Mad Over Stories Team For decades, a cold case with no witnesses and no confession often stayed cold forever. That changed with the rise of forensic genetic genealogy, a technique that turns a small amount of crime-scene DNA into a roadmap through a suspect’s extended family tree. It has closed cases that sat untouched for thirty or forty years, and it has raised real questions about genetic privacy along the way.
How forensic genetic genealogy actually works
Traditional forensic DNA matching only works if a suspect’s profile is already in a law-enforcement database like CODIS. Genetic genealogy takes a different route. Investigators convert crime-scene DNA into the kind of detailed profile used by consumer ancestry services, then upload it to databases that permit law-enforcement searches, historically GEDmatch and FamilyTreeDNA. Instead of a direct hit, the search returns distant relatives, often second, third, or fourth cousins.
From there the work is old-fashioned genealogy. Analysts build family trees outward from those matches, cross-reference public records, and narrow the field until a single person fits the DNA, the geography, and the timeline. A discreetly collected sample, from a discarded cup or tissue, then confirms or clears that person against the original evidence.
The Golden State Killer: the case that changed everything
The technique moved from theory to headline news in April 2018, when investigators arrested Joseph James DeAngelo, a former police officer, and identified him as the Golden State Killer, an offender responsible for a string of murders and rapes across California in the 1970s and 1980s. The break came after his crime-scene DNA was uploaded to GEDmatch and matched to distant relatives, with genealogist Barbara Rae-Venter helping build the tree that led to him. DeAngelo pleaded guilty in 2020. It was the first high-profile case solved this way, and it set off a wave of similar breakthroughs.
Not just one case
Within a year, genetic genealogy produced its first conviction at trial: William Earl Talbott II, found guilty in 2019 of the 1987 killings of a young Canadian couple in Washington State, a case identified through the same family-tree method. Since then, agencies across the country have used the approach to name suspects and to identify unknown victims whose remains sat unidentified for decades. What these cases share is a common thread, evidence that had gone as far as conventional methods could take it, revived by a distant relative’s DNA.
The privacy questions it raises
The power of the method is also its controversy. When you upload your DNA to a consumer service, you are also exposing information about relatives who never consented, and some of those relatives may become the thread that leads police to a family member. The Golden State Killer search, conducted before most users understood law enforcement might access their data, prompted a backlash. GEDmatch later shifted to an opt-in model for police matching, and in 2019 the U.S. Department of Justice issued interim guidance limiting the technique largely to violent crimes and unidentified remains. The debate over consent, warrants, and how far investigators should reach into shared genetic data is still unfolding.
For a broader look at the unsolved cases still waiting on evidence like this, browse our true crime coverage.
Related Reading
Research Notes
- Pew Research Center
- Edison Research True Crime Consumer Report
- Google Trends Summergeist
- AP / Publishers Weekly best-seller list
Editorial note: Mad Over Stories prioritizes source-aware storytelling. For true crime and real-life topics, we avoid naming private individuals unnecessarily, distinguish confirmed facts from interpretation, and update articles when better information becomes available.
