The investigative implication of these relationships is that a crime-scene profile compared against a genealogy database can produce not an identification but a set of relatives — people who share genetic material with the unidentified person. Those relatives, by themselves, are not suspects. They are starting points for genealogical research. The task becomes building family trees from the relative matches, using traditional genealogical methods — birth records, census data, marriage records, newspaper archives, obituaries — to map the relatives’ family structures and identify who in those structures might be the person the investigators are looking for.
This is investigative genetic genealogy: the use of consumer DNA database matches combined with traditional genealogical research to develop an investigative lead in a criminal case.
Paul Holes contacted genealogist Barbara Rae-Venter in March 2017. Rae-Venter had a background in molecular biology and patent law and had developed expertise in genetic genealogy, including in complex cases involving adoptees seeking biological relatives. Holes asked her to apply her skills to the Golden State Killer’s DNA profile.
By October 2017, she was working with the investigative team on the case.
Building the Family Tree From Distant Matches
The specific genealogy database used in the Golden State Killer investigation was GEDmatch. At the time the investigation was underway, GEDmatch was an independently operated, volunteer-run website that allowed users to upload raw DNA data files from commercial testing services and compare those files against others in the database. It was not primarily a commercial consumer service; it was a tool for genealogists who wanted access to matching capabilities that extended beyond any single company’s user base.