How a Family Tree Catches Killers

For most of modern policing, DNA could only catch someone who was already in the system. A profile from a crime scene was compared against the Combined DNA Index System (CODIS), the national database of offender profiles. If the person who left it had never been arrested and swabbed, the profile sat there — accurate, damning, and useless. That is how cases went cold for forty years with the killer’s own DNA in the evidence locker.

Investigative Genetic Genealogy (IGG) changed that. It is not magic and it is not a conviction machine. It is a method, and it works in five steps. Here is each one, plainly.

The method, step by step

Step one: the crime-scene DNA. Investigators start with biological evidence from the scene — blood, skin cells, a stain on fabric. A specialized laboratory builds a much richer profile from it than the standard CODIS profile: hundreds of thousands of genetic markers called single nucleotide polymorphisms (SNPs), the same kind of data a consumer ancestry test produces.

Step two: the genealogy databases. That profile is uploaded to genealogy databases whose users have consented to law enforcement matching — not Ancestry or 23andMe, which do not permit these searches, but databases like GEDmatch (where law enforcement matching is opt-in) and FamilyTreeDNA, plus purpose-built databases such as DNASolves. The search does not return the killer. It returns relatives: usually distant cousins who share small segments of DNA with whoever left the sample.

Step three: the family tree. This is the genealogy part, and it is slow, careful, human work. Researchers build the family trees of those cousin matches backward — to shared great-great-grandparents — and then forward again, branch by branch, looking for the place where the branches converge. They narrow by what the evidence already says: the person’s likely age, where they lived, sometimes ancestry markers. Eventually the tree points to a household, a sibling group, sometimes a single name.

Step four: the lead. That name is a lead. It is not proof of anything, and responsible investigators treat it that way. Nobody is arrested because a family tree converged on them.

Step five: traditional confirmation. Police go back to ordinary, time-tested police work. They obtain a direct DNA sample from the person — often from discarded items, sometimes by court order — and compare it to the actual crime-scene profile using the same forensic standards that have governed DNA evidence for decades. The genealogy gets investigators to a door. The confirmation match, and everything that follows — interviews, records, prosecution, a courtroom — is what closes a case.

Three cases where it worked, all the way through

The Golden State Killer. For more than four decades, the man responsible for a string of murders and rapes across California left DNA at scene after scene, and CODIS never returned a name — he had never been swabbed. In 2018, investigators uploaded a SNP profile built from crime-scene evidence to GEDmatch. Distant cousin matches led genealogists to build trees that narrowed, eventually, to Joseph James DeAngelo, a former police officer of the right age living in the right places at the right times. Officers confirmed it with DNA collected from items he discarded. In June 2020, DeAngelo pleaded guilty to thirteen counts of first-degree murder and admitted to dozens of uncharged crimes. He is serving consecutive life sentences without parole. This was the founding case — the proof that the method works.

Rachel Morin. Rachel Morin, a mother of five, was killed on the Ma & Pa Heritage Trail in Bel Air, Maryland, in August 2023. DNA from the scene matched no one in the offender database, but investigators linked it to an unsolved home invasion in Los Angeles earlier that year — same unknown man, no name. The Federal Bureau of Investigation (FBI) supported a genetic genealogy effort that produced a lead: Victor Martinez-Hernandez, who had no American arrest record to match against. He was arrested in Tulsa, Oklahoma, in June 2024, and his DNA was confirmed against the scene evidence. In April 2025 a jury convicted him on all counts, and on August 11, 2025, he was sentenced to two consecutive terms of life without parole plus forty years. The genealogy found a man no database ever could have. The jury verdict came from evidence, not from the family tree.

The Idaho student murders. On November 13, 2022, four University of Idaho students — Madison Mogen, Kaylee Goncalves, Xana Kernodle, and Ethan Chapin — were killed in their home in Moscow, Idaho. Investigators recovered male DNA from the snap of a knife sheath left at the scene. The FBI used investigative genetic genealogy on that profile, and the lead pointed to Bryan Kohberger. Police then confirmed it the traditional way, including DNA recovered from trash at his family’s Pennsylvania home, which showed a close male relative of the sheath contributor. Kohberger pleaded guilty in 2025 to all four murders and is serving consecutive life sentences. Again the pattern holds: genealogy produced the lead; confirmation DNA and conventional investigation produced the case.

What this method is not — and the limits worth saying out loud

A lead is not proof. This is the single most important sentence about IGG, and it protects real people. When a tree converges on a family, every innocent member of that family briefly sits inside the investigation’s frame — a brother, a cousin, a son who happens to share DNA with someone who did something terrible. The confirmation step exists precisely so that no one is accused on the strength of a family tree. Done properly, the public never learns the names the genealogy passed through. Trees contain errors; records contain misattributed parentage; a lead can point to the wrong branch. The method’s honesty lies in knowing that and building the safeguard in.

Consent matters. IGG only searches databases whose users have agreed to law enforcement matching. After public debate in 2019, GEDmatch made law enforcement matching opt-in rather than automatic. The method runs on volunteered DNA — people who decided their genetic information could help solve violent crime. That consent is the method’s legal and ethical foundation, and it should be.

It does not replace police work. Every case above ended the old-fashioned way: confirmation samples, warrants, prosecutors, juries or guilty pleas. IGG shortens the distance between a cold profile and a viable suspect. It does not shorten the distance between a suspect and a conviction. That distance is the justice system doing its job, and it should stay exactly as long as due process requires.

Where Othram fits

Much of the field’s hardest laboratory work runs through Othram Inc., a Texas laboratory built specifically for forensic cases. Consumer DNA labs were designed for clean saliva samples; Othram built its sequencing process for the evidence real cases actually produce — degraded, contaminated, decades old, sometimes vanishingly small. The company has helped law enforcement agencies resolve hundreds of cases, including identifications of victims whose names had been lost for decades, and it operates DNASolves, a database where contributors explicitly volunteer their DNA for this work. We cover Othram’s cases because they are, plainly, the standard the field is measured against. (Othram was not the laboratory in the Morin case — that lead came through the FBI’s genealogy effort.)

If you want to see the method up close

We wrote a full case file on the Rachel Morin investigation — the timeline, the forensic chain, what law enforcement did at each step, kept clearly separate from our own analysis. It is free, because the point of this work is understanding, and understanding should not have a paywall in front of it.

Get the free Rachel Morin case file

If you would rather have this kind of work arrive on its own, the newsletter is where we publish first: one case, examined honestly, no hype. You can unsubscribe in one click and we will not take it personally.