Some people, no matter what they do, simply cannot fall asleep until the wee hours—and do not feel rested unless they get up much later than most of us. These night owls may have a common form of insomnia called delayed sleep phase disorder (DSPD), which studies have suggested is at least partly heritable. Now researchers at the Rockefeller University and their colleagues have uncovered a genetic mutation that could elucidate what causes these often awkward sleep schedules.

Of course, DSPD is not a problem for everyone who has it: if you work as a bartender or a musician, you might never seek a diagnosis or treatment, says lead study author Alina Patke, a sleep researcher at Rockefeller, who self-identifies as a night owl but does not have the mutation. Yet for others, especially college students or office workers, the condition can be torture. The new study centered on a 46-year-old female subject with lifelong sleep problems. “Typically she would go to bed at 2 or 3 A.M., sometimes as late as 5 or 6,” Patke says.

The woman lived under observation for 14 days in a room with no clocks or windows. Not only did she produce the sleep-inducing hormone melatonin five to seven hours later than a typical person in similar previous studies, but her sleep was also oddly fragmented, sometimes coming in short naps. When the team analyzed her DNA, they found a mutation in a gene called Cry1 that also showed up in her family members who reported sleep problems. This gene encodes a protein that is known to suppress the action of the core circadian clock proteins CLOCK and BMAL, which activate a wide variety of genes—including some related to wakefulness—during the day. The mutation caused the deletion of a portion of the CRY1 protein's tail, making it even more effective at suppressing CLOCK and BMAL. The team sifted through a genetic database and found 39 other people with the mutation. Most of them also had relatively late bedtimes and wake-up hours.

Daniel Kripke, a psychiatrist who has studied sleep and a professor emeritus at the University of California, San Diego, who was not involved in the work, points out that studies that scan large groups of people for links between a particular trait and a genetic variant have found no connection between this mutation and DSPD. Still, he says, the new paper presents convincing evidence that it could be behind some cases of the disorder.