Blood Work: The Citizen Sleuth Using Genealogy to ID the Dead

Margaret Press is transforming forensic genealogy one Doe at a time.
By Leah Worthington

ON APRIL 24, 1981, THE BODY OF A YOUNG WOMAN with auburn braids and a fringed jacket was discovered off the side of a road in Troy, Ohio.

She had been strangled to death only hours before. Authorities took DNA samples but couldn’t find a match for the woman. For decades, she was described only by the clothes on her back: “Buckskin Girl.”

In March of 2018, 37 years later and more than 2,000 miles away, Margaret Press sat behind the glow of her computer. It was nearly 1 a.m., but she wasn’t going to sleep anytime soon. She watched as the genomics site, GEDmatch ran through Buckskin Girl’s profile, comparing it to hundreds of thousands of other DNA profiles in the public database. As closer and closer matches appeared—a fourth cousin, a third cousin, a second cousin once removed—Press recalls “hitting refresh over and over again.”

“People love doing this. We have a waiting list of about 200 genetic genealogists who want to donate their time to help us.”

A 72-year-old retiree and mystery writer, Press has spent the last two years of her life identifying bodies. Along with her partner-in-solving-crime, Colleen Fitzpatrick, she co-founded the DNA Doe Project, a nonprofit that uses forensic genealogy to piece together the stories of the nameless dead, the Jane and John Does of the world.

Since 2018, Press has helped identify more than 20 such nameless victims, many of whose cases had gone cold decades ago. Her successes have earned her recognition in the media, where she has been described as a “DNA crime solver,” “volunteer sleuth,” and (her favorite) “Sonoma County grandmother.” With her short-cropped white hair, downturned mouth, and penchant for mysteries, it’s hard not to think of a modern-day Miss Marple.

A self-described hermit, Press conducts business from a quiet cottage at the end of a long driveway in Sebastopol. When we met there last June, she led me straight to her office, an airy room that’s empty, save for two computer monitors, a tetris of wooden tables, and scattered filing cabinets. The only color is a splash of pink and yellow Post-its on the wall, one for each Doe. Their scribbled names, or monikers rather (they’re here on this wall precisely because they’re nameless), are her constant companions as she works all day and late into the night researching family lineages, building genealogical trees, and managing a team of more than 60 volunteers from all around the country. An “ensemble cast,” she calls them, describing the eclectic mix of genealogists, accountants, artists, and ex-law enforcement personnel who communicate almost exclusively through private Facebook groups.

“People love doing this,” Press says. “We have a waiting list of about 200 genetic genealogists who want to donate their time to help us.” She thinks they’re drawn to the humanitarian cause and the allure of solvable mysteries. “That’s the Las Vegas combination that keeps people pulling the slot machine, that keeps people totally addicted.”

PRESS GREW UP WITH MYSTERIES. She read the Hardy Boys and started her own detective agency to solve neighborhood crimes by “sneaking around other people’s backyards” and “going around the house finding all the lost money” (which, to her mother’s chagrin, was often “lost” in her purse). As a teenager, she spent long weekends with her grandmother, an amateur genealogist who’d been researching their family’s history for years. What could be better than investigating a mystery about her own family?

Her sleuthing was put on hold when she left Pasadena to study linguistics at UC Berkeley. After graduating in 1968, she moved east and began a career as a software engineer. Several years later, divorced and raising a daughter, she moved to Salem, Massachusetts, drawn there by the town’s mystique and by her own family history. She’s a descendent of a man named William Towne, whose daughters were accused of witchcraft.

On one of her first visits to a park in town called Salem Willows, she noticed a little green restroom by an old waterfront arcade. In the winter, everything is shuttered, “and the leaves are blowing around and there’s all these willow trees and water on three sides,” she remembers. “And I said, ‘I’m gonna write a mystery. I want to put a body right here.’”

Her directness catches me off guard, and she laughs at my surprise. “Well that’s how mystery writers think!

“I thought, ‘OK, if I was gonna kill somebody in the park and stick them in the men’s room, [they] wouldn’t be discovered until spring.’”

To gather material, she began hanging around the Salem police department and chatting up the detectives. “I really loved the excuse to immerse myself in what essentially was another culture for me because I had no experience with law enforcement at all.” Soon she was a regular on the night shift, riding in the back of the patrol cars, tagging along on drug busts, and gathering tidbits for her stories (“panning for gold,” she calls it). By her account, it was a sight to behold, the middle-aged woman running around with the cops. “I had this sort of self-deprecating image of me, you know, waddling along behind them with my purse.”

Press remembers one night when they were called to a break-in at an apartment. “So we traipsed in, you know, me with my funny old lady purse and them in their leather jackets and guns.” The owner, a nurse, was out working the night shift, and the detectives wandered through the apartment inspecting the scene.

“As they were going about their business and dusting for prints, they said, ‘OK, Margaret, so here’s the broken window in the bedroom. What do you think?’ And then they said to me in all seriousness, but not seriously—you could see the twinkle in their eyes, ‘So, where do you dust next?’ And I’d say, ‘Well, the door to the apartment was slightly ajar when we came, right? I would dust the doorknob.’ And they’d say, ‘Good idea!’”

Press published her mystery novels—Requiem for a Postman and Elegy for a Thief—in the early ’90s, followed a few years later by a true crime book entitled Counterpoint: A Murder in Massachusetts Bay. The murder case, which dealt with an artist, a sailor, and a psychic, was one of Salem’s most bizarre. Press called it a “very, very lucky opportunity” because “the victim and the killer were both in my neighborhood, within half a mile.”

“It was a Hail Mary. They had DNA; no one else could do anything with it. So why not give it a shot?”

Press’s first foray into genomics came after she read a 2006 article in The New York Times about consumer DNA testing. She immediately ordered a kit and used her results to confirm much of her grandmother’s genealogical work. “That was very, very exciting for me,” she says.

She spent the next few years diving back into ancestry research and helping people, mostly adoptees, track down lost family members. In 2015, she retired and moved back to California thinking, “Now I get to read and I get to write and I get to do fun genealogy and spend time with my grandkids.”

She sighs, pauses. “Little did I know.”

The idea of using genealogy to find Does came to her while reading Sue Grafton’s alphabet mystery series. She had just finished Q is for Quarry, about a real-life Jane Doe from Lompoc, when it occurred to her, “If we can do this for adoptees, why can’t we do this for unidentified remains? It’s the same idea: You’re looking for parents,” she says.

Press first reached out to the Santa Barbara County police with her idea, but they never returned her calls. So she turned to someone who could help: Colleen Fitzpatrick.

Fitzpatrick, who wrote the book on modern forensic genealogy, had the network that Press lacked. She saw potential in the idea and got in touch with two friends—a U.S. marshal who’d been investigating a suicide victim from Ohio and a detective in Connecticut who’d been trying to identify a dead baby. Press and Fitzpatrick offered the officials an exchange: You give us the DNA, and we’ll pay for the analysis. They gladly agreed.

In both cases, Press says, “they figured it was a Hail Mary. They had DNA; no one else could do anything with it. They tried it through CODIS [the FBI’s criminal DNA database], they tried everything else. So why not give it a shot?”

In the summer of 2017, the crime lab in Ohio sent the last bit of remaining DNA to be sequenced. The Doe had shot himself more than a decade earlier, and his DNA was significantly degraded. But by late September, they had a file to upload to GEDmatch, and genetic relatives started popping up right away. They pieced together the family trees in what Press calls a sort of ancestral “triangulation,” and a name emerged: Robert Ivan Nichols. Press and Fitzpatrick were in awe. If even with such a low-quality sample, they were still able to find a match, this idea of theirs might actually work. The DNA Doe Project was born.

IN EARLY 2017, A FORENSIC ANTHROPOLOGIST named Dr. Elizabeth Murray heard Fitzpatrick speak about forensic genealogy at a conference and immediately reached out. She knew of a cold case from her district in Ohio involving the murder of a young woman wearing a tasseled leather jacket. One of the state’s longest unsolved cases, Buckskin Girl continued to haunt investigators, and Murray hoped the Doe Project could help.

Almost 40 years after her death, all that remained of Buckskin Girl was fingerprints, dental work, and a single tube of blood. Several labs expressed doubt that there would be anything left to analyze. But expert Dr. Weining Tang was able to salvage enough DNA to put together a “library” of the victim’s genetic material.

“One could be just sheer luck. And to have a second one—I think it told us that this was going to work, that we had really stumbled on something and had a tiger by the tail.”

This genetic “library,” Press explains, is sent off for sequencing and converted into massive, computerized files. These are shared with a bioinformatics expert who extracts smaller datasets containing only the crucial genetic regions. (In forensics, over 99 percent of DNA is useless because there isn’t enough variation in most genes to conduct meaningful comparisons between individuals.)

About two months after accepting the case, Press and Fitzpatrick finally received the files to upload to GEDmatch and begin the genealogical work. “That’s the moment where all the volunteers have been assembled,” Press says excitedly. “They’re all over the country. They’re all on their computers. And they’re all camping out in our Facebook group.”

GEDmatch is a much more powerful tool than law enforcement databases. It looks at thousands of genetic markers, where the FBI’s CODIS looks at 20. And where CODIS only contains criminal profiles, GEDmatch casts a wider net. It takes about 24 hours for GEDmatch to crank through the nearly 1 million people in the database and identify all the genetic matches.

Press and Fitzpatrick waited as the first hits started to appear.

“We had a team all lined up,” Press says. “It was almost like a delivery room where the mother’s in labor and she’s coming in and the team is all ready and we’ve got our hands out waiting for the baby to pop out.”

The matches come in backwards, beginning with distant relatives who share comparatively few markers with the target. Then the volunteers start googling furiously, looking for Facebook connections, email addresses, family listings, any clues that might help in the search.

After a couple of hours, the Doe Project hit gold: GEDmatch had identified a half first cousin or first cousin once removed, someone “close enough that they would know” the victim.

Press clicked on the father’s tree and saw four kids, three of whom were still living. The fourth read, “death date unknown, assumed dead.”

“And that’s when I was on the phone with Colleen, and I said, ‘I just had shivers go down my spine. This is her. This is her.’” She was in tears, laughing. After decades without progress, the mystery of Buckskin Girl’s identity took mere hours to solve. “It’s like, we cannot believe this actually effing worked!”

If the case of Robert Ivan Nichols had shown promise, finding Buckskin Girl sealed the deal. “One could be just sheer luck. And to have a second one—I think it told us that this was going to work, that we had really stumbled on something and had a tiger by the tail.”

First thing the next morning, they called Dr. Murray, the forensic anthropologist in charge of the case. “She was blown away,” Press remembers. The mother, who was still alive, was notified by the sheriff’s office and taken into the police station for a DNA swab. Five days later, in April of 2018, they had official confirmation—and a name: Buckskin Girl was Marcia L. Sossoman King.

THEIR ANNOUNCEMENT WAS THE FIRST OF ITS KIND. But just three weeks later, a similar project made headlines: Barbara Rae-Venter (ex-wife of J. Craig Venter, a driving force in the Human Genome Project) and her investigative team stole the national spotlight with the arrest of the Golden State Killer, a notorious serial murderer and rapist who had terrorized California in the late ’70s and early ’80s. For a while, identities of serial criminals and victims were being published week after week. Some struck closer to home than others. The suspected NorCal Rapist, arrested in September 2018, was a 26-year employee of Berkeley.

“We’re not the only ones to have cracked open this massive thing,” Press says. “We were just the first ones that the world heard about. And we are still at the forefront.”

Yet what seems, from the outside, like a success for forensics has become a major point of controversy among those doing the work.

“I’m of a camp that says privacy is your own responsibility. If privacy is a concern for you, stay off the Internet, don’t test!”

“The huge split right now in the genetic genealogy community is, was this ethical to do or not?” Press says. The question centers around “informed consent” and whether the thousands of people on GEDmatch knew what they were signing up for. The debate reached a boiling point this spring when GEDmatch added an “opt-in” button, which requires users to actively choose to share their data with law enforcement. Overnight, forensic genealogists around the country lost access to much of the data they’d been using to crack cases.

Press and Fitzpatrick knew their work would be controversial, which was why they decided early on to stick with identifying Jane and John Does. They hoped people would be less concerned about their DNA being used to track down the identities of victims than those of perpetrators. (Right now, they’re working with officials to identify victims of the 2018 Camp Fire that ravaged Butte County.) Even celebrity genealogist Barbara Rae-Venter tried to keep her work on the Golden State Killer a secret. “She was concerned about the same thing we were, which was, if there’s enough controversy, GEDmatch may shut down … and if GEDmatch shuts down, we have nowhere to go.”

Press describes the forensic genealogy field today as the “crazy Wild West. Everybody’s in the game.” It started with citizen scientists, but soon professional genealogists, forensic labs, and detectives were rushing in to get a piece of the action. And law enforcement’s involvement, in particular, has raised concerns among people who aren’t sure they want police looking at their genes. This is frustrating for Press who doesn’t believe that DNA should be treated any differently from other information we share online. “We use Facebook all the time to track down our candidates, just the same way that law enforcement does. There’s a ton of information you can get from Facebook where people don’t even realize it.”

Press, unlike many of her colleagues, believes the burden is on the user to understand the risks of uploading their DNA. “I’m of a camp that says privacy is your own responsibility. If privacy is a concern for you, stay off the Internet, don’t test! But once your results are anywhere, no one can guarantee it won’t be hacked.”

That said, she worries that forensic genealogy may someday be used to implicate perpetrators of lesser crimes, or even non-crimes. “If they start allowing assault, then what is it going to be next—parking tickets? Is it going to be, more seriously, people who show up at certain rallies, protesting the government? Is it going to be—and we’ve seen this start—women who’ve had abortions?” The fact is, she says, genealogy sites could be used for “a lot worse things” than hunting down a serial killer.

PRESS’S FAVORITE PART OF HER WORK is making the call to law enforcement. In the case of Marcia King, some of the detectives involved had dedicated years to the search and were blown away by the news. “They never believed that we could really do it,” she says. “One guy said to me, ‘You can’t even begin to explain to me what you did. Because it’s witchcraft.’”

King is still the project’s fastest solve—done in about four hours. Asked whether police resented being outsmarted by a couple of elderly women, Press says no: An answer, no matter how it’s found, or by whom, makes the detectives’ years of angst and effort worthwhile. “Even though secretly we know, in the end, none of that mattered. None of it mattered. All we needed was the DNA.”

Although identifying King was a huge break in the case, there are still unanswered questions: Who killed her? What were the circumstances that led to her death? Press compares it to reading 19 chapters of a 20-chapter mystery—a special kind of torture for puzzle-driven minds. But there are always more cases to be solved.

With more than 28 ongoing investigations (and another 23 in the pipeline), workshops to teach, conferences to attend, and all the recent media attention, Press lives and breathes her work. She’s at her desk often 15, 16 hours a day, hardly stopping to put on a fresh shirt. She checks her email religiously and since the Doe Project formed, hasn’t been unreachable for a full 24 hours. Even talking with me for an hour or two uninterrupted made her nervous.

“I feel like right now I’m an air traffic controller. And while we’re talking, planes are crashing.”

Tending to her cases means neglecting just about everything else. “I forget to do laundry, and I forget to go to the bank, I forget to go get more coffee, and I’ll regret that tomorrow,” she says. “Another week has gone by and I haven’t talked to my grandkids.”

She’s joking, but if there’s one thing she’s learned in her work, it’s the importance of family.

Despite what the media might have us believe, the well-loved, well-off murder victim is the outlier. Mostly it’s the drifters, the loners, the people who’ve lost their families or been estranged from their communities who end up in the morgue. They’re often in with the wrong crowd and forgotten at the time they disappear.

“I feel like right now I’m an air traffic controller. And while we’re talking, planes are crashing.”

Marcia King had been hospitalized with mental health issues and had run away a few times before she disappeared for good in the spring of 1981. Given her history, no one took the missing-person report seriously. Still, her mother kept looking; she never moved or changed her phone number in the hopes her daughter might someday find her way home.

Sometimes, after the Doe’s identity is announced, relatives will come to Press and Fitzpatrick—who are intimately familiar with the victim’s life—looking for more answers. Other times, the families just need to express their pain. While Press is careful not to get too emotionally involved, these encounters can be a kind of reckoning. King’s mother made it clear that for the families, there’s no such thing as a happy ending.

“It’s not closure, it’s bringing up the pain all over again,” Press remembers her saying. “You will move on to your next case; I’ll never move on.” 

Leah Worthington is senior online editor of California. She hopes to be Nancy Drew when she grows up.

From the Winter 2019 issue of California.
Image source: Detail of photo by Marcus Hanschen
Share this article:
Google+ Reddit

Add new comment