How police are using DNA from ancestry websites to catch murderers

“Technology is neither good nor bad,” wrote the late historian Melvin Kranzberg: “[but] nor is it neutral.” We find it hard to predict all the ways in which technology can be used and the effects those different uses will have. The double-edged nature of pretty much every technological advance since his death in 1995 is testament to his wisdom. The latest such advance to make the headlines is called investigative genetic genealogy (IGG), and combines DNA analysis with traditional genealogical methods to solve crimes.

The technique first came to widespread public attention four years ago, when it helped to identify Joseph DeAngelo as the so-called Golden State Killer who committed 13 murders and 51 rapes in California between 1974 and 1986. DeAngelo was only caught after distant relatives of his were identified on GEDmatch, an online DNA and genealogy database. The year after his arrest, 2019, only 28 cases worldwide acknowledged the role of IGG in generating the initial lead: the number is now more than 200 and set to rise.

On the face of it, IGG is a good thing. American cases solved by a mix of DNA testing and genetic genealogy this year alone include the hitherto unsolved murders of Nancy Bennallack (stabbed in California in 1970), Anna Kane (strangled in Pennsylvania in 1988), and Stacey Lyn Chahorski (strangled in Georgia, also in 1988.) In all three cases, however, the killer has also since died, meaning that justice delayed is to some extent also justice denied.

We think of unsolved crimes as ‘cold cases’, and they are a source of enduring popular fascination. A US drama of that name ran for seven seasons between 2003 and 2010: Waking The Dead and New Tricks have covered similar ground on the BBC, with almost 200 episodes between them. But Tony Blockley, former head of crime at Derbyshire Constabulary, says, “I don’t think that’s [cold cases] an appropriate term. They’re just unfinished cases. They’re constantly being reviewed: they’re not put on a shelf and just left. For those victims and the families of those victims that haven’t had final resolution, they’re still living those crimes on a daily basis. If it happened 40 years ago, it’s like it happened yesterday.”

Ethical concerns

But dig deeper into those cases solved at least partly by IGG and there are some serious ethical concerns. The genealogy sector is varied and increasingly crowded: companies include Ancestry, 23andMe, FTDNA, MyHeritage and Living DNA. There are estimated to be over 100 million people in these databases. There has been vertiginous growth on two fronts: the DNA databases on one hand, and on the other, the availability of genealogical records. Records dating back centuries from all around the world have been digitised, indexed and put online. Births, marriages and deaths, census records, electoral registers, newspaper articles and other historical records are all a few clicks away. What was once a laborious process involving distant travel and recalcitrant archivists is now as easy as an online supermarket shop, and with fewer substitute items to boot.

The vast majority of DNA data uploaded to such websites is done voluntarily by people seeking to know more about their own family histories, though whether they know exactly what they’re potentially exposing themselves to is another matter. There are parallels with the ways in which so many of us voluntarily give up our data to social media and shopping sites, with little control over the ways in which that data may eventually be used.

In the case of IGG, law enforcement involvement is the elephant in the room. Policies vary widely between sites. Some work with law enforcement and some don’t, unless compelled to by court order or similar: many have changed their terms and conditions more than once in the past few years. Standards and methodology vary too: companies use different ways of finding matches and set different thresholds for what they count as a match. As a result, the relationship predictions can vary from company to company.

Global impact

There is also a disconnect between the global nature of the databases and the national constraints of law enforcement. Someone in one country might upload a DNA test and find that a distant relative becomes involved in an investigation in an equally distant country – one, perhaps, with different human rights and sentencing standards, up to and including the death penalty. Anyone who tests is making a decision not just on their own behalf but all of their genetic relatives too.

“People using genetic genealogy databases for their own purposes never anticipated this kind of access to their genetic information or that information being used to identify people they’re related to,” says Amy McGuire, director of the Center for Medical Ethics and Health Policy at Texas’s Baylor College of Medicine. “There’s a genuine tension between wanting to protect consumers and be respectful of their wishes, and recognising that working with law enforcement provides a social benefit.” Surveys in the USA have found widespread approval for law enforcement using consumer DNA databases to solve violent crimes, but much less approval for non-violent crimes.

Scotland Yard announced last month that it is planning to run an IGG trial here, though so far details remain sketchy. Debbie Kennett, a genealogist and expert in DNA at University College, London, is not convinced that we need it. She points to major differences between the USA and here, particularly “the problems in their criminal justice system. They have huge backlogs of untested sexual assault kits and tens of thousands of prisoners who have not been added to CODIS [Combined DNA Index System, America’s federal DNA database].

“They’re also not making good use of the familial searching technique which was pioneered in the UK and involves looking for close matches to the perpetrator in the criminal DNA database. Only a handful of US states use familial searching in their state DNA databases and it’s not permitted in the federal DNA database. A significant percentage of the US cases could have been solved years earlier without resorting to IGG if they had dealt with all the systemic problems and used familial searching.

“The UK DNA Database is very efficient, and most serious crimes will be solved using existing conventional methods. The hit rate [percentage of profiles which get a match when uploaded] in the UK database is 66 per cent compared to just 35 per cent in CODIS. This, of course, doesn’t mean that IGG wouldn’t have a role to play in the UK, but the number of cases where it might be used is likely to be smaller – perhaps a few hundred cases compared to tens of thousands of cases in the US.”

The Biometrics and Forensics Ethics Group (BFEG), a non-departmental public body supported by the Home Office, concurs. “If a genealogy approach is used, a proportion of the potential relatives will not be UK-based and this could add significantly to the genealogical effort and cost. The approach should only be used if it can be shown to be based on clear evidence, verified by an independent body, that the established methods already in use for these law enforcement purposes are no longer adequate or effective. Otherwise, the use of any such novel processes would not meet the tests of necessity and proportionality.”

Complex decisions

So what should those who wish to upload DNA tests to these websites do? Firstly, check whether opt-in or opt-out is the default position, and this can vary not just between companies but even within them depending on where the consumer lives. The European Union’s General Data Protection Regulation, for example, obliges companies to make opt-out the default for EU citizens ( they can opt back in if they choose).

Critics say there needs to be a greater obligation on genealogy companies and the legislative framework around them to ensure openness, transparency and accountability.

“There is an urgent need for forensic scientists, bioethicists, law enforcement agencies, genetic genealogists and other interested parties to work together to produce international guidelines and policies to ensure that the techniques are used responsibly and effectively,” Kennett says. Addressing the issue at more macro levels also brings into play ethical debates surrounding public interest, societal good, state power and oversight mechanisms.

Kennett also points out the “ethical issues relating to the cases on which IGG is used. In America, they are using IGG to identify and prosecute women who abandoned babies decades ago. These women are being charged with homicide and put in prison, whereas in the UK we treat these cases as infanticide and are sympathetic to the women. The reversal of Roe v Wade also raises scary possibilities of IGG being used to identify and prosecute women having miscarriages and abortions.”

As ever, our capacity to do things runs ahead of our capacity to decide whether we should. What’s new is that, for the first time, the past is also the present, and the implications of that run far wider, longer and deeper than the simple solving of a decades-old murder case.