Millions of people around the world are purchasing DNA testing kits and uploading their results to genealogy websites to trace their family trees. Police services are making use of this genetic information too, looking for matches with DNA samples left at crime scenes to solve cold cases. Commissioner Kosseim speaks with renowned DNA forensics expert Dr. Frederick Bieber about the use of these investigative techniques by police and the potential privacy risks for people who participate in recreational genealogy, as well as their family members.
Dr. Frederick Bieber is a medical geneticist at Brigham and Women’s Hospital in Boston, Massachusetts and an associate professor of pathology at Harvard Medical School.
Resources:
Info Matters is a podcast about people, privacy, and access to information hosted by Patricia Kosseim, Information and Privacy Commissioner of Ontario. We dive into conversations with people from all walks of life and hear stories about the access and privacy issues that matter most to them.
If you enjoyed the podcast, leave us a rating or a review.
Have an access to information or privacy topic you want to learn more about? Interested in being a guest on the show? Send us a tweet @IPCinfoprivacy or email us at podcast@ipc.on.ca.
Info Matters Podcast S2 Episode 2: Family ties: Using investigative genetic genealogy to solve crimes
Guest: Dr. Frederick Bieber, medical geneticist and Harvard professor
Patricia Kosseim:
Hello, I'm Patricia Kosseim, Ontario's information and privacy commissioner. And you're listening to Info Matters. A podcast about people, privacy and access to information. We dive into conversations with people from all walks of life and hear real stories about the access and privacy issues that matter most to them.
Hello listeners, and welcome to another episode of Info Matters. You've often heard about them, maybe even purchased one yourself. Consumer DNA testing kits have exploded in popularity in recent years, thanks to heavy online marketing and TV ads that encourage you to learn more about your heritage. In 2019, it was estimated that more than 26 million people worldwide had taken an at-home DNA test like Ancestry or 23andMe. And global sales of these kinds of consumer DNA testing kits are expected to grow another 12.5% from now to 2026.
More and more people are choosing to upload their DNA test results for free on genealogy websites. These websites allow people to compare their results with that of millions of others to look for matches. It's a way of finding long lost family members, constructing family trees and learning more about where you come from.
While the vast majority are genealogy enthusiasts who use these websites for recreational purposes, others, like police services, have started using them for investigative purposes to help solve cold murder cases by looking for patches with DNA samples left at a crime scene. The Golden State Killer was the first example of a decades old case that police in the US were able to crack by searching a DNA profile against a consumer genealogy database. Since that first case in 2018, police services in Canada have also started using these techniques.
While investigative genetic genealogy has ultimately led to the arrest and conviction of society's most horrible criminals, it raises serious privacy issues, particularly for innocent family members who unwittingly get caught up in the web of a murder investigation as possible suspects. They become subject to surreptitious police surveillance for no other reason than the fact that they may be a distant biological relative of the actual murderer.
What kinds of ethical controls should be in place around police use of these techniques to ensure that broader societal goals of public safety are met without unduly infringing on the privacy rights of innocent individuals? My guest for this episode is Dr. Frederick Bieber. He's a medical geneticist at Brigham and Women's Hospital and a member of the faculty of medicine at Harvard. Originally from Canada. Dr. Bieber is an internationally renowned expert on forensic DNA and a member of Canada's National DNA Data Bank Advisory Committee, a committee I also served on for a number of years. Dr. Bieber, welcome to the show.
Dr. Frederick Bieber:
Thank very much commissioner. I'm happy to be here.
PK:
So, to help get things started, Dr. Bieber, or Fred, if I might call you so. Can you tell us a little bit about yourself? How did a boy from the Canadian prairies become one of the world's leading pathologists in the field of forensic science?
FB:
Well, I think it was a bit of good luck. After my parents moved to the states, I was able to study at universities here in the US. And actually, it was a collision with my English professor during a snowstorm in my freshman year where he said to me, "My good man, what are you reading these days?" I said, "Well, my assignments, of course." He said, "That will never do. Here, take this book. It's a New York times bestseller. I want you to read it and meet me for tea next week."
And the book happened to be The Double Helix written by James Watson, who discovered the structure of DNA. And I was so fascinated by the book I decided I better study genetics. And so, that chance collision in a snowstorm really changed my life. And I've had a lot of good luck along the way to have great professors and mentors. And I've been here now at Harvard as a faculty member for 42 years. And my students are smarter than I am, so I learn a lot from them every day.
PK:
And we learn a lot from you as well. In simple terms, Fred, what is this new emerging field of investigative genetic genealogy and what are some real life applications that you're seeing?
FB:
Well, let's remember that we inherit all of our DNA from our parents. Half from our mom, half from our dad. And so, we share a lot of DNA with all of our ancestors and with our close relatives. Our siblings, our children, if we have them, our cousins, our uncles, aunts, nephews, nieces, and so on. And the modern way of DNA typing looks at many more different locations along what we call the genome. We're not just looking at 46 chromosomes anymore to see if we have our male or female. We're looking at tens of thousands of these small markers that we call SNPs, S-N-Ps, that stands for single nucleotide polymorphism for changes. And most of our DNA is very much the same because our genes encode the important proteins that create our structure and also hormones and all the important proteins that we have. But a lot of the DNA is variable person to person.
And we inherit these variations from our mom and dad when they make sperm and egg cells. And by comparing them, we can look for regions of similarity between two DNA samples. We could get DNA from the skeletal remains and type it for these tens of thousands of markers. Again, that we call SNPs. And we could compare it to people in a family that may have lost their cousin or brother. So, it could be a missing person's investigation. We do this after mass disasters, plane crashes, fires, floods, and tsunamis, where there are a lot of victims of natural disasters. So, in many instances, we're using these technologies to identify people for humanitarian purposes.
But there are other applications as well, as you point out. If there is a criminal investigation after a homicide or in many instances, a sexual assault, and we have no suspects, no particular suspects. One could obtain the DNA, isolate the DNA from the crime scene. It could be blood on a knife. It could be blow back from a firearm, discharge blood from the discharge onto someone's clothing. It could be the so-called rape kit that our nurses and doctors obtain in the emergency room after a sexual assault. And we could compare the DNA results either to persons of interest, persons who might be suspects in the case.
But if there are no suspects that are obvious to law enforcement, the results can be uploaded into one or more of these databases that collect DNA and look for commonality, for close relationships, genetic relationships up to about fourth degree relatives. And by that I would mean cousins, for example, are what we call third degree relations. And it could be a first cousin once removed or a great aunt or a great, great grandparent. Someone that distantly related, if there's enough areas of similarity between two profiles, one in the database, and one from remains or blood or semen from a rape kit, we could infer that these two individuals are related.
And the person who's related is usually not the perpetrator of the crime, but it could be a close relative. At that point, once the DNA comparisons are made, then those results are turned over to professional genealogists. And they have all sorts of tools, as you might expect, looking at birth records, death records, marriage court records, obituaries to find who are the relatives, and they can recreate family trees based on publicly available information. And if they are successful, they may find that, "Gee, there's a woman in Alberta who's closely related to the DNA that we found at the crime scene."
And even though she may be of the wrong age and wasn't in the location where the crime happened, we could ask the genealogist, "Does she have relatives who were living or are living in the area where the crime happened?" And through their careful work, they can turn over a name or names to law enforcement who then might go and start a surveillance of those persons or person and wait till he, or potentially a she, discards a Tim Horton's coffee cup or a cigarette on the street. They would surreptitiously collect that evidence, that discarded DNA, what we call discarded DNA, and extract the DNA from that and see if there's a match or not to the real crime scene evidence. So, it's a multi-step process.
PK:
Fascinating. I mentioned earlier the case of the Golden State Killer, which I understand was the first time this new field of investigative genetic genealogy was used in the US to successfully resolve a cold murder case. If you were describing that case to a neighbor, say, how would you explain it in simple terms step by step?
FB:
During a 10-year period between 1976 and 1986, there were over 100 residential burglaries, over a dozen murders and kidnappings, and over 50 sexual assaults that were unsolved in seven different counties in California. They were all linked together by the common DNA, but there were no viable suspects. Fortunately, one of the retired pathologists who had since moved to New Jersey from California, he remembered that he had some extra DNA left over from one of the autopsies, because he always collected two samples. And he had been reading about the new tools in genetics. He contacted some of the investigators back in California and they discussed the idea of doing this genealogy type search.
They were able to type that original DNA sample from the rape homicide case and typed it for, again, tens of thousands of these markers that we call SNPs. They turned their results over to a retired patent attorney, whose name is Barbara Rae-Venter. And she did her work as a genealogist, gave a name to the law enforcement investigators, a specific person named Mr DeAngelo, who interestingly enough had been a police officer at one point in his career.
They surreptitiously collected his DNA for a direct comparison to the old crime scene evidence and found a direct match. The courts then obtained a warrant or the judge issued a warrant for his arrest. At which time, the law enforcement was able to get a direct sample cheek swab sample from him, not just the discarded material, and compared it and found again, confirmed the match. His arrest happened, as you pointed out, in April of 2018. And since then, an explosion of interest has happened by law enforcement groups, especially in the United States and also to a somewhat lesser extent, but still there's active interest in Canada, of course, as well to address unsolved cases.
PK:
There have been several Canadian examples where this type of genetic investigative method has been used and most notably the Ontario case of Christine Jessop that led to conclusive evidence, which identified the real murderer more than 30 years after the crime had happened, finally giving Christine's family the kind of emotional closure that they had long yearned for, for decades. And the case was quite high profile here as well, Fred, because it also led to conclusive exoneration of Guy Paul Morin, who had been wrongfully and tragically imprisoned for years for a crime he didn't commit. I think that's an interesting case because it really does show some of the important societal benefits that this kind of investigative technique can lead to.
FB:
Well, indeed. That particular tragic case is important in so many ways. One, to bring some level of closure to the family of this poor little girl who was murdered. It turns out the person who was responsible for her death, himself, is deceased. But as you point out, the exoneration of someone who was incorrectly accused and incarcerated is a key element of justice in our system. And despite the best intentions of our justice system, there are indeed people who are incorrectly accused and convicted. Many before the era of DNA testing, where well-intentioned jurors and well-intentioned witnesses simply get it wrong.
We know, for example, from a lot of the post-conviction exoneration's that have happened in the Americas, that many times, one of the most fallible types of evidence is eyewitness testimony. And it turns out jurors, understandably, want to hear the witness say, "It was him. I'll never forget that face." But oftentimes, lighting conditions are bad, stress affect someone's memory and well-intentioned witnesses often get it wrong. They identify a person who they said committed the crime, or they sought the crime committed against another person. And they're simply wrong. They misidentified somebody.
But jurors, understandably, are often swayed by that compelling evidence and they vote to convict. Or the court, if it's before a judge or a magistrate, he or she decides that a conviction is in order. And there have been now hundreds of individuals, mainly men, often men of color, who have served decades in prison for crimes they did not commit. And DNA has been what has helped exonerate them if biological evidence was part of the case.
PK:
So Fred, how widespread is investigative genetic genealogy being used by police today?
FB:
At this moment, I'm aware of over 500 investigations that are going on in north America. Some of them are identification methods, a missing persons IDs. There are two major websites that law enforcement uses now to do this investigative genetic genealogy. One is called GEDmatch, G-E-D match, that's owned by a company called Verogen. The other is Family Tree DNA. And they both have, I think Verogen has almost 2 million samples in their database where people have voluntarily uploaded their own profiles.
For example, if you bought a kit in the grocery store or in the pharmacy for 23andMe and you got your results from 23andMe and used them for whatever purpose you intended, you could then choose on your own to upload those data onto GEDmatch, as a way of finding more of your relatives that might have people who might have uploaded their profiles to GEDmatch. GEDmatch has an opt-in, opt-out option for you, where you could allow law enforcement to use those data for criminal investigations, where you could say, "No, I'm opting out. I only want to use it for my own personal purposes or research purposes," for example. And according to my recent conversations, about 75% of the people who upload their DNA to GEDmatch, elect to opt in. They allow law enforcement to use those data. And about 25 to 30% do not.
I think it's important for people who are thinking about participating in these projects, could be recreational, could be a birthday present you got, or a holiday present you received from a family member. I think before you do that, you should really take a moment or two or three to read the fine print. Have a conversation with your own family about it. Because as I tell my students here at the U, when you give your DNA sample somewhere, Patricia, you're giving half of your brother's DNA, half of your fathers, and half of your moms, half of your children, one eighth of your cousins, because cousins share about one eighth of your DNA by virtue of common ancestry. If you're an identical twin, to take this to the logical extreme, you're giving all of your identical twins DNA to somebody, whether it's a research team studying cancer at the Ontario Cancer Institute, or whether it's for a school project, or if it's for 23andMe. You're giving all of your twins DNA and they may not know about it. So, the family connections are obvious and deep. And I think a lot of people don't think about this and I think it is important to do that.
PK:
That's really important. Let's unpack that a little bit. I had a look at the privacy policy of the GEDmatch genealogy website. For example, it contains about 12 pages of text, informing users that one of the possible uses of the genealogy data uploaded could include familial searching by third party, such as law enforcement agencies, to identify the perpetrator of a crime or to identify remains. And then as you said, it somewhere along the process provides them with an opportunity to opt in to such a use. But do you think that clause or that information is enough to fully inform users of the risks associated with, say, getting implicated in a law enforcement investigation or implicating your family members in such an investigation?
FB:
I'm glad you asked the question because it is important. Let's take the example where the amount of DNA that's in common with crime scene evidence suggests that it's a third degree relative. And those data are turned over to a genealogist and say, "Let's find anyone in GEDmatch or Family Tree DNA, either company, that could be related as a second or third or fourth degree relative to this DNA profile." And they give a list of names to the genealogist. And these are names of people who voluntarily put their DNA in there. And the genealogist gets busy and he or she constructs a five or six generation pedigree that has a hundred people in it. And that's fairly easy for them to do because again, public records, obituaries, et cetera.
So, they put together this long, wide and deep pedigree, three, four generations deep and half a mile wide. And they turn this over and they say, "Oh my gosh, in order to narrow this down somehow, we need to go secretly get DNA from this person or that person. We know they didn't commit the crime. They had nothing to do with it. But if we rule them in or out as closely related, then we can eliminate that whole side of the pedigree and focus on the other side of the pedigree."
Well, sometimes Patricia, those people that they want to go collect from, they could be elderly in a nursing home. Are they able to give informed consent about giving a cheek swab? You approach someone and say, "We think in the greater family that you may not even know, you may not have ever met them, might be a suspect in a crime. Do you mind us taking a cheek swab from you?" But instead of doing that, they may just wait till that person drops a cigarette or drops his Tim Horton's cup outside the market. And they may be collecting that surreptitiously without a warrant, without permission.
And that's where I have a bit of a problem. And I've made this discussion with our FBI and with other law enforcement agencies that they really need to think carefully about how they surreptitiously or clandestinely collect these samples that are really elimination samples. They don't think that the third cousin committed any crime or the old person in his rocking chair in a nursing home, but it would help them with the investigation to rule in or rule out, it's one side of the family or another. And this happens quite often when the pedigrees get wide and deep. So, that's a very important matter that hasn't really been wrestled with, in my opinion, well enough.
PK:
That is an issue that concerns me as well, because as a registered user, say, of one of these platforms, you may consent and opt in to having your sample used by law enforcement, but your family members, the innocent ones, as you say, who are part of the large pedigree and are being surveyed surreptitiously in order to be ruled out, certainly didn't consent and have no knowledge or information, or any kind of insights into what is happening. These innocent family members as genetic suspects or informants.
I think the examples you give really shed a light on the privacy implications for those family members. But there are more than privacy implications for family members. There have been cases where family members who have been approached by police for questioning about a father or brother or a cousin they never even knew they had, which results in sometimes some surprising and inadvertent disclosures that can completely disrupt families and upend their lives as they once knew it. Now you've had experience with some of these inadvertent disclosures. Can you describe some of the situations that have resulted from family members being approached by law enforcement about family they never really knew they had or were related to?
FB:
The most striking example I can tell you about is a colleague at work told me that both he and his sister sent their DNA off to one of these companies, but they discovered in the process that they weren't siblings and their parents were dead. They didn't realize till then that cousins then came forward, once they announced this to the greater family, that his older sister had been adopted because the parents had a fertility problem. And they used to refer to him as the miracle baby, because they had gone to a fertility doctor to achieve a pregnancy that resulted in his birth. And then he started getting emails from people through this website saying, "We share half of our DNA. We're related." And it turns out, at last count, Patricia, he has 50, five zero, half sibs.
It turns out that the fertility doc that his parents went to... And he didn't know anything about this, his parents are now dead, that went to in Philadelphia, Pennsylvania, turns out used his own semen to impregnate his patients as part of the fertility treatment. And all these children born, or many of the children born are half sibs, half siblings. They shared this doctor as their biological father. And I asked him, "How do you feel about this?" And he said, "I'm still... I have post traumatic stress disorder finding this out. Some of the half sibs at a meeting we had together, were happy because they were only children and now they feel like they have a big family. But others are emotionally wrecked because we didn't know if our parents knew that the doctor wasn't using our father's material."
So, sometimes the news is welcomed and interesting and helpful to families. But other times, as you point out, it can be quite disconcerting, if not disruptive to family relationships. And not all surprises are pleasant ones.
PK:
And it really speaks to this notion of a right not to know, in some cases, particularly for family members who choose not to upload their DNA or get their DNA test results in the first place, let alone upload them on a genealogy website. So, I think there are many privacy and ethical issues that arise for those family members who are not direct participants in these recreational websites. So Fred, what are some ethical safeguards that can be adopted to ensure that these techniques and technologies are used and governed responsibly?
FB:
Well, again, the first success in the US of investigative genetic genealogy was in 2018 with the arrest of Mr DeAngelo. And by September of 2019, just a little over a year later, our US Department of Justice proposed what was then called an interim policy on forensic genetic genealogical DNA analysis and searching. The interim policy provided some guidance to law enforcement and crime labs with regard to oversight, which type or category of cases it could be used for in order to narrow the search for possible perpetrators of the crime.
The policy recommended obtaining a warrant for the actual DNA or SNP analysis of the discarded DNA, but only after it had been surreptitiously collected, and I would add without consent. And this sequence of steps seemed reasonable to the Department of Justice, but seems not entirely logical to others. And this is something that is sort of an active discussion now in the states, whether or how to obtain permission to collect these surreptitious samples without consent from distant relatives who they know had nothing to do with the crime. Simply to rule in or rule out whole categories of the family tree.
PK:
So no doubt, a very active policy debate going on in the US and we will watch closely to see how this interim policy evolves over time and what the experience is, in order to inform similar governance processes here in Canada.
Fred, once again, thank you so much for joining us on Info Matters and helping us get the conversation started. Your expertise in this area is truly unparalleled and we're extremely fortunate to have you join us on the podcast. You've helped explain a very complex science in a very simple way. Investigative genetic genealogy is a rapidly evolving field with new tools and technologies that can be used for humanitarian purposes, as well as for criminal investigations. And you've definitely given us lots to think about.
For listeners who want to learn more about privacy access and protecting personal information in the context of law enforcement, I encourage you to visit our website at ipc.on.ca. You can also call or email our office for assistance and general information about Ontario's access and privacy laws.
Well, that's it folks. We've reached the end of another episode of Info Matters. Thank you for listening and until next time.
I'm Patricia Kosseim, Ontario's information and privacy commissioner, and this has been Info Matters. If you enjoy the podcast, leave us a rating or review. If there's an access or privacy topic you'd like us to explore on a future episode, we'd love to hear from you. Send us a tweet at @IPCinfoprivacy, or email us at podcast@ipc.on.ca. Thanks for listening and please join us again for more conversations about people, privacy and access to information. If it matters to you, it matters to me.