MORNINGS STARTS AT 3 A.M. for Shawn Radcliffe. Without setting an alarm, he wakes up like clockwork in the pre-dawn hours, does some writing and then some yoga, sometimes followed by a run. “I really like the mornings,” Radcliffe says. “I’m alone and it is a beautiful time.” He goes to bed between 9 and 9:30, which means he rarely sleeps more than six hours a night. But at 48 his health is good, he feels happy and well rested, and his unconventional sleep cycle doesn’t seem to have any ill effects.

Radcliffe’s relationship with sleep hasn’t always been so harmonious. In college he stayed up late to study, drank tea and soda to keep going and suffered from insomnia and fitful, fragmented sleep. He finally solved those problems when he started practicing yoga and meditation, limited his caffeine intake and learned how to switch off his mind at night. But he also had to accept that sleeping at an unusual time and for fewer hours than most people do was part of his internal machinery, and it wasn’t going to change.

Evidence is growing that genetic makeup has much to do with when, how long and how well a person sleeps, and also plays a role in some types of insomnia—the inability to fall asleep or stay asleep. Just this past October, sleep researchers established that yet another gene—the third discovered so far—appears to share responsibility for the kind of “short sleep” that Radcliffe currently experiences, which appears to carry no negative health consequences. Insomnia, in contrast, does impair health and well-being, and it affects one-third to half of the U.S. population, making it the most common sleep disorder.

“Insomnia can harm people mentally and physically,” says Philip Gehrman, a psychologist who works in the department of psychiatry at the University of Pennsylvania and whose current research involves the genetic roots of insomnia. “It is tied to depression and anxiety and can depress the immune system,” he says. And current therapies leave much to be desired. Sleeping pills, the most common treatment, don’t work for everyone, may lose their effectiveness over time and can lead to daytime drowsiness, attention deficits, memory loss and even dangerous episodes of sleepwalking. A second approach, cognitive behavioral therapy, involves trying to change attitudes about sleeplessness with the goal of reducing sleep anxiety. This treatment, which also involves establishing rigid sleep routines, isn’t easily accessible to most people and can be challenging to do without support.

“If a treatment for insomnia means you have to wake up every day at the same time, flip on a light, sit in front of that light for half an hour”—a typical cognitive behavioral treatment—“you’re asking an awful lot from a person,” says Jacqueline Lane, a geneticist at Massachusetts General Hospital. “These treatments can be really difficult to maintain.”

Better alternatives are needed, and getting there means delving deeper into the genetic underpinnings of sleep. Yet that, too, poses challenges. “Every major neurotransmitter is involved in regulating sleep, and sleep is the product of so many different biological systems that there are bound to be a large number of genes that influence it,” Gehrman says. “It makes for a very complicated situation.”

Moreover, environmental and social factors can also influence when humans sleep and for how long. “There is a biological tendency for us to go to sleep and wake up, but we often override that, because we have to stay up to meet a deadline, to study for an exam or to spend time with our families,” says neurologist Louis Ptáček of the University of California, San Francisco, co-senior author of the October 2019 Science Translational Medicine study that identified the third gene implicated in short sleep.

Such factors can make untangling the genetic complexities of sleep even more difficult—and during the past 35 years, as modern genetic tools have revolutionized research into many neurological disorders, discoveries about sleep have lagged behind. That’s beginning to change, largely thanks to a giant, recently established database that pairs genetic information about people with data about their sleeping habits. The resource has helped researchers probe the core components of sleep—sleep-wake times, sleep length, sleep quality—and is slowly helping them parse the intricacies of insomnia.


Art by Flashpop/Getty Images

EVERY MINUTE, THOUSANDS OF tiny biological “clocks” are at work in the body. Found in nearly every tissue and organ, the clocks control the daily rise and fall of body temperature, metabolism, the release of hormones—and sleep. They’re coordinated by a single master clock, a cluster of 20,000 neurons in the brain’s hypothalamus called the suprachiasmatic nucleus, or SCN. The SCN uses signals from the eyes, cued to the waxing and waning of daylight, to “tell the time” and synchronize the internal army of clocks. The system is responsible for circadian rhythms, plus it controls the production of melatonin, a hormone that helps humans sleep. When morning light hits the optic nerves, this master clock reduces the output of melatonin, and as daylight fades, it produces more.

The other major process affecting sleep is sleep homeostasis, an internal biochemical system. This “sleep drive” is low after a good night’s sleep, but as the day progresses, the chemicals behind the drive build until—normally late at night—the urge to sleep becomes overpowering. In the 1980s, scientist Alexander Borbély from the University of Zurich in Switzerland proposed a two-process model that married the two systems. Circadian time oscillates like a wave in humans, making us sleepy at night and wakeful in the morning. On a separate track, the longer people are awake, the more they feel their “sleep drive.”

This accounts for the fact that people who miss a night’s sleep may fall asleep during the day, even though the body’s clock is saying “be awake.” While Borbély’s model is highly simplified—it doesn’t take into account the nature of stress, arousal or other factors—researchers have found it a useful framework for investigating the genes behind sleep, which may play into two or more interlocking mechanisms.

Forays into identifying individual genes associated with the timing of sleep started in the 1970s. By 1984, scientists Jeffrey Hall and Michael Rosbash at Brandeis University in Waltham, Massachusetts, and Michael Young at the Rockefeller University in New York City had found the period gene in fruit flies. Mutations of this gene could shorten or lengthen the flies’ daily cycles of activity. These scientists later discovered additional genes affecting circadian rhythms and eventually were able to explain the key workings of the biological clock—an achievement that led to a Nobel Prize in Physiology or Medicine in 2017.

AS HELPFUL AS THESE discoveries have been, translating molecular findings about fruit flies into a meaningful understanding of human sleep has proved difficult. One tool for doing that is the genome-wide association study, or GWAS, which lets researchers compare the sequenced genetic data from large numbers of people—some who have a particular health problem and some who don’t—and look for potentially significant differences. By 2015, however, only two GWAS’s had been conducted on insomnia, and neither gleaned significant insights.

One likely reason for the inconclusiveness of those studies was the relatively small number of participants. One looked at 10,038 people in Korea, and the other considered 2,323 Australian twins. “To do this type of analysis you need very large numbers—often more than a single research group can come up with,” says Allan Pack, founding director of the Center for Sleep and Circadian Neurobiology at the University of Pennsylvania Perelman School of Medicine.

It turns out that one resource—the UK Biobank, created in 2006—had those large numbers. The database boasts detailed health information, including genetic data and answers to a questionnaire about sleep habits, for 500,000 people. In addition, one in five of the UK Biobank participants spent a week wearing a monitor that recorded information about activity, rest times and sleep patterns.

Recently an international group of researchers, led by the University of Exeter and Massachusetts General Hospital, analyzed the genetic data of 453,379 people from the UK Biobank who had responded to the question, “Do you have trouble falling asleep at night or do you wake up in the middle of the night?” Nearly a third had answered “usually,” which researchers took as an indication that they suffered from symptoms of insomnia.

The scientists also looked at activity tracker data, which told them how efficiently the research subjects slept and whether they were awake in the middle of the night. Using a GWAS approach, they identified 57 regions of the genome, containing 236 genes, that were associated with insomnia symptoms. That marked a significant advance on previous studies, which had found only seven areas of the genome linked to insomnia. The team published its findings in Nature Genetics in March 2019.

These early results were encouraging—all of the identified regions could be possible therapeutic targets—and the team continued to look into related issues, including how many hours a study participant habitually slept. They found 78 associated gene regions, some of which overlapped with the areas implicated in insomnia. Sleep quality was the next target, which they explored using activity tracker data showing when people went to bed and when they finally got to sleep, how long they slept and how broken up or fitful their sleep was. Inspecting data from the UK Biobank as well as information from three additional studies, the team ultimately associated sleep quality with 47 gene regions.

Of particular interest was an uncommon variant of one gene, PDE11A, which appeared to affect both sleep quality and sleep length. “Occasionally you find one thing that looks very clear in a study and it’s like, ‘Yes. This is the spot. This is it,’” MGH geneticist Lane says. Previous studies had suggested that this gene could be a good target for treating neuropsychiatric disorders such as depression or anxiety, so its connection with sleep was illuminating.

Apart from these specific insights, the researchers also came away with a more general sense of how genes affect sleep. Many of the gene regions identified by the activity tracker data, for instance, were linked to the production of serotonin, a neurotransmitter known to play a role in the sleep cycle, bolstering the case that it may promote deeper sleep and improved sleep quality. They were also able to pinpoint what insomnia wasn’t. Previous researchers had mapped regions associated with chronotype—someone’s propensity to go to sleep earlier or later than most people do. The newly identified gene regions for insomnia and poor sleep quality had little overlap with those.


Art by Jose Luis Pelaez/Getty Images

LANE BELIEVES THIS NEW evidence, taken together, suggests that there are important distinctions to be drawn between unconventional sleep cycles and true insomnia. Some people seem able to sleep less or more than average and may have earlier or later wake-up times, with no ill effects whatsoever. “But if someone takes a long time to fall asleep, and wakes up five times during the night, then it seems very clear that there are health consequences,” Lane says. “And these findings indicate that insomnia is a true disorder.”

Perhaps most illuminating is that researchers also found significant overlap between the genes implicated in insomnia and those related to depression, anxiety and other psychiatric conditions. “Now the question is whether some sleep traits are closer to psychiatric traits than they are to other sleep traits,” Lane says. “When we talk about sleep and psychiatric traits, are we talking about different things? We think of them as very separate beasts, but they are not all that separate.”

Such findings indicate that beneficial treatments for insomnia might focus on reducing anxiety and treating mood disorders. They could explain why cognitive behavioral approaches—which can focus not only on sleep habits but also on a person’s thoughts and emotions involving sleep and sleeplessness—are often successful.

It is more than likely that there are different types of insomnia, Lane says, driven by discrete genetic forces. For example, one gene associated with insomnia is also involved in restless legs syndrome, a sleep disorder characterized by an uncomfortable, irresistible urge to move the legs. A genetic approach could try to identify different forms of the disorder so that physicians can offer more personalized diagnoses. “It will help to subdivide people who might be better candidates for one kind of treatment versus another,” Lane says.

But tailoring treatment is still a long way off. The recent GWAS’s are a starting point, and the next step will be to pinpoint the active genes and understand how they influence sleep. “If you find the causative gene, then that takes you into exploring new pathways, and it opens up an entirely new area of biology,” says Pack from the University of Pennsylvania.

Pack has established a pipeline with other researchers to follow up on the GWAS he conducts. Starting with a gene region, one of his collaborators will home in on a gene of particular interest. Then Alex Keene, associate professor of biological sciences at Florida Atlantic University in Boca Raton, tests the effects of variants of that gene in animal models.

Keene suggests that this methodology may be particularly helpful for a complex disorder like insomnia that’s likely to involve many genes, each of which makes only a small contribution to the final trait. “It is not going to be one gene or one genetic mechanism that explains why you or I might sleep less than the average person,” Keene says. It is only in working through the possibilities, one by one, that researchers can slowly build up a genetic picture of how the disorder occurs—and eventually arrive at an age of new therapeutic targets, which may bring a better night’s sleep for everyone.