Cousins in the Code

Folders in the Dust

Eli: In 1994, Paul Holes was opening an unused file cabinet in a library when he noticed old folders labeled with the letters EAR. Inside were forgotten case files on the East Area Rapist.

Mara: A listener asked: How did a genealogy website finally catch the Golden State Killer, decades after his crimes? Today's angle is from a cold case dead end to a living family tree.

Eli: Finding those folders pulled Holes into what would become a twenty-four-year pursuit as he led the Contra Costa County effort on the case. For years, the crimes looked like unconnected sprees in different jurisdictions.

Mara: Forensics eventually changed that picture.

Eli: Forensic evidence proved that Northern California rapes and Southern California homicides were committed by the same individual. By March 2001, DNA testing directly connected the two strings of crimes to a single perpetrator.

Mara: Yet having a genetic profile did not solve the mystery.

Eli: Standard law enforcement databases produced zero matches. His DNA was not in CODIS, the federal database storing profiles of known offenders to link scenes to convicted individuals.

Mara: State data banks, an Ohio comparison, fingerprints, and cash rewards all came up empty.

Eli: Detectives turned to open genealogy platforms only after exhausting every traditional forensic database and running out of conventional leads. A different path surfaced through the work of investigator Peter Headley.

Mara: Headley had been working an entirely different kind of cold case.

Eli: Headley worked with civilian genealogist Barbara Rae-Venter to identify Lisa Jenson, a woman who had been abducted as a toddler and abandoned in California. By building genetic trees, they discovered her real identity was Dawn Beaudin.

Mara: Those were methods built for families, not crime scenes.

Eli: Investigators adapted adoptee searching, using techniques originally taught by genealogists to help adopted individuals locate biological parents through voluntary genetic networks and public archives.

Mara: Applying family-reunification techniques to unidentified killers had never been tested on this scale.

Eli: Holes called Rae-Venter in 2017 after reviewing that breakthrough. After she checked his case history, she agreed to help build trees for the Golden State Killer, Mara.

The Backup Swab

Mara: To search genealogy websites, the team could not use the old forensic markers. Standard criminal databases look at twenty specific points of variation known as short tandem repeats.

Eli: While consumer databases require hundreds of thousands of markers.

Mara: They needed SNP profiling across the entire human genome to assemble a dense digital file. But decades of testing had depleted almost all crime-scene evidence.

Eli: Until Holes found an overlooked preservation habit in Southern California.

Mara: In the 1980 Ventura County double murder of Lyman and Charlene Smith, pathologist Peter Speth had collected two rape kits. He gave one to police and kept an untouched duplicate in the coroner's office in case the first was lost.

Eli: That backup kit sat preserved for nearly four decades.

Mara: In January 2018, specialists pulled semen from that vaginal swab, extracting enough pristine DNA to compile a comprehensive single nucleotide polymorphism data file, mapping single-letter variations in the DNA.

Eli: Major consumer genealogy services generally required a warrant or court order. So Holes turned to GEDmatch.

Mara: GEDmatch operated under very different rules.

Eli: GEDmatch was a free, open website where hobbyists uploaded raw data from various commercial services to find relatives across different platforms. In 2018, it could be queried without any court order.

Mara: Holes set up an undercover account to search the file.

Eli: The platform never showed investigators personal profiles or raw DNA sequences. It generated a match list showing shared segments measured in centimorgans, a genetic unit quantifying shared DNA to indicate biological closeness.

Mara: Could Holes really work with that?

Eli: The challenge was converting those modest numbers into an identity.

Building Downward

Mara: This was the turn in the investigation: an untouched duplicate 1980 rape kit yielded pristine semen for an autosomal profile, allowing Barbara Rae-Venter to map distant matches back to nineteenth-century ancestors and build reverse family trees downward.

Eli: Tracing backwards from distant cousins rather than downwards from known suspects.

Mara: She relied on adoptee searching, tracing shared consumer matches through genealogical records to identify unknown biological parents. Here, the team constructed a reverse family tree, following matches back to shared ancestors and mapping every descendant downward.

Eli: They examined marriage certificates, census rolls, obituaries, and social media.

Mara: The tree reached great-great-great-grandparents in the early 1800s. Hundreds of relatives appeared across several branches, and volunteers narrowed the candidates based on age and whether their histories placed them in California during the crime waves.

Eli: A crucial break came from an additional search.

Mara: In February 2018, Rae-Venter searched the profile through her MyHeritage account and uncovered second-cousin matches. That narrowed the vast family network down to a specific branch.

Eli: On March 18, 2018, investigative assistant Monica Czajkowski alerted the team that they had found another man.

Mara: That man was Joseph James DeAngelo. His name fit the family tree, but documentary records alone could not confirm he was the killer.

Trash at the Curb

Eli: Detectives began vetting DeAngelo against physical and biographical clues. Barbara Rae-Venter emphasized that genetic genealogy provides an investigative lead rather than courtroom proof, requiring traditional detective work before making an arrest.

Mara: Physical traits from the genetic data matched him closely.

Eli: Investigators used DNA phenotyping, analyzing genetic markers to predict physical traits like eye color, which indicated blue eyes.

Mara: His past employment raised immediate alarm.

Eli: DeAngelo was a former police officer who was fired from the Auburn Police Department in 1979 after shoplifting a hammer and dog repellent. Willick's account matched what Holes associated with the Golden State Killer, becoming a turning point in deciding to obtain his DNA.

Mara: Surveillance teams followed him to collect confirmatory evidence.

Eli: Officers shadowed him to a Hobby Lobby and swabbed his car door handle. That sample produced an initial mixed DNA reading, which was promising but insufficient to justify an arrest.

Mara: They waited for a clean sample discarded in public.

Eli: On April 23, 2018, investigators retrieved a piece of discarded tissue from the trash can placed outside his home. The crime lab immediately ran a confirmatory test.

Mara: They conducted STR confirmation, comparing standard forensic markers directly against the original 1980 semen swab to establish legal identity.

Eli: The profile was a one hundred percent match. It tied DeAngelo to three Contra Costa rapes and eight Southern California murders.

Mara: The genetic trees and the physical evidence had converged completely.

Eli: On April 24, 2018, deputies arrested seventy-two-year-old Joseph James DeAngelo outside his home in Citrus Heights.

Mara: Officials announced the capture days later, crediting Rae-Venter's genealogical analysis.

Eli: Holes said he thought it was the first time an unknown offender had been identified through consumer genetic genealogy, Mara.

The Opt-In Line

Mara: Investigators caught the Golden State Killer by uploading crime-scene DNA to GEDmatch, tracing distant cousin matches back to shared ancestors, building family trees down to Joseph James DeAngelo, and confirming his identity with discarded trash DNA.

Eli: The method spread quickly across law enforcement, resolving hundreds of violent cold cases that had stalled for decades.

Mara: The arrest also provoked immediate backlash. GEDmatch co-founder Curtis Rogers stated that law enforcement had never contacted the site about the investigation, and its unannounced use occurred without platform knowledge.

Eli: Legal analysts debated whether searching genealogy databases violated Fourth Amendment rights, though legal scholars noted that voluntarily sharing genetic information with a third party usually diminishes privacy expectations under established doctrine.

Mara: Different views emerged at once. Investigators like Paul Holes, genealogist Barbara Rae-Venter, and supporting prosecutors viewed searching open databases as an essential public safety measure, using voluntarily shared public data to solve violent crimes.

Eli: On the other hand, GEDmatch co-founder Curtis Rogers said he had not known the site was being used to pursue the killer and insisted that it did not hand out data.

Mara: The industry shifted rapidly. GEDmatch adopted opt-in consent, requiring existing users to actively permit law enforcement to compare files against their profiles.

Eli: Users had to deliberately check a box for that search.

Mara: Hundreds of thousands of users opted in, but the policy locked nearly ninety percent of existing accounts away from criminal investigators.

Eli: Federal standards followed soon after.

Mara: In late 2019, the Department of Justice issued an interim policy, establishing federal rules limiting genetic genealogy searches to violent crimes or unidentified human remains only after traditional CODIS searches fail.

Eli: Decades after Holes first found those files, closing the case required preserved evidence, open databases, and traditional legwork.

Mara: Whether future disclosures reveal the exact laboratory protocols used to extract low-level DNA from the 1980 semen swab remains a signal to watch.

Eli: For listeners wanting to look into the method, reverse family trees.

Mara: Follow Law & Justice as investigators navigate genetic privacy.

Eli: That's our story for today. I'm Eli.

Mara: And I'm Mara. Thanks for listening to Angle, and until next time.