's good sometimes it's bad. ¯\_(ツ)_/¯

there is i tell you ;) at least 20 years less off your life.

The last point isn't surprising at all. I think there is a subset of the population that matures quickly, reproduces quickly, and dies quickly.

From my own observations in middle school and high school, those that mature earlier are far more likely to get into drugs, forgo education, etc... all things that are measures of success in our current society

Success in our society is not a good measure of evolutionary success. Evolution is only concerned about the number of viable offsprings you produce, not how well off they are.

For reference, see Idiocracy opening sequence... https://www.youtube.com/watch?v=YwZ0ZUy7P3E

Smart people delay, wait and then end up not... "dumb" people just hook up and go to town...

I think the fact that Evolution is still spreading "successful" genes and "unsuccessful" genes become less prevalent is obvious.

What isn't obvious, is as Idiocracy so succinctly puts it, is that gene that are "successful" doesn't necessarily translate to intelligence, good decisions, stable lifestyle, etc.

That successful genes spread is a tautology. The frequency of a gene in a population compared to its alleles is its success. I wish we could retire the idea of success in an evolution and just talk about gene frequency. Then the irrelevant ideas of goodness and progress don't come up so readily in people's minds.

I agree with the premise, these reports seem to focus on the act of passing your genetic material to a child as the final act. Anyone who has had kids knows that passing on your genes is the easy part, raising them to be productive members of society who will in turn procreate takes a long time and you continue to influence and guide them. So something that shortens your lifespan would reduce the survival edge of the remaining offspring compared to a group with that guiding influence.

This is a pretty clear social structure in humans. I understand not all or even most animals have that same social structure. Can someone explain to me why they would assume remaining alive after you dump off your genetic material would be neutral? (from an evolutionary perspective)

This is addressed in the article:

Why these late-acting mutations might lower a person’s genetic fitness — their ability to reproduce and spread their genes — remains an open question.

The authors suggest that for men, it could be that those who live longer can have more children, but this is unlikely to be the whole story. So scientists are considering two other explanations for why longevity is important. First, parents surviving into old age in good health can care for their children and grandchildren, increasing the later generations’ chances of surviving and reproducing. This is sometimes known as the ‘grandmother hypothesis’, and may explain why humans tend to live long after menopause.

Can someone explain this to me? The article says: "People who carry a harmful genetic variant die at a higher rate, so the variant becomes rarer in the older portion of the population." However, this is the opposite of how I understand evolution: "harmful" variants (from the evolutionary point of view) should become rarer in the younger portion of the population- whatever happens to their original carriers is irrelevant.

Also "researchers scoured large US and UK genetic databases"- well, I wouldn't be too sure that the biggest part of evolution happening in the human species can be seen inside the UK and US genetic databases.

> one of the first attempts to probe directly how humans are evolving over one or two generations.

Considering the extraordinary amount of time it takes for evolution to do anything at all, that they can quantify it in what is relatively zero time in the span of life on Earth, seems to remarkable to be true.

This is a very, very stubborn idea about evolution that has proven to be kind of false. Rapid evolution exists, and we can observe evolution happening in many species:


As I understand, the simplest explanation for species being so stable over millions of years (like, say, sharks) is that while evolution can happen quite fast, it "averages itself out" - there's a kind of regression-to-the-mean effect.

Also everyone forgets evolution is about adaptation. It's not about "levelling up" across-the-board like in a video game. If a species is already optimized for the environment, and the environment remains stable, of course you won't see much evolution!

I have been taught something different in school, but that was a while ago, so perhaps the field has evolved some in recent years. That, depending on the species, evolution rarely occurs between generations, and that it leans on large changes that are sudden and far between, and these changes are not at the same time for each species. That is to say, evolution is not usually a linear process for which at any point in time there is a delta. It is more of a stair step.

That is part of what makes the report hard to believe, even if it is a novel idea and has some other merits.

Not to say it couldn't be true, but as I expressed, it seems remarkable that such evidence was found in a single generational study.

You're right that it is a relatively recent development: the first papers (cautiously) going against this idea of evolution happening slowly started appearing in the middle of the last century. The first paper flat-out stating evolution was observed within a few generations was in the late 1970s (mentioned in the linked article in my previous post). So you can imagine it would take a while for the field to accept this new attitude, let alone how long it takes for this new knowledge to then diffuse into the rest of society.

Either way my comment certainly wasn't intended as a critique aimed at you!

Also, thinking about it some more, it's not that strange that the idea of slow evolution hasn't died out yet, because it's not so much wrong as that reality more complicated and really counter-intuitive than what you think if you only look at fossils[0]:

> In the 1950s, the Finnish biologist Björn Kurtén noticed something unusual in the fossilized horses he was studying. When he compared the shapes of the bones of species separated by only a few generations, he could detect lots of small but significant changes. Horse species separated by millions of years, however, showed far fewer differences in their morphology. Subsequent studies over the next half century found similar effects — organisms appeared to evolve more quickly when biologists tracked them over shorter timescales.

> Then, in the mid-2000s, Simon Ho, an evolutionary biologist at the University of Sydney, encountered a similar phenomenon in the genomes he was analyzing. When he calculated how quickly DNA mutations accumulated in birds and primates over just a few thousand years, Ho found the genomes chock-full of small mutations. This indicated a briskly ticking evolutionary clock. But when he zoomed out and compared DNA sequences separated by millions of years, he found something very different. The evolutionary clock had slowed to a crawl.

(I have to apologise for not remembering this article yesterday, it would have made things more clear)

So in retrospect my "very stubborn idea" assessment is probably a bit unfair.

But, back to what we were taught in school. Caveat: I'm not a biologist, let alone evolution expert, just a nerd who is into this topic. There are a number of issues with the description you gave (which, for the record, fits with how it was taught to me in high-school)

It is true that evolution is not linear, but being taught that evolution is a stair step is just as bad. It (implicitly) assumes evolution is a form of progress. It's not, it's just adaptation. The simplest of life forms that exist nowadays are just as successful as we are, in the sense that they also didn't go extinct and will continue to thrive in the future. On the other end of the spectrum you actually can have runaway adaptation, especially when there's a kind of "arms race" between predator and prey. The cheetah is so hyper-specialised for catching its prey by speed alone that it's a kind of evolutionary dead-end; it can't zig-zag like the gazelle can. Anyway, the stair-step attitude originates from outside of biology, from Platonic thinking that there are "higher" ideal forms.

And the sudden changes turn out not to be "few and far between", they just don't seem to stick. Most adaptations average out, because they're highly contextual to their short time periods. As mentioned, since we based our first ideas about evolution on fossils, we only noticed the slow evolution. Sex actually seems to play a large part in the filtering process[1]:

> For a species whose numbers show no signs of collapsing, humans have a shockingly high mutation rate. Each of us is born with about 70 new genetic errors that our parents did not have. That’s much more than a slime mold, say, or a bacterium. Mutations are likely to decrease an organism’s fitness, and an avalanche like this every generation could be deadly to our species. The fact that we haven’t gone extinct suggests that over the long term, we have some way of taking out our genetic garbage. And a new paper, recently published in Science, provides evidence that the answer may be linked to another fascinating procedure: sex. (...) As the number of nasty genetic errors in a population rises, natural selection will sweep large rafts of them out of the genome together. And in sexual organisms, because of the ways that mutations from each parent can recombine randomly onto the same chromosomes, the synergistic elimination of bad mutations can happen even faster.

Anyway, hope this addresses some of your understandable scepticism!

[0] https://www.quantamagazine.org/evolution-runs-faster-on-shor...

[1] https://www.quantamagazine.org/missing-mutations-suggest-a-r...

I mean, I'm assuming these people are professional biologists of some sort, that have probably spent decades in their field. Not only that but they would know more about how evolution works than the average college educated layperson.

I'm a little frustrated to point that out. However my frustration isn't enough to motivate my next question. What did you hope to accomplish, in light of these problems, by criticizing these scientists' work with what is ostensibly, cursory popular science knowledge?

I guess I just want to understand why there is a mistrust for professional scientists even on a site as technologically inclined as HN.

> I guess I just want to understand why there is a mistrust for professional scientists even on a site as technologically inclined as HN.

I think you are new to HN. There is a long history of healthy skepticism by the readers here of studies posted in a wide range of fields. Healthy skepticism, or at least, pointing out remarkable and hard-to-believe aspects of a reported study (as I did here), is quite commonplace all over HN.

In fact, if you read any of the nearly hundred other comments in this thread, you will see equal amounts of skepticism, some more verbose, some more terse, than what I expressed. Fortunately, you did not go and criticize those commenters also.

In the case of evolution, I was taught years ago in school that evolution is not a gradual and linear process, and that often very long stretches occur with zero evolutionary action, then a single evolutionary event causes a gigantic leap of changes in a species, a single spike, which then plateaus again for a long time another spike.

I am not an evolutionary scientist of course, but I was leaning on that, and layman's perception of evolution, in my comment.

I'm frustrated to have to point this out. What did you hope to accomplish by criticising my reaction to the article? Are we not all here to comment on our opinions and reactions to the items posted? Just because you do not agree with my particular reaction to the article does not mean you should make such wide assumptions about my own motivations (or those of any other reader you choose to attack in the future).

Imagine a conception occurs with a random mutation that causes a miscarriage. That's evolution in a few months.

Imagine one that causes someone to be infertile. The balance of genetics changes (evolution) in a generation.

Africa went from 1/2 the population of Europe in 1950 to 1.6x the size today. The gene frequencies on those two continents are different - thus there has clearly been differential fitness (in the hundreds of millions in a lifetime)

> That's evolution in a few months.

A single isolated event with no impact on the species is not the definition that most would use for evolution. And even if it had an effect on that individual family, evolution is about nature selecting that particular mutation, which may not happen at all. So I don't support the premise of your argument.

Evolution is the change in gene frequencies over time. Accepting my premise is just knowing the textbook definition.

I think what you are trying to describe is "novel fitness increasing mutations" - these won't be detectable for ages as it takes time to spread through a population.

Single isolated events matter - All mutations start off in an individual, they then increase or decrease in frequency.

The natural time period of evolution is the generation; meaningful change must occur over that sort of period, or it is not going to happen at all. Evolution is going on all the time; it's just the major effects, such as the establishment of a new family, that usually takes a long time (I say 'establishment' over 'emergence', as talking about the time for emergence runs into the sorites (you have to draw a line somewhere) problem.)

> The natural time period of evolution is the generation; meaningful change must occur over that sort of period

Unless you have specific education in this area, this is contrary to the classes I took in university on the topic of animal biology. There is no "natural time period" for evolution, as you say. Meaningful change usually does not happen over any arbitrary period. It is typically a random event that spikes rarely in a species, and usually there are long periods of zero change until such spikes occur between these changes.

I suspect you are postulating on what evolution is without actually knowing how it most often emerges.

Firstly, a matter of terminology: my use of 'natural period' may be wrong here, if it implies a cyclical process; I am not sure what might be the correct term, so I shall use 'characteristic period'.

You are confusing two different but related issues, and may have misunderstood your professor. The timing of large effects (speciation, for example) is unpredictable, and there are long periods in the history of any given line of descent during which no speciation-level evolution occurred (note that the article is about evolution at a much smaller scale than speciation. These events are unpredictable in detail, but occur much more frequently.)

As for the process by which it happens - variation and natural selection - the generation is, to a first approximation, a characteristic unit of time. A mutation occurs at a point in time, but it is only subject to being selected by natural selection if and when it is passed in the next generation, and natural selection works by creating more of the next generation on account of it.

This is somewhat complicated by the appearance of neutral mutations, which do not have effects that are selected for or against. These take the genome on a continual but evolutionarily directionless random walk, for which the characteristic time period is the generation. There is the possibility that, as a consequence of these random walks, any given allele's frequency in the population will drop to zero, after which it will not reappear unless it is recreated by another mutation.

While neutral mutations, by definition, have no evolutionary effect at their inception, they may become significant either as a consequence of a change in the environment or through the allele's interaction with, or modification by, a subsequent mutation. Whenever that happens, however, natural selection still works by changing the occurrence of the allele from one generation to the next.

On reflection, my phrase "meaningful change must occur over that sort of period, or it is not going to happen at all" is inaccurate; to express what I was thinking, I would now write that "when evolution occurs, it is manifest from one generation to the next."

I see that you are aware that evolution has a tendency to occur in spikes, so it seems odd that you would write, in your initial post, "Considering the extraordinary amount of time it takes for evolution to do anything at all..." There is no requirement for there to be long periods of stasis between events.

It takes a long time for the changes to be obviously observable. But at least some change has to occur in any given generation, in order for it to build up to 'significant' change over time. So now that we can look at genes directly, it makes sense that we would potentially be able to identify these changes.

> But at least some change has to occur in any given generation

This is not usually true and not typically how many species have evolved. Typically a long sequence of generations will exist with zero change, and then individual evolutionary events occur to create very rapid change in a very short period of time before stabilizing with long periods of no change again.

A lot of people here are expressing what they seem to think evolution is without apparently having ever actually studied it; that is my take-away anyway.

Interesting. I'm unfamiliar with that position, but as you say, I'm certainly not an expert in the area. A bit of googling didn't turn up evidence one way or the other; do you happen to have a reference?

My main reference is what I was taught in an animal behavior course several years ago at university. There are some interesting books on this subject on my reading list, but I haven't yet gotten to them.

I'm confused about how the inferences are being made. If genes are found in older people, okay, so what? That doesn't mean anything. But then the article says:

> But if that were the case, there would be plenty of such mutations still kicking around in the genome.

So does that mean the gene is less common in children than parents? If so, what does it matter that older people don't have the gene? Why does age factor into this at all?

* This is not, by itself, evidence of evolution at work. In evolutionary terms, having a long life isn’t as important as having a reproductively fruitful one, with many children who survive into adulthood and birth their own offspring. So harmful mutations that exert their effects after reproductive age could be expected to be ‘neutral’ in the eyes of evolution, and not selected against.


We want to live more (and more). But that's not the main point of life, as science (and most religions) say.

We are in our best form when we reproduce and give our best to our children (and family). So they can do the same in the future.

Today's society tries to sell the idea of small families (or no family at all ) and long lives as the best option. But maybe it's not the case.

This reminds me of my core mission: sterilize myself or self terminate.

Everything I've experienced and felt since youth indicates that my seed is unwanted by humanity, and it's clear suicide helps evolution by eliminating bad genes. I think about suicide constantly and have done so since youth. This is my brain telling me to help humanity by leaving. Hoping to complete that mission ASAP.

I wouldn't necessarily go along with your logic here. If you decide not to have children for one reason or another you can implement that decision without suicide. A lot of people have suicidal thoughts at some stages in their lives, then recover and are glad they didn't take a non-reversible action.

Anyway, it's too dangerous a topic for you to work things out without a bit of "peer review". So I strongly recommend to contact a counsellor.

This isn't occurring st some subset of stages; it's been steady throughout my entire life. And I've grown more and more cynical with age. It's a decision that I've long intellectualized.

People who don't survive aren't around anymore to discuss whether they are satisfied with their decision. Why would I waste my time talking to someone who intends to talk me out of something that isn't a debate to me? It's a personal decision.

Have you considered Buddhism, the entire purpose of which is to learn to overcome suffering? Here's a summary: http://www.bcc.ca/buddhism/fournobletruthsandeightfoldpath.h...

Suicide doesn't help evolution. If you have bad genes, you die before you can reproduce, or your children do. If you and your children survive, those were good genes. You don't have to do anything to manage evolution, it takes care of itself by its own.

I disagree. With a high sex drive, it's probable that I would reproduce eventually, risking passing on my genes. Suicide ensures that procreation won't occur.

If you suspect your offspring (or your genes) would somehow be harmful to humanity, no one is asking you to produce any. But calling suicide your "mission" just betrays spite and resentment. What's so urgent about it?

Isn't it really about wanting to escape suffering?

People are having children later on in life so it would make sense.

It is odd that longer lifespans would be selected for. Given two otherwise equivalent populations, the one with the shorter lifespan will come to dominate.

This does not seem intuitive to me. Would not having a longer lifespan allow a subset of a population to better care for itself through social means?

This study absolutely does not show that humans are evolving

yeah - a gene that kills you is less common in people over the age that it tends to kill you? SHOCKER! Also, super high testosterone dudes probably inseminate more women than betas and they probably die younger too.

Unless the statistical distribution of alleles is exactly the same (unlikely) from one generation to the next then humans would have to be evolving, wouldn't they?

Genetic drift is not usually considered 'evolving' because the variants are selectively neutral. That's why finding selected variants isn't as simple as 'SNP X is 0.00001% more common in generation 2 than generation 1, case closed!' - you are looking for variants which are becoming more/less frequent than would be plausible under simple drift. Just like an experiment where you know the effect will never be exactly 0 because of chance imbalances and random noise.

An explanation not explored in the paper: the social pressure to have children later in life favors people living older.

we're evolving mostly through sexual selection now as opposed to natural selection but still some of that too

What's the difference between sexual & natural selection ?

Sexual selection can still happen even if humans become immortal.

Natural selection cannot.

You do not even need a scientist to see that there is a massive degeneration and retardation going on in the US gene pool currently. Influx of foreign genes was the solution, but now dumbness itself will trigger its own growth even stronger. If you want a better gene pool for your descendants it is time to leave US now. Good Luck!

This comment is way outside the standards of civility and substantiveness here. We ban accounts that continue to post like this.


Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact