The result didn’t make sense. The researcher kept scanning the mouse’s pregnant belly, back and forth, back and forth. They could see the embryos on the ultrasound screen, cocoon-like, in grayscale — but there were no heartbeats. It should’ve been visible by now, at eight and a half days, a rhythmic flashing, a clue that the organ was on its way. Instead, each outline was utterly still, dead before it had come to life.

That couldn’t be right. These mice carried the exact same genetic mutation as a family who lived about an hour away, in the county outside Pittsburgh, near the West Virginia line. The humans definitely had beating hearts. Nicole Burns worked at an assisted living facility, hoisting residents in and out of bed, guiding them as they shuffled to dinner with a firm hand on the back. Her daughter Peyton loved hunting. Her son Cameron loved playing basketball. Their genome couldn’t contain something so lethal: They were all very much alive.

The researchers tried again, inserting the family’s genetic variant into the DNA of a mouse egg, injecting that into the coils of a mother’s oviduct. The same thing kept happening, sound waves bouncing back and revealing embryos dead in the womb.

“This is basically not a survivable mutation,” said Cecilia Lo, chair of developmental biology at the University of Pittsburgh, who led the research. “If you have it, you’re pretty much dead.”

So how had Burns and her relatives survived? Lo wondered if, against all odds, they’d inherited not just one but two ultra-rare mutations, the second shielding them from the deadly effects of the first. It seemed stunningly improbable — but that is, in fact, what she, postdoctoral researcher Xinxiu Xu, and their colleagues showed in a paper published Tuesday in Cell Reports Medicine.

Out of 589,306 people, 13 were medically unbelievable. But the data were anonymized; there was no way for them to go back to those patients and find the protective variants they suspected were lurking in their cells.

Now, almost by accident, Lo found herself analyzing eight genomes that seemed to tell the story Friend had been speculating about. Her team dove back into the code, searching for the mutation that might be responsible for the rescue. There were a handful of other variants shared by all eight relatives — but only one of those genes was called into action in the heart. Bingo. Sure enough, when the researchers bred new genetically modified mice, this time with both variants, the embryos survived. Some also showed the same heart defect as Burns and her kids. In other words, imperfect as it was, the second variant was protective.

It felt like a scientific triumph. Often, the narrative arc in biomedicine is a quest for disease-causing genes. Understand harmful alterations, the thinking goes, and you can hopefully reverse them — and there are reams and reams of papers about these “bad genes.” There are “good gene” storylines, too. In 1996, a genetic variant was described that could stymie HIV — an observation that became a class of drugs. But Lo’s paper was genre-bending, in a way. The “good gene” they’d pinpointed wasn’t safeguarding against something extraneous, like a virus; it was softening the damage of another gene. “That’s the untold story of human genetics,” she said.

To Friend, the paper could provide a road map to look for other, similar instances. But it was also a technical feat. “It’s almost like, ‘Here’s what you could do today that you couldn’t do 10 years ago.’ It’s like, whoa! Just beautiful,” he said.

But that wasn’t exactly how Burns felt about what was going on inside her genome.

Nicole Burns lives in her great-grandfather’s house, in Amity, Pa., about an hour south of Pittsburgh. He’d had over 100 acres of rolling Appalachian foothills, with sheep, horses, cows, pigs, though some of the land has since been sold off. What’s left from the farm is mostly scrap, a rusted plough, an old wheel sitting at the edge of the yard where it fades weedily into woods. Stuff her grandfather wouldn’t get rid of, for sentimental reasons, and that now she won’t get rid of, in memory of her grandfather.

The land wasn’t the only thing he’d passed down. Her other, less appealing birthright was a hole in the heart. For her, actually, it was several, the biggest the size of a quarter. Normally, blood would pass from the heart’s right side to the lungs, from the lungs to the left side, and from the left side to the rest of the body. But these openings allowed some to burble backward, oxygen-rich blood leaking from left to right, to get resent to the lung. Sometimes, those holes close up on their own. If they don’t, the inefficiency can slowly, over decades, stretch out the tissue, changing the heart’s shape, potentially shifting its rhythm. In some cases, a doctor can thread a patch in through an artery or vein, but that doesn’t always work.

In her family, heart surgery is a kind of ritual, a rite of passage. Her great-grandfather was the exception, diagnosed too late to operate, and he died of a massive heart attack. After that, everyone ended up under the knife. Her grandfather, a metallurgist at a steel mill, got surgery as an adult. Her mom, aunt, and uncle all got it around their teens. She and her sister got it, both her kids had to have it, and it seemed like if they were to have children themselves, they’d be destined to need heart surgery, too.

But it only came to a geneticist’s attention because of Burns’ son Cameron. He was born six and a half weeks early, by C-section. “I knew something was wrong, because they wheeled me down in the bed so I would be able to see him,” she said. He was already hooked up to too many machines for her to hold him. A minute later, they rushed him to the children’s hospital.

She discharged herself two days later. Her doctors would have liked her to stay — she’d just had surgery, and could hardly walk upright, even clutching her belly — but Cameron’s team kept calling her in her hospital room, asking permission for procedure after procedure, and she felt like she had to be there. His body kept filling up with fluid. This wasn’t because of a hole. His heart just wasn’t pumping right.

The cardiologist said she had a choice to make. They could take him off all the machines and medications — nothing seemed to be working anyway — and see if he could pull through. Or they could try yet another tack, inserting a drain tube, to see if they could siphon away some of his swelling. She chose the drain. “It was selfish on my part, because I was thinking, you know, any time I can get with him — God forbid, before something happens — that’s what I want,” she said.

It worked. But during those discussions, the cardiologist also heard something that intrigued him. Often, inheritance patterns of congenital heart defects were murkier, weirder, harder to predict. “Like, not everybody in the family has it. That’s what struck me,” said Brian Feingold, director of the pediatric heart failure center at University of Pittsburgh Medical Center. “That’s why I said to Dr. Lo, ‘You have the expertise to kind of help unravel what’s going on here, because there’s something different than what we typically see going on in this family.’”

The researchers drove out on a snowy day in 2011. Four generations of the family were there: Eight biological relatives who had holes in their hearts and three relatives by marriage who didn’t. The kitchen became a makeshift lab for swabbing noses and collecting blood. Burns’ grandmother had bought pop and a tray of sandwiches. The snow was picking up. Once they had their samples, the scientists didn’t linger. After they left, the house quieted. Burns was glad that part was done. Now, she thought, maybe they’ll be able to start figuring out some of these mysteries. “I was like, ‘I don’t know what we all did in a past life. But we got cursed with something,” she said. “I’m going to need somebody to take it back.”

For Friend and his colleagues, the idea of looking for mutations that suppress bad counterparts began with lab animals. Experiments in the classic creatures of genetics — yeast and roundworms, for instance — showed how much interplay there was between different parts of DNA. A complex choreography of many genes is often the backstory to a particular trait.

That gave them an idea. So much of the search for drugs involved looking for disease-causing variants. Repairing the damage is often tough — and Friend wondered if there might be cases where toying with a different gene could make it easier to stop some of the first one’s troublemaking. Imagine a broken machine on an assembly line, twisting a piece unusably out of shape. Getting that device back to normal is an intricate fix. But if there were another, earlier point on the conveyor belt where you could reroute around that station, maybe you could save some of the damage and keep things rolling.

That’s part of the appeal. But Eric Minikel isn’t totally convinced. He’s a grad student at the Broad Institute, working with his wife, Sonia Vallabh, to search for therapies for the fatal mutation she carries. People often ask if he’s looking for this kind of protective variant. He isn’t. He’s excited when that sort of result is found, but he doesn’t foresee a huge wave of them emerging and transforming medicine. “There are many diseases (including prion disease) where the original disease gene has been known for decades and yet no one has successfully drugged it, so the problem isn’t a lack of genetically validated targets, it’s a lack of drug development,” he wrote in an email.

Another challenge is finding the people who have unidentified protective variants. “They’re not going to be coming into clinics, because they’re not sick,” explained Friend. “They’re not under the microscope, so to speak.”

Even if they are sick, they’re exceedingly rare. As Benoit Bruneau, director of the Gladstone Institute of Cardiovascular Disease, pointed out about Burns’ case, “This is one family in the entire history of studying congenital heart defects that have this.”

The rarity itself can create a methodological problem. It can be near impossible to test a hypothesis with an n of one, or three, or five. As Minikel explained, “The fundamental problem is that the dataset would be wide (there are billions of different places to look in the genome) but shallow (the disease is rare).” He went on: “So if you look at a person who has supposedly-lethal mutation X but is mysteriously alive, there’s no quick way to know which, if any, of their rare genetic variants kept them alive.”

In the case of Burns’ family, it seemed possible, but slow. At the time, sequencing whole genomes was prohibitively expensive. The team could read only short snippets of DNA and then try to piece together the part of the code that contained instructions for making proteins — a technique in which tiny alterations might be missed.

They were looking for a mutation shared by the eight affected relatives that might explain their heart issues. They found nothing. Years passed. Burns wondered whatever happened to the blood and saliva she’d given away. Her doctors had no news when she asked. Her mom died. Then her aunt. Both had heart attacks.

But, by five years ago or so, sequencing methods had become fast, accurate, and cheap enough for the team to go back into the freezer and get out the family’s vials. When they found mutations, they had CRISPR gene-editing tools, allowing them to easily and quickly insert the alterations into mice. When they took bits from a family member’s samples and chemically coaxed them into heart muscle cells, they saw only a mild structural defect, and the cells could still beat, could still help generate a pulse.

There were limitations. When they spliced the first, lethal variant into African clawed frogs, another common lab animal, the embryos had a heartbeat that was reduced but not eliminated. And there were questions the study didn’t answer. How exactly this pair of mutations was giving rise to the strange set of symptoms in Burns’ family remains unclear. The first gene was already known to play an important role in the growth of heart tissue. But why would this combo so insistently give rise to heart holes?

To Loydie Jerome-Majewska, a developmental biologist at McGill University, that unanswered question meant the possibility for some sort of treatment was still far off. “Before you can think of therapy, you need to know how things are happening,” she said.

For now, the idea retains an aura of what-if, a garden of forking paths that is the germ of all research, at once captivating and dismaying. A few weeks ago, Lo called Burns to share what she’d found in the family’s genomes. “I told them, ‘You know, you’re actually very blessed,’” Lo recalled. “‘Think about it. None of you would have been born if you didn’t have the second mutation.’”

When she was about 10 years old, Burns’ mom took her along to bring-your-kid-to-work-day. Work was a skilled nursing facility. That was when she fell in love with the job, decided that she wanted to do the same thing — a different kind of inheritance, of sorts. She found the scrubs enchanting, the way they brought you into the world of doctors and nurses. Seeing them, as a kid, she thought: That’s somebody of importance. The residents were so happy to see her, and showered her with M&Ms and Hershey’s. When she started work at 18, she was taken aback at how difficult those lost to dementia could be, how aggressive. She asked her mom how come there was no lockdown unit where they’d spent the day when she was a kid. “She was like, ‘Oh there is. I just didn’t take you there.’ I had to figure that out on my own.”

She laughs about it. But when it came to her family history and her own kids, she didn’t want to pass down any surprises. So, a few years ago, when she was watching a movie with her daughter, she decided they were going to have The Talk. She was sitting in her grandfather’s old chair — fluffy and burgundy, wide enough to have fit him and his two Boston terriers — and her daughter was in a recliner. As she started in, her daughter gave her one of those looks. “Like, ‘I know what you’re going to say, and I really don’t want to hear it.’” Burns forged ahead.

“I said, ‘You’re getting to that age, you know, you do go out with your friends. If you ever happen to be with these boys that you claim’ — and I used the air quotes — ‘are all just friends—’”

“Mom, I already know all this, we have sex ed at school.”

What Burns was getting at was a bit different. She wasn’t going to tell her kids not to have kids. She wanted grandkids. But it seemed like their heart trouble was getting worse from generation to generation. That wasn’t necessarily something that Lo and Feingold knew to be true; it was just a suspicion Burns had. Cameron had been so sick at birth. He’d needed heart surgery as a baby, to repair his holes. Then, years later, at Christmas, he’d been rasping for breath, and he ended up hospitalized for heart failure again. She just wanted her kids to think about that, about all the time she’d spent teetering closer and closer to loss, about the family’s history of miscarriages and stillbirths. In a way, she was relieved when her daughter said she didn’t want to have kids, that instead she might adopt. It also made her sad.

When she’d finally heard about the results from Lo, she felt similarly conflicted. On the one hand, it felt good, to have some sort of closure, some kind of answer. On the other, she wished her grandfather and mother and aunt could have been alive to hear it. It was cool to know you were so scientifically unique. She’d told some of her friends at work. “They just kind of look at you like, holy cow. Really?”

But there were also downsides to being such a mystery. She knew there wouldn’t be a therapy anytime soon. The biggest change was that the cardiologist would want to follow up more, to keep an eye on their hearts, powering them through work shifts and classes and basketball games and hunting trips, all normal but utterly improbable.