OM wants to use private DNA databases from the US to solve cold cases

The police and the judiciary want to use private, genealogical DNA databases to solve deadlocked murder cases. Justice will file a claim with the examining magistrate later this month asking for the use of the American companies GEDMatch and FamilyTreeDNA in two murder cases. The use of private databases would be a first. The top of the Public Prosecution Service has approved its deployment.

The deployment is part of a pilot project intended to clarify the role that DNA databases can play in solving serious criminal cases. That is what national forensic public prosecutor Mirjam Warnaar and DNA expert Lex Meulenbroek of the Netherlands Forensic Institute (NFI) say to NRC.

Hundreds of cold cases have been solved in the US using private databases

The use of private, genealogical databases in America has solved hundreds of cold cases, some of which were decades old. The method gained public prominence through the unmasking of the Golden State Killer, a serial killer responsible for a string of rapes and murders in the 1970s and 1980s. The perpetrator was found in 2018 via a match in a private DNA database: a former police officer. Investigators and forensic specialists wanted to quickly deploy the method in the Netherlands, and are now being given permission by the top of the Public Prosecution Service to submit this to the court.

A gold mine

People upload their DNA profiles to private databases, for example because they want to learn more about their family history. Thanks to the technology used, profiles in the database can be linked to even very distant relatives: for example fourth or fifth degree cousins, or further.

Combined with genealogical research, the databases are therefore a gold mine for investigative services: a perpetrator can be traced through a few very distant relatives. It is not known how many Dutch people have uploaded their DNA profiles, but even without the presence of those profiles, Dutch perpetrators can be traced, for example via the European ancestors of Americans.

In the past year and a half, two Dutch murder cases have been selected in which the method will be used first. A case revolves around a hitherto unidentified victim who was violently killed, and whom the Public Prosecution Service hopes can be traced through the databases. In the other case, the judiciary wants to find out the perpetrator of the crime via the private databases.

Track to the culprit

It is emphatically a means of investigation, according to the Public Prosecution Service, and not a means of evidence. If a suspect is found using the method, regular DNA testing will still have to be done to see whether the trace of the perpetrator found at the crime scene, for example, matches that person.

Illustration Dewi van der Meulen

The use of private databases is sensitive. It is a collaboration with private organizations. Most of their clients did not initially upload their DNA profile to solve crimes. To avoid that problem, the Public Prosecution Service says it only cooperates with the databases GEDMatch and FamilyTreeDNA. At these companies, users can explicitly indicate (via a so-called ‘opt-in’) whether their profiles may be used for police investigations. At GEDMatch and FamilyTreeDNA, approximately 1.7 million people have made their profiles available in this way to, among other things, solve murders.

The Public Prosecution Service concludes that the law allows the use of the drug. “We as the Public Prosecution Service believe that there is enough legal basis to use the databases,” says national forensic prosecutor Mirjam Warnaar. “This is allowed by law and that is what we stand for. The law does not specify what kind of DNA relationship testing you are allowed to do.” DNA expert Lex Meulenbroek is hopeful. “These are new developments that many relatives and survivors who struggle day in day out with their suffering have pinned their hopes on. For that reason alone, this pilot is something we really need to do.”

Golden State Killer

In the American case that marked the breakthrough for the use of the databases, everything had already been tried. Witnesses had been interrogated, composite drawings had been made, dozens of detectives had worked on it for years. There were many DNA traces of the perpetrator, but that did not match a person in the DNA databases of the US police. Who the Golden State Killer was, the man responsible for dozens of rapes and thirteen murders in the American state of California in the 1970s and 1980s, remained unknown for years.

Until American detective Paul Holes realizes something in 2017. Private DNA databases are then emerging in America, in which people upload their DNA profiles because they are curious about their family history. And if only a handful of those people match the serial killer’s track record—because they’re distant relatives—then they could find the Golden State Killer. It is decided in the deepest secrecy to take the gamble.

And it works. Through the private DNA database and family tree research, a number of people emerge who could be the Golden State Killer. One is an elderly former police officer, Joseph DeAngelo. If the police then compares his DNA profile with that of the DNA traces of the perpetrator, there is a hit. He is sentenced to life imprisonment.

Twenty year old murder

The arrest of the former cop sparks a revolution in the forensic world. In the US, a breakthrough is then forced in hundreds of unsolved cases via private DNA databases. In Sweden, the investigative method will lead to a confession in 2020 from a suspect in one of the most poignant murder cases in Swedish history, that of eight-year-old Mohammed Ammouri and teacher Anna-Lena Svensson sixteen years earlier, in 2004. In Norway, the technique also leads to success. At the beginning of this year, the police solve a more than twenty-year-old murder of a 71-year-old linguist there.

“In the Netherlands, two private databases have therefore been selected,” says Warnaar. “GEDMatch and FamilyTreeDNA are private databases, where people can explicitly indicate whether they want their DNA profile to be used for criminal investigations. That was very important to us, which is why we chose only these two databases.”

They are American companies. How will that help you in Dutch cold cases?

Meulenbroek: ‘The vast majority of people in these databases have North-Western European roots. Although a large proportion of the participants are Americans, they often have European ancestors. The strength of these DNA databases is that the profiles in them have enormous clout. Fourth or fifth degree cousins, as well as beyond, can be found. Through these databases, there is a good chance that you will find distant relatives, and if you are lucky, close relatives.”

What are the unsolved cases?

Warnaar: “In this pilot we want to investigate the value of the databases in two types of cases. These are serious criminal cases where we want to try to find the perpetrator. And murder cases where we do not know who the found dead victim is, and we want to try to find out via the databases. These are completely deadlocked cases, which are very serious.”

How does that work?

Meulenbroek: “The plan of action is already ready. The DNA profile that we want to upload in the private databases is being drawn up in collaboration with a Dutch laboratory. The file of this profile is a long letter code that we then upload to GEDMatch and FamilyTreeDNA. Uploading is anonymous. We will then see who the profile matches. From these individuals we get information about the degree and category of relatedness, based on the amount of DNA that matches.”

And if you found, say, four distant cousins, then what?

Meulenbroek: “We then have to find out who their common great-great-grandfather is, for example, and then work that family tree all the way down to the present, to arrive at a possible perpetrator.

Warnaar: “That research will be carried out by the MEB, the Center for Family History, a government service that promotes genealogical research and family history. It will be digital and paper research, in archives, family trees.”

It is of course possible that those genealogists get stuck, and that they have to approach people, for example to request their family tree.

Warnaar: “In principle, we are not going to approach anyone. If the researchers get stuck, there is a new situation. Then a public prosecutor would have to give permission to contact those people.”

Suppose the family tree research succeeds, and a number of possible suspects roll out, what happens then?

Meulenbroek: “I expect that eventually there will be one or at most a few. In principle, you then have a regular criminal investigation. So then the police can get to work.”

Warnaar: “For example: which of these people is already included in the file? The family tree research can also lead to labeling someone as a suspect, arresting someone and taking their DNA. Then it becomes clear through regular DNA testing whether someone’s DNA matches the perpetrator’s trail. This new method will therefore play a limited role for evidence in court. It is a means of investigation, not evidence.”

But can it do what you want?

Warnaar: “There will undoubtedly be a discussion in court about this, and that is only good. We believe that the legal basis is really sufficient. You have to stand for that. We at the Public Prosecution Service are convinced that this is allowed. The law does not state what kind of DNA profile you are allowed to make, and the law does not specify what kind of relationship testing you may or may not do. A question that a judge can ask is: have the interests of a suspect been violated? I do not think so.”

You say: it is not in the law that it is not allowed. So it’s okay.

Warnaar: “The legislator has deliberately stayed away from boxes and categories about DNA. The legislator has taken a decision in principle that kinship testing is allowed and has seen no reason to put all kinds of boxes in it. We think it is allowed.”

Nevertheless, you are going to upload a DNA profile of a possible perpetrator and a victim to a commercial company.

Meulenbroek: “We do that very carefully. We do not send DNA to these organizations, but upload the DNA profile of the unknown person using a code. After we have received the results, we remove the profile from the database. We do not store DNA profiles in that database.”

The more people upload their DNA, the greater the chance that cases with a DNA trace will be solved. Do you also encourage that?

Warnaar: “I would never really do that. That’s not our role. People make their own decision whether to hand over their DNA profile to a private DNA database. We don’t want to push anyone.”

Meulenbroek: “Participation in these databases is something that people should carefully consider for themselves. We also do not want social pressure to arise to do or not to do this. With this drug, there is not the social pressure that people can experience during a large-scale DNA relationship study, where people are asked by the judiciary to provide their DNA for a population study, as in the Nicky Verstappen case, for example.”

We owe this to society

NFI employee Lex Meulenbroek

Mirjam Warnaar and NFI member Lex Meulenbroek have noticed in recent years that this subject plays a major role among relatives of murdered people and those left behind of missing persons. They see that there are developments that increase the chance that the murderer of their father, mother or sister will be found. “This new method offers opportunities in cases that have completely stalled,” says Meulenbroek. “We owe it to society to try out such new developments, if only for all those people who have pinned their hopes on this.”

ttn-32