The DNA Doe Project:

Giving the Unidentified a Name

Ken Doyle, Promega

Share this article

Buffalo Cave John Doe

Dubois, Idaho is a small, rural town in Clark County that’s home to Buffalo Cave—part of a large network of lava tube caves that extend across the southern portion of the state. The cave is a popular site for family excursions. In March 1991, Lynn Thomas took his daughter and granddaughters on a trip to the cave to search for artifacts. Eleven-year-old Anna did find something unusual—but not what the family expected. She uncovered a human hand, well preserved in the cool and dry air of the caves. The hand appeared to have a piece of a red sweater still visible on the attached arm.

The family notified the Clark County sheriff’s office. Further investigation uncovered another arm and two legs wrapped in burlap. The sheriff’s team determined, based on the clothing, that the limbs matched a torso discovered by another family on an excursion to Buffalo Cave almost 12 years earlier. The limbs appeared to be cleanly severed from the torso, which had been recovered from a site deeper in the cave. After making little progress on the case, the sheriff’s office turned it over to an anthropology team at Idaho State University (ISU). Although the team managed to determine a few of the victim’s physical characteristics—he had reddish-brown hair and was of European descent—the absence of a head made positive identification impossible. With progress on identifying the “Buffalo Cave John Doe” at a standstill, the ISU team placed the limbs in cold storage, and the case was forgotten.

Not entirely forgotten, however. In November 2014, a sheriff’s deputy in Clark County, John Clements, was reviewing the county’s relatively short list of cold cases. The Buffalo Cave John Doe case stood out for him. He reasoned that the advancements in forensic technology since 1991 might provide a fresh opportunity to solve the case.

Clements was invited to speak about the case at ISU by then assistant professor of anthropology, Kyra Stull. His talk piqued Stull’s interest in the case, and she decided to investigate the site further. Led by Stull, an ISU team conducted additional excavations at the cave and used laser scanning devices to make a three-dimensional map of the scene, but the victim’s head eluded their search. The absence of the head meant that the team had to base their study on the limbs and torso alone. They estimated that the man was in his mid-thirties to mid-forties at the time of death and between 5’ 6” and 6’ 2” in height. They also concluded that the victim’s remains could have lain buried in the cave for up to 10 years. Stull submitted the information to the National Missing and Unidentified Persons System (NamUS).

Buffalo Cave John Doe (Image Credit)

In 2019, Amy Michael, a lecturer in anthropology at the University of New Hampshire, decided to discuss the Buffalo Cave mystery with her students. Michael had become acquainted with the case while she was a visiting assistant professor at ISU. She invited Anthony Redgrave and Lee Bingham Redgrave, forensic genealogists, to speak to her class.

Genetic genealogy uses the results of DNA analysis to search genealogy databases and public records, in order to construct family trees. Forensic genealogy applies this technique to aid in forensic identification. (See Forensic Genealogy: What Your Second Cousin’s DNA May Say About You in The ISHI Report, April 2019.)

Anthony and Lee Redgrave were volunteers working with the DNA Doe Project (DDP), a nonprofit organization dedicated to identifying missing persons and unknown crime victims. Michael asked whether the DDP would consider accepting the Buffalo Cave John Doe case.

“We were able to facilitate the submission of the case,” Anthony Redgrave says, “and were the team leads for the DDP genealogists on the case.” The DDP team coordinated their efforts with the Clark County sheriff’s office and ISU forensics researchers.

A collaborative effort among several DDP partners for whole-genome sequencing and bioinformatics analysis resulted in a GEDmatch-compatible file. GEDmatch is a publicly accessible database that can be used to compare DNA profiles from different consumer DNA testing companies. The DNA profile was also uploaded to Family Tree DNA. The DDP team then spent the next three and a half months using forensic genealogy to develop family trees that ultimately included over 30,000 individuals.

At last, in December 2019, the DDP team was able to put a name to the victim—in fact, several names, since he used a number of aliases. They made the startling discovery that the remains in the Buffalo Cave were over a hundred years old and belonged to Joseph Henry Loveless, born on December 3, 1870. The identity was confirmed by a DNA test by Loveless’s 87-year-old grandson.

The DDP research on Loveless revealed a colorful past. At the time of his death, he was wanted by law enforcement for the murder of his second wife, whom he dismembered with an axe. He had, on several occasions, escaped from jail by cutting through the bars with a saw that he kept hidden in his boots. He spent much of his life as a bootlegger and assumed numerous false identities. Although the motive for Loveless’s assumed murder and the identity of the perpetrator remain unknown, it’s likely that he was killed by local residents in retribution for the brutal murder of his wife.

The DNA Doe Project

The Buffalo Cave John Doe case was the oldest case solved by the DDP using autosomal DNA analysis—a relatively recent method of human identification employed by forensic genealogy. Colleen Fitzpatrick, Co-Executive Director of the DDP and founder of Identifinders International, is widely considered to be the founder of forensic genealogy. Her pioneering work helped solve the notorious Phoenix Canal Murders in Phoenix, Arizona.

Fitzpatrick and DDP co-founder Margaret Press were drawn together by their shared enthusiasm for genealogical research. In 2017, Fitzpatrick recalls, there was an active discussion in the genetic genealogy community about using the results from direct-to-consumer DNA tests for forensic case work. Press—who is also a fiction writer—had been helping some friends search for their biological parents. “I happened to read a novel, Q is for Quarry, by one of my favorite mystery writers, Sue Grafton.” The novel was based on the real case of a Jane Doe murder victim who had never been identified. “A lightbulb went off in my head,” Press says. “Jane Doe could be solved with the same techniques, by finding her biological parents, too. I dropped the book and ran to my computer.” Fitzpatrick and Press connected on Facebook in February 2017, and the DDP was born from that initial conversation.

The DDP relies on volunteers to solve cases, and there’s no shortage of people wanting to help. “At the moment, we have about 300 candidates on our waiting list to choose from,” Fitzpatrick says.

“People are eager to help,” Press adds. “The key qualification we look for is experience in unknown parentage cases; for example, adoptees and foundlings.” In addition, the DDP accepts volunteers with specialized skills, such as bioinformatics, or someone who is working for a law enforcement agency responsible for one of their cases.

Colleen Fitzpatrick presents at the 30th International Symposium on Human Identification

Margaret Press presents at the 30th International Symposium on Human Identification

The “Belle in the Well”

In 1981, a woman’s body was found in a well in Windsor Township, Lawrence County, Ohio. She had been strangled, and her body was thrown into the well with a rope around her neck tied to a cinder block. When the Lawrence County coroner’s office investigated, they found the body was badly decomposed, and they were unable to identify the victim based on her few personal effects. Informed by dental analysis, the investigators assigned an age range of 30–60 years. The case soon became known as the “Belle in the Well”, and it haunted the lead investigator, Bill Nenni, for 38 years and even after he retired.

One of Nenni’s former team members, Elizabeth Murray, met Fitzpatrick at a conference in 2017 and discussed the case with her. At the time, the DDP was just getting off the ground, and the founders were looking for pilot cases.

The case was one of the most challenging ever undertaken by the DDP, taking approximately 14 months to solve. During the construction of the initial family trees, it became apparent that endogamy—the custom of marrying only within the limits of a local community, clan, or tribe—was widespread in the community to which the Belle in the Well had belonged.

“When a person shares a single set of common ancestors, the DNA they share with each other falls within a predictable range that allows us to make a guess of how they are related to each other, before we ever build out their family trees,” Anthony Redgrave explains. When two people are connected through more than one set of common ancestors, the amount of DNA they share is typically more than if they only connect through one pair of ancestors. Most relationship estimates based on shared DNA assume just a single set of common ancestors, making it appear that those ancestors are more closely related than they really are. Together with the degraded state of the victim’s DNA, this complication made the DDP team question the accuracy of any information they pieced together during the first year of their analysis.

A picture and rendering of the "Belle in the Well" (Image Credit)

Other challenges the team faced were organizational. “There was a lot of streamlining of methods and retrofitting of standards of practice that we developed as we went along,” Lee Redgrave says, “as this was one of the very first cases taken on by the DDP.”

In March 2018, the DDP team uncovered information for a woman named Louise Peterson, based on a fresh round of DNA sequencing that they uploaded to GEDmatch. However, Peterson was older than the predicted age for the Belle in the Well, and the team continued to search for a better fit. A new match appeared in January 2019 on GEDmatch—a man unrelated to the other matches for cousins of the victim.

“This DNA match had a third great-grandfather, Gustav Peterson, who moved with his family from Germany to Canada,” Anthony Redgrave says. Gustav Peterson had a son, Charles, who moved from Canada to the US and married Ora Mynes—the mother of Louise Peterson. “We knew, at the moment we saw this marriage record come up, that this was the connection we were waiting for.”

The DDP team considered the possibility that Louise Peterson might have been the mother of the Belle in the Well, because of her age being outside the predicted range. However, if this were so, there would be matches related to Belle through Louise’s husband, the father of that child. The DDP analysis showed no connection to Louise’s husband, indicating that Louise was indeed Belle in the Well.

The DNA Doe Project is an all-volunteer organization dedicated to using genetic genealogy to identify John and Jane Does. Working with law enforcement, they are able to find success even when the DNA has been highly degraded. Their DoeFundMe program allows for public donations to solve cases when resources may not be otherwise available.

We sat down with Margaret Press and Colleen Fitzpatrick, co-founders of the DNA Doe Project. They describe how the DNA Doe Project was formed, how they’ve learned to overcome obstacles (such as degraded DNA), the differences in working with Doe cases versus criminal cases, and what it means to them to be able to provide identities to the unknown.

This video has been disabled until you accept marketing cookies. Manage your preferences here or directly accept targeting cookies.

Julie Doe

In September 1988, a man searching for cypress wood to build lawn furniture discovered a body in the woods off County Road 474, in rural Lake County, Florida. The body appeared to have been dragged off the road into a secluded location—concealed in thick, five-foot high weed growth—and there were signs of potential sexual assault. When Lake County deputies investigated, they reported that the body was badly decomposed. They determined that the victim had undergone several surgical procedures, including breast implants, and estimated the woman’s age at 22–35.

Investigators compiled information on a number of leads from local residents but, on following up, each potential match was tracked down and identified as safe. The autopsy initially identified the victim as a woman who had given birth to at least one child, based on observed pitting of the pelvic bones. Although the cause of death was not determined, homicide was suspected due to the circumstances surrounding the discovery of the victim’s body. Forensic students working on the case named the victim Julie Doe. At the time, DNA testing was uncommon and expensive, and no further testing was performed.

Almost three decades later in 2015, when rapid advances in forensic DNA technology had both lowered the cost and made forensic DNA analysis more accessible, testing by the University of Florida human identification laboratory revealed that the victim had an XY genotype. Evidently, Julie Doe was a transgender woman who had undergone gender reassignment surgery and hormone therapy that had caused the pelvic bone pitting observed earlier. In July 2018, University of Florida forensic researchers used isotope testing and discovered that Julie Doe was a South Florida native at the time of her death.

Initial attempts at DNA analysis were unable to extract DNA of sufficient quality from Julie Doe’s skeletal remains. In January 2020, working with a DNA analysis laboratory and bioinformatics expert, the DDP was able to obtain a DNA profile suitable for GEDmatch. Genealogical analysis is currently underway.

A rendering of Julie Doe (Image Credit)

The Trans Doe Task Force

The Julie Doe case served as the catalyst for Anthony and Lee Redgrave to form a new organization, the Trans Doe Task Force. They took on the case in 2018, recognizing the need to address unique issues faced by transgender victims. “The Trans Doe Task Force was formed to perform services not within the scope of practice of the DDP,” Anthony Redgrave says. These services include potentially working to ensure that, after legal identification is made and confirmed by law enforcement, “we are able to find an individual’s preferred name if they had one and reach out to their chosen family and community as well.” Transgender individuals face a higher incidence of violent crime compared to the general population, particularly transgender women of color. The task force provides education and outreach surrounding the wider issue of violence against trans people.

Lee Redgrave points to the discrepancy that may arise between how a transgender unidentified person is documented in a law enforcement database, compared to how a family member or loved one may report them as missing. “Families may also be more likely not to report a trans- or gender-variant person as missing.” Rejection by family is a major cause of homelessness among the LGBT community.

“Additionally, trans people are far more likely not to have records or identification that matches their lived identity,” Lee Redgrave explains. Since each form of identification has to be changed separately, the process requires access to resources that many people do not have.

Future Directions

The world of forensic genealogy is continually evolving and adapting to change. One example is the landmark decision by GEDmatch in 2019 to change its privacy policy. In the past, users uploading their DNA profiles automatically consented to use of the data by law enforcement agencies, to trace potential family relationships with victims of violent crime or to rule out suspects in criminal cases. GEDmatch changed the policy after public backlash in a couple of high-profile crime cases, so that all profiles were automatically opted out of law enforcement use, and new users had to opt in. The database was recently sold to genomics company Verogen, which has stated that it intends to maintain the current GEDmatch terms of service.

This decision resulted in an increase in the average amount of time required for the DDP to solve a case. However, because the DDP is volunteer-powered, Fitzpatrick reiterates that the decision hasn’t impacted the success rate. The DDP has also turned to other genealogy databases. “We now routinely rely on what we can get from uploading to the Family Tree DNA database as well, though that is quite expensive,” Press says.

In some ways, the DDP has become a victim of its own success. Fitzpatrick points to the many cases the DDP has solved, as well as to their expertise in working with compromised DNA samples, stating that the DDP is “overwhelmed with requests from law enforcement agencies.” Another contributor to this problem is the organization’s generosity. “We are the go-to organization that takes on the most challenging cases for free,” Press says. Many of these cases require thousands of hours of time from professional genealogists, putting them beyond the financial reach of other organizations. Press adds, “We don’t turn away any case, regardless of how tough.”

Like many nonprofit organizations, the DDP has not been immune to the economic effects of the current COVID-19 pandemic. “Agency budgets are starting to disappear due to the pandemic,” Press says, and donors have cut back on contributions.

Both Press and Fitzpatrick remain committed to the DDP while pursuing other causes. Fitzpatrick’s work at Identifinders International complements what she does for the DDP. In addition to working on cases where identification may lead to criminal charges, Fitzpatrick says she is involved with cases of “Baby and Child Does, where the child may have been harmed by their parent or caregiver.”

When asked whether she continues to write mystery novels and true crime, Press expressed her regrets, explaining that she hasn’t had time to write at all. Further, she says, “Writing true crime is fraught with land mines—privacy issues, risk of potential lawsuits, and talking to real people in real pain.” She agrees that her work has taught her that truth is stranger than fiction: “Absolutely. No question.”

Fitzpatrick and Press are also looking forward to a new adventure: a television series, Blood Relative, based on the work of the DDP. “It will raise awareness of genetic genealogy and educate the general public on how it is used for forensic work,” Fitzpatrick says.

Press explains that they are embracing the opportunity to steer the writers in the right direction and ensure that the show is based on real science. At the same time, she says, “We also understand that this is TV…and sponsors will want to see something more engaging than characters mousing over a chromosome browser.”

Meanwhile, Anthony and Lee Redgrave are also dealing with a time of transition. They have temporarily taken down the website for the Trans Doe Task Force while they plan for a relaunch to focus on further creation of educational programs. They’re also working with other groups researching missing LGBT people. “We hope to create a database to cross-reference these cases in ways that currently can’t be accomplished with standard resources,” Lee Redgrave says.

Ultimately, the vision of the DDP and the Trans Doe Task Force is that we can live in a world where the lists of John and Jane Does shrink and, eventually, disappear. As Fitzpatrick and Press state on the DDP website: “No one should die without a name.”