Consumer genetic testing services like 23andMe and AncestryDNA generate profiles of this kind from saliva samples, examining hundreds of thousands of positions across the genome. GEDmatch and similar platforms allow people to upload profiles obtained from different services to a common database where comparisons can be made across the entire pool.
For forensic purposes, a profile developed from an unknown individual’s biological material — blood, tissue, hair roots — can be compared against the profiles in GEDmatch or similar databases to find biological relatives among the database’s users. The relatives’ known identities then serve as starting points for genealogical research: building family trees from the relatives’ known ancestors, tracing those trees forward to the generation that the DNA evidence indicates the unknown person belongs to, and identifying candidates whose biographical details align with what is known about the deceased.
The technique is powerful because it does not require the unknown individual to have a prior record in any law enforcement database. It works as long as at least some biological relatives have submitted their DNA to a consumer testing service or genealogical database — a condition that is increasingly likely to be met as those databases have grown to include tens of millions of profiles.
The ethical dimensions of this technique have been actively debated. When a person submits their DNA to a consumer service for genealogical purposes, they are typically consenting to comparisons against others in the same database to find relatives. They may not be explicitly consenting to having their genetic relationship to an unknown individual examined by law enforcement investigators. Their DNA effectively implicates their biological relatives in ways that those relatives may not have agreed to.