The result was a profile that was substantially below the quality of what consumer genetic testing produces from a saliva sample, but that was sufficient to be compared against the GEDmatch database — an open platform where individuals who have obtained genetic profiles from various testing companies can upload their results and compare them against a large pool of other uploaded profiles to find previously unknown relatives.
When the composite profile from the Chandler sample was uploaded to GEDmatch, it began producing matches. The matches were distant — third cousins and beyond, consistent with the limited amount of shared DNA that the degraded profile could reliably detect. But they were real matches, confirmed by statistical analysis, and they were the starting points for genealogical research.
The 1,500 hours of research that Dr. Press described as invested in the case represents the work of building and exploring family trees from those distant starting points. The DNA Doe Project used volunteer genetic genealogists — experienced researchers who donated their time to work on the case — to trace the family trees of the identified matches backward and forward in time, looking for the point where those trees intersected with a candidate who fit the profile of the deceased.
The evidence converged on a family in Indiana: the descendants of Silas and Alpha Nichols. The surname match — Nichols, consistent with Fitzpatrick’s 2016 Y-STR result — was confirmed. The genealogical research pointed to a son born in 1926 in New Albany, Indiana, who had served in the Navy during the Second World War, who had last been definitively placed by family in the mid-1960s, and who had never been heard from again.