Edit 2025-01-28: Due to the skeptical GWAS reported in the top comment, I've downwardly updated some of my probabilities.
I think it's unlikely (10%) that just directly applying the DEC2, ADRB1, or GRM1 mutations would make a difference in the majority of humans, since that's what the GWAS says. There is a positive GWAS (-20 minutes sleep duration) for an NPSR1 mutation (though a different one), so I'll say 30% that the below NPSR1 mutation would give you at least 1 hour of shortened sleep.
I think that, when the DEC2 mutation fails in the general population, it's likely via "that mutation not leading to high orexin levels", not via "high levels of orexin failing to promote wakefulness". Which means there's a potential path to victory by administering orexin (or orexin agonists), or the other neurotransmitters corresponding to the other mutations, to adults. Caveat: it could be unsafe to do this with adults whose brains didn't grow up that way, and also there could be other necessary mutations we don't know about. Overall, I give 60% likelihood to this being workable.
The rest of the post is left unchanged below.
There is decent evidence that there are "short sleeper genes": individual genetic mutations that cause a human to naturally get only 4-6.5 hours of sleep per night (instead of ~8) while remaining healthy, happy, and productive. There don't seem to be any negative side effects, although there might be positive ones.
So far, 5 short-sleeper mutations across 4 genes have been identified. Probably at least one of these is a good candidate for developing a drug to mimic the effects of the gene.
Such a drug would be enormously valuable. If it worked, presumably it would give people at least one extra waking hour per day, and maybe the "higher resilience and productivity" side effects would be real too. If you care about overall utility[1], then consider the value of 1 extra waking hour per day, times 365 days per year, and imagine that 100 million people ended up taking the drug; if we say each hour is worth at least $1 (a lowball), then that's $36 billion per year.
Just to illustrate the point, if an investment yielding 3% returns were that profitable, the investment would be $1 trillion. Even if you assume the probability of success were as low as 1%, that'd be worth $10 billion. Such an opportunity is clearly worth looking into.
I've skimmed several of the papers on the subject of familial natural short sleepers (FNSS), most of which were written by UCSF professor Ying-Hui Fu and colleagues. (Disclosure: I've emailed Ying-Hui, video chatted with her, and consider her something of a friend.) As I understand it, what they've done is collect people who might be natural short sleepers, validate that they are, look among family members, find candidate short-sleeper genes, and test those genes in mice or similar organisms, and publish papers if an effect is demonstrated in the mice.
The DEC2 mutation reduced sleep in mice by about an hour. The ADRB1 mutation reduced by 55 minutes. NPSR1 by 71 minutes, and the two GRM1 mutations (roughly equivalent to each other) by 25 minutes. Ying-Hui says it's expected that the effects in mice are smaller than in humans.
All the genes in question are known to affect wakefulness. For example, the DEC2 protein suppresses expression of a neuropeptide called orexin, and mutant DEC2 suppresses it less; orexin is known to promote wakefulness (e.g. people who can't produce orexin are narcoleptic). This part of the mechanism seems reasonably clear.
What's less clear is how these genes accelerate the repairs/maintenance the body performs during sleep. (If they did not do any such acceleration, then you'd expect these people to be miserable, unhealthy insomniacs.) But there is at least one experiment that seems to show that they do accelerate some type of repair:
They took mice that were engineered to generate amyloid plaques and tau protein tangles—pathologies that seem associated with Alzheimer's disease (though the causality seems unclear and you should take such research with a grain of salt). Anyway, they gave some of them the DEC2 and others the NPSR1 mutations, and found that both genes significantly reduced those Alzheimer-correlated pathologies.
I'll just directly quote this summary article:
[Short sleepers] report greater flexibility around sleep timing and less subjective deficit after sleep deprivation. They often deny experiencing jetlag (unpublished data). However, it is not just sleep duration that characterizes this group of individuals. There is a high behavioral drive among those with FNSS, and individuals report a need to always be mentally active resulting commonly in high profile, high pressure jobs or holding multiple jobs. FNSS individuals also appear to have high pain thresholds and relative resilience to life stressors (unpublished data). Thus, the phenotype does not only encompass a sleep pattern but also daytime behavior. Individuals with short sleep and increased behavioral drive have been described in the literature as early as 1972 when Hartmann et al., while studying sleep need variation, described those who sleep less than 6 h as “smooth [and] efficient” and with greater tolerance and flexibility and less depression compared with long sleepers [123]. It is possible that DEC2, ADRB1 and other causative mutations all lead to a higher behavioral drive, which allows those with FNSS to overcome increased homeostatic sleep pressure though much work needs to be done to test this hypothesis.
And, from a 2015 BBC article profiling a short sleeper:
A positive outlook is common among all of the short-sleepers that Fu has studied. “Anecdotally,” she says, “they are all very energetic, very optimistic. It’s very common for them to feel like they want to cram as much into life as they can, but we’re not sure how or whether this is related to their mutations.”
More from the summary article:
FNSS is defined as a stable trait of sleeping 4–6.5 h/night without daytime sleepiness or clear impairments (Table 1). These individuals are sometimes seen in sleep clinics because they are told they “need” more sleep—they are sometimes considered insomniacs. However, when these individuals are asked how they feel after sleeping 4–6.5 h, they often report feeling “great” and “well rested.” [...]
Individuals with FNSS report decreased sleep need throughout adult life, some dating to childhood. They report sleep durations ranging from 4 to 6.5 h per night without daytime sleepiness or reported deficits from sleep deprivation. FNSS individuals are in part distinguished from facultative short sleepers by their lack of catch-up sleep on weekends and free days.
Ying-Hui says, in private communication: "From our observation on those natural short sleepers during past 15 years, there is really no one negative thing that we can put our fingers on." Also: "We have carefully characterized more than 100 (somewhere around 160) FNSS and have not seen major negative problem. However, it’s not impossible that we have missed some very subtle negative effect."
If there were long-term negative consequences from the lack of sleep, I think we'd expect them to accumulate over time (and therefore be worst in older people). Ying-Hui likes to mention an over-80-year-old triathlon competitor and some other impressive old people, and says "Many of the FNSS individuals have remarkable memory and cognitive capacities (unpublished data)". I believe her opinion is that, if anything, FNSS people are healthier on average because they're better at self-repair through sleep.
I do suspect there are selection effects: people who would go visit a sleep clinic despite feeling fine (and therefore be discovered as FNSS) seem likely to have more disposable cash, and perhaps be more medically conscientious (or scientifically curious) than average, which probably correlates with various advantages; one should bear this in mind if comparing against the general population. This would extend, less strongly, to the FNSS relatives one discovers from the first one—but equally to their non-FNSS relatives, who are an ideal comparison group. I don't know if that comparison has been made.
Overall, I suspect that, if there are any negative side effects, they are (a) small to very-small and (b) outweighed by the positive side effects. (This is where "having a drug you can start and stop taking when you like" may be superior to actually having the gene.)
Common wisdom is that some essential maintenance and repair processes occur during sleep; also that if you don't get "enough" sleep (and <6 hours is usually not "enough"), it causes misery and cognitive underperformance in the short term, and there's good reason to expect longer-term negative mental and physical consequences. (Some have expressed contrarian views.)
Wakefulness-promoting chemicals, which the mutations seem involved with, could probably suppress the "short-term misery and cognitive underperformance", but it's unclear how they would accelerate the repair processes. And since some of the mutations are only a single base pair, it would have to be the very same protein that caused both the reduced tiredness and the accelerated repair. Once could be a coincidence, but four times on different genes suggests something more fundamental. Does wakefulness promotion inherently accelerate repair? How?
A few possibilities come to mind:
Or maybe the genes don't accelerate repair. This is contraindicated by the Alzheimer mice result, but let's say we dismiss that. Possibilities that come to mind:
Readers may form their own probability estimates. I think I give at least 50% that the genes work as advertised—and I stand by my conclusion that it's well worth investigating.[3]
There are several things one can follow up on. For those who want to validate the premise that short-sleeper genes work as advertised, I mentioned a few tests and readers can probably think of more (if one can gain access to population-wide genome databases and get data about others with the known mutations, that could be very nice).
For those who want to work towards developing a drug, there are various paths of advancement. For example, with DEC2, one could check (first in mice) if administering orexin (which can be done with nasal spray[4]) at the right dose and schedule has the same effect as having mutant DEC2; if so, that could be the drug. If not, then you look for something that sensitizes the orexin receptor—or something, this isn't my field. Ying-Hui and her team have plenty of ideas about how to proceed; that's not the bottleneck here (though it is fun to think of ideas).
The big problem is funding. Ying-Hui was saying on Reddit in 2015 that funding was insufficient, and it seems to have gotten worse since then; she has said she's "pretty much given up on trying to get NIH to fund my research" and she's had to lay people off from her sleep lab. The cost to run the lab is in the low millions per year. Compared to the expected value of the research, the cost is laughable—but unfortunately it would drain my personal savings pretty quickly, and I must appeal to others.
So. I'm trying to tell people about this trillion-dollar bill lying on the sidewalk. Who's going to pick it up?
Wikipedia on FNSS: https://en.wikipedia.org/wiki/Familial_sleep_traits#Familial_Natural_Short_Sleep
FNSS in general (search page for FNSS): "Genetics of the human circadian clock and sleep homeostat" (2020): https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6879540/
DEC2: "The Transcriptional Repressor DEC2 Regulates Sleep Length in Mammals" (2009): https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2884988/
"DEC2 modulates orexin expression and regulates sleep" (2018): https://www.pnas.org/doi/10.1073/pnas.1801693115
ADRB1: "A Rare Mutation of β1-Adrenergic Receptor Affects Sleep/Wake Behaviors" (2019): https://www.sciencedirect.com/science/article/pii/S089662731930652X
NPSR1: "Mutant neuropeptide S receptor reduces sleep duration with preserved memory consolidation" (2019): https://www.science.org/doi/abs/10.1126/scitranslmed.aax2014
GRM1: "Mutations in Metabotropic Glutamate Receptor 1 Contribute to Natural Short Sleep Trait" (2021): https://www.cell.com/current-biology/pdfExtended/S0960-9822(20)31441-X
Alzheimer mice + FNSS experiment: "Familial natural short sleep mutations reduce Alzheimer pathology in mice" (2022): https://www.cell.com/iscience/pdf/S2589-0042(22)00234-6.pdf
Press release: https://www.ucsf.edu/news/2022/03/422416/when-it-comes-sleep-its-quality-over-quantity
2015 BBC article that profiled a short sleeper, plus quotes from Ying-Hui: https://www.bbc.com/future/article/20150706-the-woman-who-barely-sleeps
2015 Reddit AMA with Ying-Hui: https://old.reddit.com/r/science/comments/3kj669/science_ama_series_im_yinghui_fu_i_study_the/
Nasal spray orexin: "Systemic and Nasal Delivery of Orexin-A (Hypocretin-1) Reduces the Effects of Sleep Deprivation on Cognitive Performance in Nonhuman Primates" (2007): https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6673447/
For those who care only about research projects such as AI safety, I'd still say it's pretty valuable. Imagine if all your current and potential researchers had an extra hour per day, and were resilient to sleep deprivation, and had "a need to always be mentally active resulting commonly in high profile, high pressure jobs or holding multiple jobs"-tier motivation.
A more precise model, which I suspect does describe some sleep functions, would be: During wakefulness, some trash (metabolites, toxins, whatever) accumulates at rate R. During sleep, the body pumps in clear liquid and pumps out the impure liquid, which, assuming perfect mixing, means the metabolites drain at rate K*M, where M is the current level of metabolites. As calculus students learn, implies : exponential decay. But the takeaway is that delaying this type of repair makes it more efficient.
Of course, there are other possibilities. The straight linear model is: "During wakefulness, resource X (e.g. cell glycogen) is depleted at a fixed rate, and during sleep, it's replenished at a fixed rate." Probably this describes some other repairs.
Finally, there's the unpleasant model where if you put off the repair for a long time, it becomes harder to repair, and in the worst case might become unrepairable by normal means; examples include scarring and dental plaque accumulation. I do fear that some sleep repairs are like this, that being extremely sleep-deprived for some number of days could cause permanent damage.
Even if these short-sleeper genes fail us, I strongly suspect that some dedicated effort could find out some of the bottleneck sleep repair processes (e.g. tissue repair is thought to occur during deep sleep, partly because that's when growth hormone is secreted); and either speed them up, or make them occur during other sleep phases, or even when resting but awake; and probably buy us at least 2 hours per day at no cost to health. These short-sleeper genes just seem to be the most convenient, already-prototyped way to do it.
There's a fascinating 2007 study where they administered orexin-A to sleep-deprived rhesus monkeys, and found that it masked the symptoms of sleep deprivation on a memory test.
Has there been a past success story where a drug was developed to mimic the effects of a gene and successfully improved a complicated phenomena (in this case, sleep)?
I'm unfamiliar with drug development, but my limited knowledge of genetics and sleep suggests this would be complicated. A past success story would sway my mind a little bit.
Presumably, it depends a lot on what the gene does. If it affects how your bones grow during childhood, and you're now an adult, then you're out of luck. If it produces a protein that has effects in the present, is easy to synthesize and deliver to the body where it matters, then it's straightforward. (For example, if you want to digest lactose as an adult and lack the right genes, you can buy lactase pills apparently for 10¢.)
I'd also say that the complexity of the phenomenon may be very different from the complexity and difficulty of the cure: e.g. if you had some broken enzyme that caused your body to waste half of some essential vitamin, the consequences to your body might be very complex, but the solution might be to just take a vitamin supplement. At any rate, Wiki says that there are over 6000 genetic disorders, and that over 600 are treatable. The citation on the latter describes a database of known treatments for genetic disorders, and has some interesting numbers:
Enzyme replacement therapy (ERT) is available for 28 disorders. [...] Supplements are important in the treatment of 175 disorders [...] Medications, vaccinations, and blood products play a role in the management of 328 disorders. This includes 154 medications, 3 vaccines, and 9 different blood products.
Regarding the FNSS genes. For DEC2 in particular, it's the interaction with orexin that makes the difference (the second DEC2 paper showed that an orexin receptor antagonist turned off the short-sleeper effect), and the orexin producers and receptors are all neurons in the brain. And ADRB1, NPSR1, and GRM1 all affect some kind of wakefulness-related receptor (it's the R in all their names), which I expect are also in the brain. Delivering to the brain is more delicate than to the stomach, but it's certainly doable.
My impression is that, by trying out different chemicals, you can hope to find one that binds to the receptor and either increases or decreases the activity of it (an "agonist" or "antagonist") and ideally doesn't bind to anything else; Ying-Hui said something along those lines. That's roughly the extent of my knowledge.
I have an undergraduate degree in Neuroscience and I am very skeptical that such a drug can be found. Can't talk to these specific genes in particular but genes are often turned on in different ways and in different parts of the brain, and lead to different effects based on which genes are turned on or off with them. Now because of the gene interaction in the background, the same receptor can cause a reverse effect when activated in one part of the brain or another. Additionally, each neurotransmitter has upwards of 20 different receptors in the brain. Also, every single synapse has multiple different receptors.
Drugs on the other hand hit all the receptors of a certain type in all of the brain. It's a very indiscriminant effect. The diseases that medications seem to be capable of fixing are diseases that lead to a brain-wide shortage of a neurotransmitter, which are fixed by prescribing the transmitter or a medication that causes its recycling and reuptake (like levo-dopa for parkinson's).
I've head a neuroscientist say once that treating disease with medication is like "fixing a car by pouring oil indiscriminantly under the hood," and I agree. One example for that is SSRIs for depression. Serotonin does a lot of things in the brain - it also has an important role in motor functions. SSRIs help depression, but the effect isn't very direct (people describe having some additional energy, not necessarily being 'cured') and has many, many side effects because they also affect all the other serotonin concentrations all over the brain.
Even when compared to mood control, sleep is also a very delicate, very carefully-orchestrated process in the brain. It's controlled by similar centres as breathing, heart rate, and balance. Sure there are some rare genes that contribute to less sleep but I doubt you'll find a drug that will not have severe side effects.
On a different note - I haven't read her studies in particular but the main reason I left neuroscience was because the standard practices, especially in human and animal research, seemed to me so lacking. The recruitment process which you mentioned alone can be extremely biased, and is extremely hard to control for. The sample sizes are also always quite small. I wouldn't be surprised if this group is skewed towards top 1/300th of people with this gene, and that they have numerous other genes and environmental factors that makes them more 'protected' from deprivation but also more likely to reach recruitment for this type of study.
TL;DR gene expression is tightly controlled over time and space and drugs are not. This is especially true for sleep. I am skeptic drugs will work at this fine scale since no current drugs do.
I did more research into the topic and while I started out as being skeptical became more convinced that we might be able to reduce sleep needs. I wrote it up as: Orexin and the quest for more waking hours
Maybe the most tractable new idea to improve lives by 5% or more at scale!
I recently came across this short-sleepers intervention idea on Twitter. https://x.com/byersblake/status/1865853596769849671?s=46
And when I started listing out big project ideas like economic/political systems change, reducing diseases/aging/major causes of death, resolving climate change, working on advanced market commitments, etc., I couldn’t think of anything that was as tractable that would significantly improve lives at scale. There are ways to improve democratic decision making & alignment with people’s values & priorities in big ways, but implementing the mechanisms is a big political challenge if not a non-starter. Attacking major causes of death is feasible if we find scalable solutions for aging, obesity, & addictions. We could likely increase lifespans/healthspans by 5% or more, but at this point we may get a decent amount of this when existing markets scale GLP1 drugs.
It seems like a short-sleepers intervention, even if it only reduced required sleeping hours from 8 to 6, would be equivalent to giving most people 6 more years of conscious life or a 7.5% longer healthy life if it could be done in an affordable drug/therapy. And it seems most challenges for an intervention would be narrow technical challenges & known regulatory challenges instead of unwieldy, widespread political/societal challenges.
I really like the potential here. I’ll keep reading more on it & look forward to any updates or opportunities to support it.
Looking at 160 people with the mutation that go to a sleep lab is not a good way to find out whether there are negative effects of a gene. Instead of solving that problem in a sleep lab it makes sense to look at huge genomic databases such as the UK biobank to study the effect of the genetic mutation.
For those who want to work towards developing a drug, there are various paths of advancement. For example, with DEC2, one could check (first in mice) if administering orexin (which can be done with nasal spray[4]) at the right dose and schedule has the same effect as having mutant DEC2; if so, that could be the drug.
I would expect that orexin does not cross the blood-brain barrier.
So. I'm trying to tell people about this trillion-dollar bill lying on the sidewalk. Who's going to pick it up?
The researchers at Takeda tried. They found an orexin agonist and ran it through clinical trials and the FDA gave them Breakthrough Therapy status last year:
The Food and Drug Administration (FDA) has granted Breakthrough Therapy designation to TAK-994 for the treatment of excessive daytime sleepiness in patients with narcolepsy type 1.
They ran into some trouble with their trial but there's hope:
After Takeda stopped trials of TAK-994 last year the future for treating narcolepsy with orexin-2 agonists looked bleak. But hopes were rekindled yesterday when Jazz endorsed the approach with a $50m up-front payment to Sumitomo Pharma to license the latter’s DSP-0187. The companies describe DSP-0187, which recently entered phase 1 in Japan, as a highly selective oral agent, and perhaps the hope is that selectivity will circumvent the toxicity signal that appeared to trip up TAK-994 in phase 2.
You likely won't get FDA approval for a drug that reduces healthy people's sleep needs by an hour. Focusing the drug development on narcolepsy makes a lot of sense to get the drug approved. Ideally, it will become a drug that's easy to use off-label but that will depend on a lot of factors.
If there is a gene for "needing less sleep, high behavioural drive, etc", which seems like it ought to give an evolutionary advantage, and yet only a very small fraction of the population have the gene, there must be a reason for this.
I can think of the following possibilities:
Do you have any idea which of these is the case?
The gene also has some kind of selective disadvantage. (If we are lucky, the disadvantage will be something like "increased nutritional requirements" which is not a big problem in the present day.)
From looking at Wikipedia's description of orexin it does brown fat activation which is essentially burning calories for nothing but warmth.
One of the drugs which is orexin agonist that is currently in clinical trials was earlier used as a weightloss drug.
It looks like "increased nutritional requirements" is part of the selective disadvantage.
I think a version of the second and third possibilities is probably true.
For DEC2 and GRM1, the mutation decreases the ability of the gene to block wakefulness (DEC2) or promote tiredness (GRM1), and loss-of-function mutations seem easier to create. And indeed, I think one of the papers mentioned that there are other mutations in the DEC2 gene that correlate with reduced sleep, and GRM1 already has two identified short-sleeper mutations. Therefore, I expect that similar mutations have appeared in the past, but didn't catch on.
Why would evolution have rejected them? I suspect that, as you say, 10,000+ years ago (perhaps even 1000 years ago), being awake an extra two hours before dawn brought little benefit and basically just meant that you wasted calories; the tradeoff only became worthwhile very recently, in evolutionary terms. Perhaps there were also effects like "people had a higher parasite load, and/or were constantly scraping their skin or stepping on sharp objects, and thus had a higher constant need for some type of repair that the mutations don't accelerate".
Great post. You should submit it for a cause exploration prize for a small chance at Open Phil being convinced by this!
Why has this depended on NIH? Why aren't some for-profit companies eager to pursue this?
I think both Ying-Hui and I had the impression that the research had to be somewhat further along before any profit-minded people would fund it. But someone recently explained to me that Silicon Valley companies are often funded with much less scientific backing than this, so this week I've written to one venture capitalist I know, and will probably contact others. Regarding getting funding from an existing company, I don't know much about that option.
Advice is appreciated.
What disease would you seek FDA approval for? "I sleep more than 4 hours a day" is not a recognized disease under the status quo. (There is the catch-all of 'hypersomnia', but things like sleep apnea or neurodegenerative disorders or damage to clock-keeping neurons would not plausibly be treated by some sort of knockout-mimicking drug.)
Could be marketed as a medication for hypersomnia, narcolepsy, chronic fatigue, and ADHD.
Great post, I really appreciate an in-depth review of research on reducing sleep need.
I wrote some arguments for why reducing sleep is important here:
https://harsimony.wordpress.com/2021/02/05/why-sleep/
I also submitted a cause exploration app:
https://harsimony.wordpress.com/2022/07/14/cause-exploration-prize-application/
Your post includes substantially more research than mine and I would encourage you to reformat it and submit it to the OpenPhil's Cause Exploration Prize. I'm happy to help you with edits or combine our efforts!
Thanks! It's great to see that other people have discovered FNSS research and had similar thoughts about the implications. I have indeed submitted the post for the Cause Exploration Prize. At this point I feel reluctant to make significant changes to the post, but if there are major/glaring issues, those could be worth pointing out.
There is a high behavioral drive among those with FNSS, and individuals report a need to always be mentally active resulting commonly in high profile, high pressure jobs or holding multiple jobs.
Well this definitely resonated with me - I wonder how true this is for EAs in general vs the general population!
Amazing writeup! Really interesting and well-written.
I decided to crosspost it to Manifold Markets and created a prediction market. You state that you estimate a 50% chance short sleeper genes work as advertised. Curious to see what other people think!
If you have any suggestions on how the resolution criteria or anything else can be more well-defined let me know and I can make adjustments to the market.
We are currently having growth in genome sequencing as it gets cheaper. Interesting questions are whether or not any of the mutations in this post lead to statistically reduced lifespan once someone has a sample of people who died whose genes have been sequenced, where at least X had this mutation.
Thanks tou for writing this. This is an incredibly good idea for a cost-effective cause in my opinion. I need about 9 1/2 hours of sleep every night and it is possibly the single greatest obstacle to achievement I face. I think for long-sleeper people like myself, further investigation could have an especially high upside if successful.
Thanks for posting this! I think this topic is extremely neglected and the lack of side effects among natural short-sleepers strongly suggests that there could be interventions with no obvious downsides.
My main concern with your drug-centered approach is: what if the causal path from short-sleeper genes to a short-sleeper phenotype flows through nerodevelopmental pathways, such that once neural structures are locked-in in adulthood it's not possible to induce the desired phenotype by mimicking the direct effects of the genes? If this is true, then reaping the benefits of short-sleeper genes would seem to require genetic engineering (I doubt embryo selection would scale given the low frequency of the target alleles). This would obviously be politically problematic and I'm not sure it'd be technically feasible right away (last time I checked, CRISPR people were worried about off-target mutations, but I'm not up to date with that literature so this may not be an issue anymore).
Such a drug would be enormously valuable. If it worked, presumably it would give people at least one extra waking hour per day, and maybe the "higher resilience and productivity" side effects would be real too.
This would probably increase the rate of technological advancement, including the advancement of AI capabilities. I worry that increasing AI capabilities will shorten AGI timelines.
This sounds like a great stuff for a startup idea/(effective) entrepreneur ship. But It's probably not a good idea for a cause area. Public perception would be a problem.
Can you expand on this? Is it that anything to do with genes is controversial? Maybe also the possibility that success in this could increase rich people's societal advantages over poor people even more? (I listened to the post yesterday and might've forgotten some key points)
tl;dr I work in genetics, was initially very sceptical that were any genetic variants that really did cause shorter sleep, did some research which made me slightly more sceptical still, and then found a recent study which I think serves as an additional very, very strong update that the variants mentioned in this post have no such effect.
Long version: I have worked for 5 years at a company dedicated to finding associations between genetic variants and disease. I have a bit-part technician's role in our primary workflow for discovering these associations, and I have done some work alongside our clinical sequencing group (a bunch of physician-scientists contracted by clinicians around the world to try to identify genetic causes for their patient's conditions). But I will freely admit that most of the work I do is deeply non-clinical stuff like analysing DNA from the skulls of 9th-century settlers of Iceland to find out where they were born. The upshot is that I'm claiming some expertise here, but not on the level of "I literally did my thesis on this" or anything.
I came across this cause area via a tweet and was initially deeply sceptical. It really smacked of the noughties-era candidate gene studies which turned out to have very high false positive rates, as Scott Alexander described in withering detail. And sure enough, it transpires that the DEC2 P384R mutation was indeed identified through a candidate gene approach back in 2009, during the last hoorah of the candidate gene era, just as GWAS studies were beginning to take over.
I'll try to briefly convey the fundamental scientific principle behind my scepticism. Most genetic variants that you find in human populations are extremely rare, but we also share DNA with our relatives, which means that there are a huge number of "mutations" which are quite common in any one family but rare outside that family. Even if you circle all the people in one family who share some trait, like short sleep duration, and look at what "rare" variants they share, depending on the frequency cut-off you use to define "rare", that is still potentially hundreds of thousands variants across the genome. Even if you assume that the trait in question is largely caused genetically, how on Earth do you narrow this down to a single causative variant? Well, there are many populous categories of variation that you can pretty safely assume don't have any physiological impact, and if you're feeling brave you can even decide to just look at a subset of genes that you bet are more likely to affect your trait. But even after those extra assumptions, you still have many, many candidate variants. At least, you should have many candidate variants if you have done things correctly - however, in the past, researchers often bee-lined to a small number of genes that they assumed ex ante were related to the trait and only looked at variants there, based on an even less complete understanding of biological pathways than we have now, and with smaller sample sizes, and less accurate sequencing methods.
To be clear, some early attempts to link traits to genetic variants were successful, especially when you were guided by existing knowledge of biochemistry or if previous attempts had already narrowed down the search region a lot. For instance, the gene whose variants are responsible for ~all cases of achondroplasia, a kind of dwarfism, was identified in 1994 after previous work had already narrowed the search down to a small genomic region shared by all affected members within and across 18 affected families. But according to the second link, at least 5 other genes had previously been suggested as causing achondroplasia and failed to pass scrutiny.
"Carefully examining 18 families" indicates the kind of analysis needed to perform a successful linkage analysis plus candidate gene study. I read the papers in which the short-sleep variants were presented and noted that they were all nominated based on very small numbers of alleged carriers.[1]
I also wanted to see if there were any independent evidence in favour of the idea short-sleep variants. I did about 30 mins of research into this. First, I looked for any associations between variants in DEC2 (aka BHLHE41) with any trait in the GWAS Catalog, which compiles results from many genetic association studies. These studies generally look at more common variants with generally small effects, but it's an okay assumption that a gene carrying variants with large effects on a trait will also carry variants that have small effects on that trait. For instance, the gene with large-effect variants responsible for achondroplasia also has other variants associating with smaller changes in height (which may or may not be independent of each other). For DEC2, however, I saw no plausible sleep-related hits for any variant in or around this gene. And it was the same story for for the other three genes you mentioned, ADRB1, NPSR1, and GRM1. Conversely, other genes that did seem to be fairly robustly associated with sleep duration across studies include PAX8 and MEIS1. Funnily enough, the latter gene was even more strongly associated with restless leg syndrome, which may well represent the mechanism by which variants in this gene affect sleep! I found the results of my GWAS Catalog research to be a reasonably strong update against the idea of these variants being actually associated with short sleep. I also looked at a crowdsourced database called ClinVar of potentially clinically important variants, which are generally large-effect and pretty rare, and found that DEC2/BHLHE41 variants had indeed been nominated as causing short sleep - probably based on Fu's research? - but none seemed to have support from multiple submitters.
I was tempted to put a pin in this and wait until I got to work on Monday to check for associations myself between variants in these genes and sleep traits in the most cutting-edge datasets available. The cohorts I have access to are large enough that I would probably find quite a few carriers of the variants in question. However, I then came across this study published in September 2022, The impact of Mendelian sleep and circadian genetic variants in a population setting. I will let the paper speak for itself:
They also perform a "burden" test in which they gather up all the people that seem to have any high-impact mutation in a particular gene and see if those carriers as a group are different to everyone else. Again, they find no evidence that big changes to the genes carrying alleged "short-sleep" variants actually associate with changes in sleep length. Note that their tests seem to have some specificity, in that they don't reject every one of the claims of association analysed: they also examined variants that were previously claimed to associate with a different sleep-related trait - going to bed earlier or later than other people - and do find some degree of support for a few of those genes/variants.
To be clear, this paper really represents the test you'd want to do to settle once and for all whether these supposed short-sleep variants affect sleep. They usually find dozens of other carriers for these variants, and the large posited effect sizes of ~1 full standard deviation or more, which should be perfectly detectable, just don't appear. This study seems to be very strong evidence that the variants mentioned in the above post are not actually associated with shorter sleep. I suppose the negative evidence is a bit less strong for a few variants that were found only rarely in these comparison datasets, but the burden tests and my GWAS Catalog search still point towards those genes in toto not having a strong role. My steelman case for the opposing side would involve some technical arguments. For instance, maybe the variants like DEC2 P384R actually tag some haplotype in the original families which also carries some other marker which really is causative, but again the burden tests and GWAS Catalog points against this. I also suspect that the burden test might be underpowered, but I can't evaluate this easily. I think the NPSR1 variant is the most likely to be genuinely associated with sleep. It's so rare that it wasn't found in the 2022 study and so couldn't be analysed, and the gene is reported to have a few weak associations with some behavioural traits in GWAS Catalog, but I still think it's very likely to not replicate independently.
I will end quite forcefully and say that anybody who does a clear-eyed assessment of the current state of evidence should realise that it is very unlikely that any of the variants mentioned in the OP genuinely cause much shorter sleep in humans, and extremely unlikely that more than one of them does. Fu should no longer promote the idea unless they find some new, overwhelmingly strong supporting evidence.
DEC2: 2 affected and 5 non-affected in one family.
NPSR1: 2 affected and 1 non-affected in one family, additional knowledge that the candidate variant is very rare outside the family.
ADRB1: 5 carriers and 3 non-affected in one family analysed by linkage analysis, 4 of those had part of their genome sequenced. An individual who "should" have carried the variant according to linkage analysis actually slept a normal amount.
GRM1: 2 different variants in 2 different families, one with 3 affected and 4 non-affected, the other with 1 (arguably 2?) affected and 1 non-affected. 4 affected across the families had part of their genome sequenced.
You are really not narrowing down the massive genomic search very much when you look at such small numbers of people. Some of these variants aren't even that rare in the general population - 1 in 1000 people or so being carriers. At frequencies and posited effect sizes this high, the finding would be pretty easy to replicate elsewhere, and as far as I can see nobody has.
I'll use my time on the soapbox to make an ancillary point: that EAs and especially rationalists are on the whole quite naive about genetics, and in particular are too enthusiastic about the use of genetics knowledge to facilitate biological enhancements in some way or other. To be clear, I'm actually more bullish in principle than many of my colleagues regarding more widespread use of, say, polygenic scores - for reference, I'm probably a bit less positive on the idea than, say, Shai Carmi. But rationalists and some EAs seem to think that genetics is both a much more important determinant of traits than is actually true, and also that modifying genetics or applying genetics knowledge is much more tractable than it really is. Like, it's not reflexive Luddism that's preventing you from being able to improve your offspring's IQ by +10 over baseline, it's primarily the limits of our abilities at present (and even what's feasible in ten years) to link genetic changes to traits, or to accurately perform embryo selection and gene editing.
I realize I'm about 4 months late to this conversation so you're likely the only one who will ever read this. But enhancing IQ or other traits is actually quite feasible with our current technology. Shai Carmi's paper "Screening Human Embryos for Polygenic Traits Has Limited Utility" used predictors that were already two years out of date by the time he published and are even more out of date now. You could raise IQ by 2-8 points in expectation today, with the exact amount depending most on the age of the mother, number of egg retrievals completed, genetic ancestry of the parents, and quality of the IVF clinic.
The effect size for type 2 diabetes and height is even larger because our genetic predictors for those traits are better.
The main thing preventing us from getting to +10 IQ points right now is lack of data. We have all the genomes we could ever need. But no one has phenotypes them; we don't have IQ test scores, or some highly correlated proxy like SAT score.
I wrote a whole post about this on LessWrong if you want to learn more
I quite appreciate this comment, thank you!
This is a really comprehensive comment, and I'm glad my tweet led to some really useful digging :)
Thank you very much for the high-effort post and the significant findings. I agree this seems to be strong evidence that the DEC2, ADRB1, and GRM1 mutations don't work as expected in the general population. That is disappointing. A couple of questions:
How do we explain this? My brain generates "ADRB1 (or something correlated with it) works, but it requires extra mutations to work (or, similarly, there are some mutations common in the population that stop it from working); Paul married someone with concordant mutations, Brad and Janice married people with discordant mutations". Does that seem likely, or do you have other explanations in mind?