Exploring the implications of using genetic genealogy in law enforcement
By Khadija Azhar
Between 1974 and 1986, a paralyzing fear chipped away at the veneer of tranquility surrounding suburban California and settled into its picturesque neighborhoods. While the Visalia Ransacker burgled and stalked, the East Area Rapist stunned Sacramento County with a series of rapes, and the Original Night Stalker terrorized Los Angeles with a nightmarish string of murders. Victims were threatened in phone calls and the police received taunting letters, but the investigations eventually came to a standstill with no convincing suspects in sight.
It wasn’t until 2001 that forensic links were established between the seemingly disparate crimes. The Visalia Ransacker, the East Area Rapist, and the Original Night Stalker would merge into a single perpetrator: the Golden State Killer. For almost three decades, the police tried to uncover his identity but to no avail–until genetic genealogy led to a surprising arrest in 2018.
DNA profiling through the years
Even before genetic genealogy was popularized in law enforcement circles, DNA analysis was a well-established forensic technique. Its roots trace back to 1986, where crime scene DNA was first used to convict a murder suspect in Leicester, England after a mass screening of blood and saliva samples revealed a match. Since then, this method has become widely adopted, and countries around the world have developed sophisticated national databases–such as the Combined DNA Index System (CODIS) in the U.S.–to search for matches among biological samples. But there’s a catch.
CODIS only contains DNA evidence obtained from convicted offenders, crime scenes, and missing persons investigations. As a result, it also reflects and could potentially propagate racial and ethnic prejudices that are prevalent in the criminal justice system.
Alternatively, genetic genealogy uses non-forensic databases like GEDMatch and FamilyTreeDNA, which allow consumers to upload data from commercial DNA testing companies that trace ancestries when traditional documentation isn’t available. Not only do consumers provide their profiles voluntarily, but they also come primarily from Caucasian users. With an expanded trove of DNA that isn’t limited to arrestees and ethnic minorities, police can potentially solve more crimes while redressing racial disparities.
Dr. Colleen Fitzpatrick, a forensic genealogist and co-founder of the DNA Doe Project, pioneered the use of public databases in law enforcement when she realized that commercial genealogy companies analyzed the same 20 markers that forensic experts used to match DNA samples.
“The companies, when they first started off, borrowed those [CODIS] markers,” she says. “In 2011, I went to the Seattle King County Sheriff’s Department and pointed this out to them, and I eventually got Y-DNA from a cold case.”
In 1991, 16-year-old Sarah Yarborough was strangled next to a high school in the state of Washington. Exactly 20 years later, Dr. Fitzpatrick uploaded her murderer’s Y-DNA onto GEDMatch and was able to narrow down the suspect pool to distant relatives of two Mayflower pilgrims. While the killer wasn’t caught, her technique made waves in both the genealogy and forensics communities.
Catching the Golden State Killer
Investigators working on the Golden State Killer case decided to use GEDMatch in the hopes of putting a name to the reign of terror that plagued California four decades ago. Using evidence found at the crime scenes, detectives painstakingly combed every branch of the 25 family trees generated from a couple later identified as the killer’s great-great-great-grandparents. They eventually zeroed in on a 72-year-old retired policeman who not only fit the age profile, but lived in the same areas as the Golden State Killer and had purchased guns during two of the killer’s crime sprees.
In April last year, Joseph James DeAngelo was arrested in Citrus Heights after his DNA was found to be a match. While the statute of limitations had expired on the over 100 burglaries and 50 rapes he committed, he was charged on 13 counts of murder and 13 counts of kidnapping for robbery.
With an arrest as high-profile as that of the Golden State Killer, it’s easy to sweep potential pitfalls of the technique under the rug for what is deemed to be the greater good. A harder look at these oversights reveals significant hurdles in the widespread adoption of forensic genetic genealogy.
In 2014, police investigating the 1996 murder of Angie Dodge followed up on what seemed to be a promising lead, as the crime scene DNA partially matched a profile in Ancestry.com’s database. A court order forced the website to reveal Michael Usry Sr. as the DNA match, and police honed in on his son as a suspect. Michael Usry Jr. was cleared after spending 33 days uncertain of his fate as a result of a false positive. His case confirmed that a conversation about regulating the use of non-forensic databases to protect genetic privacy was long overdue.
Dr. Fitzpatrick remains skeptical about regulating the field.
“I think it will be regulated to some extent, but just because it’s so powerful and useful and available, I don’t think there’s a way to keep a lid on it,” she says.
However, it also added a clause allowing police to use its data to generate leads in cases of violent crime, removing an earlier limitation of only homicides and rapes. As was the case with CODIS, which expanded its database from sex offenders to arrestees since its inception, GEDMatch will also likely work to ease restrictions over time.
Updating terms of service may be a step in the right direction, but it doesn’t account for the misuse of these services by rogue law enforcement personnel, inexperienced genealogists, or vigilantes looking to take justice into their own hands. If these concerns aren’t enough, Dr. Fitzpatrick explains the underlying source of controversy.
“As long as you sip from a glass of water […] I could collect your DNA from that glass and find out who you are, and who your family is without you ever knowing,” she says.
The DNA Doe Project
Owing to the backlash she received following the Sarah Yarborough case, Dr. Fitzpatrick tends to steer clear of controversy. She has chosen to work with John and Jane Does instead of suspects in criminal cases.
“We felt that the Does would be much easier for the community to digest and we were right,” she explains.
The DNA Doe Project is a non-profit organization consisting of around 40 volunteers who have dedicated their downtime to solving mysteries hidden in genetic profiles and family trees. With 12 cases solved and myriads more in the pipeline, the project has only scratched the surface of the more than 80,000 active missing person cases in the U.S. (NCIC), but it’s more than enough for families that have received closure after years of searching.
Forensics isn’t as flashy as the panoply of crime shows of the noughties have led us to believe. The field has been grappling with constraints, which seem all the more restrictive given the gravity of criminal cases, but advancements in technology have alleviated some of these challenges.
In the future, techniques such as DNA phenotyping and virtual autopsies will likely become forensic staples. Until then, it’s vital to demarcate clear rules for the maintenance of genetic privacy before the debate careens out of control.
“I think it’s clear that you can’t stop this,” says Dr. Fitzpatrick, “This is the way that the world is.”
Khadija is Jumpstart’s Editorial Intern.