By Guardian - Joy Lanzendorfer - Sat 13 Jul 2024 14.00 BST

Working from her California bungalow, Margaret Press, 77, leads a pioneering team of volunteers in the field of genetic genealogy. Thus far, they’ve identified the remains of more than 100 Jane and John Does

For 37 years, the young woman remained nameless. She was known only as “Buckskin girl” for the leather poncho she wore when she was found, strangled, in an Ohio ditch in 1981.

That was until almost four decades later, when two women in California took up her case. In 2018, Margaret Press, a retired linguist, author and computer programmer, and Colleen Fitzpatrick, a nuclear physicist and forensic genealogist, ran DNA evidence from the crime scene through an ancestry database, looking for the victim’s closest relatives.

Four hours later, they had a match.

The woman was Marcia King, age 21, who disappeared after hitchhiking from her Arkansas home in 1980. Her murderer has never been caught. King’s mother still lived in the same house with the same phone number, in case her daughter called.

It was a breakthrough moment for the burgeoning field of investigative genetic genealogy, a technique that uses genealogical research to tie a forensic DNA sample to a person’s identity. Marcia King’s was the first publicly announced case in which an unidentified victim was named by this methodology.

Since then, genetic genealogy has become one of the most powerful crime-solving tools, cracking hundreds of cold cases and unmasking murderers. The most famous of the latter is the identification of the Golden State Killer, which attracted the world’s attention to the technology and revealed its potential for solving mysteries. Today the field is a veritable wild west, with amateur sleuths and armchair investigators attempting to catch bad guys, raising ethical questions of who should be involved in these literal situations of life or death.

Through it all, the DNA Doe Project (DDP), the pioneering organization Press founded with Fitzpatrick, remains one of the most effective and professional groups around. Press is admired by her peers and law enforcement alike for her efficient, proactive nature and ability to follow ideas through to their conclusion.

To date, the DDP has identified more than 120 Jane and John Does. Working with law enforcement agencies nationwide, the non-profit has produced results at an impressive rate, naming multiple victims a month, some of whom have been missing for decades.

Its success is driven by a network of more than 100 volunteers, an eclectic group that includes teachers, genealogists, police detectives, dental hygienists, private investigators and attorneys. Working in small teams online, they often haven’t met in real life, but they’re united in their goal to discover the family trees of unknown individuals.

The DDP has taken on high-profile murders, such as naming a victim of the serial killer John Wayne Gacy, and resolved long-held mysteries, such as identifying an old west outlaw whose headless remains were left in an Idaho cave in 1916. Just as often, however, the DDP is looking into accidental deaths, suicides or people who died on the fringes of society.

“These are often people who have lost their social safety nets,” says Press, 77, at her home in Sebastopol. “They’re estranged from family or they had a streak of running away or for whatever reason many of their families were not looking for them any more.”

The DDP is in the business of helping restore these lost people to their loved ones. The goal isn’t to fight crime, but to give victims back their names.

‘A lightbulb went off in my mind’

It all started with a mystery novel.

In 2015, Press had recently retired from a career as a computer programmer in Massachusetts. Wanting to be closer to family, she moved to Sebastopol, a small town surrounded by apple orchards and vineyards about an hour north of San Francisco. Press isn’t a person who wastes her time. In addition to working as a programmer and a speech and language consultant, she holds a PhD in linguistics from the University of California, Los Angeles, and has authored two mysteries and a true-crime book. On the side, she volunteered as a “search angel”, helping adoptees locate their biological parents using genealogical databases.

Why can’t we solve John and Jane Does like we do adoptees?

Margaret Press

In retirement, Press planned to slow down, hang out with her grandchildren and catch up on her reading. But two years in, while finishing Q is for Quarry, a novel by Sue Grafton based on a real cold case of an unidentified woman near Santa Barbara, Press had an idea.

“It was like a light bulb that went off in my mind,” she says. “I thought: why can’t we solve John and Jane Does like we do adoptees? We just need to find their birth parents, right? I couldn’t believe no one had ever thought of this.”

Of course, as Press is quick to point out now, other people had thought of the idea. Within genealogy communities, the prospect of identifying Does by linking their families to their DNA had been bubbling up for some time, but no one quite knew how it could be done. Press came up with the concept independently and began figuring out how to go about it, according to CeCe Moore, chief genetic genealogist at Parabon NanoLabs, a DNA technology company that primarily works with law enforcement.

“It takes a person like Margaret to not only have the idea but to put it into action,” Moore says. “A lot of us were talking about it for a long time and Margaret just made it happen. She doesn’t just let an idea sit.”

In the beginning, Press contacted prominent genealogists, including Moore, who by a twist of technological fate didn’t receive the email. Instead, Press teamed up with Fitzpatrick, a former nuclear physicist at Nasa and a forensic genealogist who, among other things, helped determine the identity of a child who died in the sinking of the Titanic. (Fitzgerald left the DDP in 2020.)

Together, Press and Fitzpatrick developed a method to ID Jane and John Does. To name Marcia King, they plugged DNA evidence provided by police into the genealogy websites Gedmatch and FamilyTreeDNA. From there, they built King’s family tree, ultimately identifying her through relatives whose DNA also appeared on the sites.

However, the DDP’s success was soon overshadowed. Two weeks after the public learned King’s name, police arrested Joseph James DeAngelo, the Golden State Killer, who terrorized California during the 70s and 80s. DNA evidence from a rape kit was linked to DeAngelo’s relatives in Sacramento, which led right to him.

Suddenly, investigative genetic genealogy was a powerful new law enforcement tool. But the DDP chose not to pursue identifying criminals. Its focus remained on learning the names of unknown people.

How volunteers identify bodies

Today Press works from her bungalow, in a bedroom she converted into an office. Her wraparound desk with laptop and double monitors looks like the command center it is. The room is filled with large whiteboards covered in multicolored sticky notes. On them, Press has written details about current cases, most of which are kept top secret as part of criminal investigations.

From this room, Press communicates with DDP researchers scattered across the country. Volunteers for the non-profit are vetted for genealogy and other relevant expertise. The organization’s success comes down to members’ dedication and hive-mind approach to working cases.

The process begins when the DDP and law enforcement agree on someone to identify. The police deliver crime scene evidence, such as blood or skin samples, to a lab, which sequences the unknown person’s DNA. The volunteers are given access to the “DNA kit”, where the genetic information has been compressed into a file that’s compatible with Gedmatch and FamilyTreeDNA. Once this is plugged into the sites, researchers can sift through data and build the family tree. The goal is to find the “union couple”: the two individuals who bring together unrelated families – for instance, the person’s parents. When that’s discovered, the Doe’s name is tracked down and handed over to officers.

However, finding the candidate’s parents, and therefore their identity, can sometimes seem impossible. Often the DDP works with spotty results, such as snippets of DNA from degraded biological material. The databases investigators use are a fraction of the size of those used by companies like Ancestry and 23andMe, which don’t allow law enforcement partnerships and are therefore off-limits.

With a smaller genetic pool to draw from, investigators frequently deal with distant relatives. They fill in gaps with old-fashioned detective work, combing through resources like yearbooks, public records, newspapers and social media. It isn’t uncommon for volunteers to analyze Facebook or TikTok accounts, noting who’s commenting and observing family relationships. Anything to complete the puzzle.

After they find a potential candidate, the DDP hands over the results and waits. It isn’t involved in informing families that their loved ones have been found – that’s the job of law enforcement. The suspense can be frustrating, says Press.

“Once we turn the case over to the agency, we feel this sense of emptiness, like we’re at the penultimate final chapter of the book [and don’t get to read it],” she says. “But we have to be very professional and say: ‘It’s out of our wheelhouse now. It’s not up to us. It’s not even our business. If the agency never gets back to us, we have no right to demand that they do.’”

For Bryan Worters, one of the DDP’s investigators, there’s satisfaction in naming the nameless. A recent graduate of the University of South Carolina with a master’s in criminology, he volunteers up to 20 hours a month for the organization. He became interested in genetic genealogy after learning about the Green River Killer, Gary Ridgway, who murdered as many as 71 women, primarily sex workers and teenage runaways, in the 1980s and 1990s. Today Worters’ goal is to bring closure to families of similar victims, however bittersweet it may be.

“It’s sometimes very heartbreaking to see how long it may take,” he says. “But no matter what the age is of a case … these are people who have lived lives, have been on the same planet as us, and they are just as deserving of being able to be identified and returned to their loved ones as much as anybody else.”

Every year, 4,400 unidentified bodies are recovered in the United States. Missing young white women have historically received the most media attention, which generates more information and leads, and means that other demographics are often ignored or overlooked. An analysis of the National Missing and Unidentified Persons System found that the remains of Native American women are 135% more likely to go unidentified than the bodies of women from any other group.

No matter the age of a case, these are people who have been on the same planet as us, and they deserve to be identified

Bryan Worters, DDP investigator

It’s a disparity that investigative genetic genealogy would seem well positioned to address. However, researchers often run into hurdles. One of the biggest, according to the DDP, is that there’s less genetic data for people of color in general. Part of the reason is historical distrust of law enforcement, as well as privacy concerns. Not only do major databases such as 23andMe ban police activity, Gedmatch and FamilyTreeDNA also allow customers to opt out of law enforcement participation, and many do. The DDP lists cases of Asian, Hispanic and Indigenous Jane and John Does as the most difficult to solve.

So far, the DDP has identified one Indigenous Jane Doe, Shirley Ann Soosay. Her body was discovered in 1980 in an almond orchard near Bakersfield, in California’s Central valley. Soosay was from the Samson Cree Nation in Alberta, Canada.

Her niece, Violet Soosay-Wolf, had been looking for her aunt ever since the family stopped receiving greeting cards from her in 1979. For decades, Soosay-Wolf drove 13 hours from Alberta to Seattle and Vancouver to search morgues, halfway houses, hospitals and cemeteries for her aunt, according to Alta Journal. In 2019, she finally gave up.

Meanwhile, by 2020, the DDP had put thousands of hours into Soosay’s murder but still hadn’t learned who she was. They knew she had grown up in the Canadian community of Maskwacis, so they put a plea on Facebook to the Indigenous people in the area asking for leads. Soosay-Wolf saw the post. She uploaded her DNA to Gedmatch, leading to a positive ID. The killer was Wilson Chouest, who was already serving a life sentence for the rapes of other women. In 2018, while he was in jail, Chouest’s DNA was linked to two Jane Does – now investigators knew one of them was Soosay. The other was an unidentified woman found in Thousand Oaks, California, who was four months pregnant when she died. The DDP is working to learn her name as well.

Even one DNA upload can be powerful, as Emily Bill, a volunteer, knows first-hand. Bill, a Los Angeles psychology professor and psychotherapist, tackles some of the DDP’s most challenging cases. Because she’s near fluent in Spanish, at least half her assignments are people of Hispanic origin. One particularly challenging case involved attempting to learn the name of a 41-year-old woman who was hit by a car in Arizona in 2004. She was from Aguascalientes, Mexico, where endogamy – marriage among members of a close-knit community – was common. Endogamy is one of the most complicated issues in genealogical analysis; in this case, the family tree was as tangled as a ball of yarn.

“I just can’t overstate how many times people were overlapping and how many people were related to each other,” says Bill. “It didn’t matter which direction you went in, you would loop back around to where you were just at, and you would be totally dizzy from being turned around.”

Three years and 1,300 hours of work went into the case with little progress toward identification. Then one day, the volunteers logged into FamilyTreeDNA and saw that someone new had uploaded their genetic material. They were shocked that a match had appeared as if by magic almost 20 years after the woman’s death. It was a wealth of information. While before researchers were working with distant relatives, this new match was closely related, which made all the difference.

They soon learned the woman’s name was Amelia Muñoz Loera. Her niece had been wondering what happened to her missing aunt and decided to upload her DNA.

A short time later, she had an answer.

Putting up guardrails

With such dramatic results, it’s no surprise that the genetic genealogy industry has boomed, which, given the sensitive nature of the cases, has some practitioners concerned.

The low barrier to entry to the field, in which people can set up shop without credentials or much money, has led to intense competition for cases, according to the journal Forensic Science International. This, in turn, has caused some unprofessional behavior. Press tells me about a self-declared sleuth, unconnected to the DDP, who told a grieving family about the discovery of their lost loved one before law enforcement had a chance to do it. In another example, an organization put out press releases announcing a solved case before it was ready to go public.

Press is working to address the lack of official standards and oversight by helping to develop some herself. She’s a member of the Investigative Genetic Genealogy Accreditation Board, which is defining best practices and a code of ethics for the field, with the goal of creating an accreditation exam for hopeful genealogists. She also worked with Moore to launch the non-profit database DNA Justice Foundation, where individuals can donate their genetic material specifically for law enforcement purposes. Almost 800 profiles have been uploaded so far.

While Press stepped away from day-to-day management of the DDP earlier this year, she’s still deeply involved in the group’s affairs. On top of that, she’s gearing up for the Hollywood treatment. The National Geographic channel and Wall to Wall Media are filming a documentary series on the DDP tentatively called Waking the Dead. Each episode will follow an investigation from beginning to end.

As a lover of mysteries, Press understands why people are drawn to this type of work. “It’s a very sexy field if you’ve got a problem solver who also wants to help law enforcement,” she says. However, the reality of identifying a victim is a more complicated experience than it may seem.

For Press, learning the name of a Jane or John Doe feels like that scene in The Wizard of Oz when Dorothy steps out of the black-and-white world into Technicolor. What started as paper files and hard facts turns into a human being. “Then we’re in a three-dimensional color world where there’s a real person,” she says.

But almost as soon as the identity is revealed and humanity is restored, there comes a poignant realization.

“It’s the moment when they get lifted up off the page and they’re a walking, talking person – except, suddenly, they’re not,” Press says.

“We see them for an instant, and then they’re dead.”

USING ChatGPT4o TO SOLVE GENETIC GENEALOGY.

Using genetic genealogy to investigate old unsolved murders, also known as cold cases, has become an increasingly successful method. This approach involves comparing DNA evidence from crime scenes with genetic profiles in public genealogy databases to identify potential relatives of the suspect. Here are some steps and considerations if you're looking to delve into this type of investigation:

1. Collect DNA Evidence: Ensure that the DNA evidence from the crime scene is preserved and available for analysis. This typically involves working with law enforcement agencies.
2. DNA Testing: The DNA evidence must be tested to create a genetic profile. This is often done using SNP (single nucleotide polymorphism) testing, which is more detailed than traditional STR (short tandem repeat) testing.
3. Upload to Databases: The genetic profile can be uploaded to public genealogy databases like GEDmatch or FamilyTreeDNA, which allow law enforcement use. Consent and legal protocols should be followed.
4. Find Matches: Look for genetic matches in the database. These matches might be distant relatives, so extensive genealogical research will be required to build family trees and identify potential suspects.
5. Build Family Trees: Use traditional genealogy methods to build family trees of the DNA matches. This may involve looking at historical records, public documents, and other sources to trace relationships and find common ancestors.
6. Narrow Down Suspects: Based on the family trees, you can narrow down potential suspects who fit the profile (e.g., age, location, etc.).
7. Confirm Findings: Once a potential suspect is identified, further DNA testing (e.g., obtaining a new sample from the suspect) is necessary to confirm the match.
8. Legal Considerations: Ensure that all steps are taken within the legal framework and with the cooperation of law enforcement agencies.

Would you like guidance on a particular case or need help with any of these steps in detail?