Today, Explained - A police sketch based on DNA

Episode Date: October 25, 2022

Earlier this month, police in Edmonton, Canada, released a sketch of a suspect. The issue is, no one knows what the suspect looks like. This episode was produced by Siona Peterous, edited by Matt Collette, engineered by Paul Robert Mounsey, and hosted by Sean Rameswaram. Transcript at vox.com/todayexplained   Support Today, Explained by making a financial contribution to Vox! bit.ly/givepodcasts Learn more about your ad choices. Visit podcastchoices.com/adchoices

Earlier this month, the police department up in Edmonton, Alberta, Canada, released a sketch of a suspect. Just another day in police town. Except... No one knows what this suspect looks like. The sketch was based purely on DNA. Edmonton police are turning towards an emerging technology with hopes of identifying this suspect in a sexual assault case that has gone cold. The emerging technology is known as DNA phenotyping, using nothing more than DNA found in an investigation

to try and figure out what a suspect may have looked like. And within days of adopting this technology, the Edmonton police had reversed course. They took their DNA police sketch down and apologized to the public. Why? Coming up on Today Explained. It's a case of bad science and poor ethics. The all-new FanDuel Sportsbook and Casino is bringing you more action than ever. Want more ways to follow your faves?

Check out our new player prop tracking with real-time notifications. Or how about more ways to customize your casino page with our new favorite and recently played games tabs. And to top it all off, quick and secure withdrawals. Get more everything with FanDuel Sportsbook and Casino. Gambling problem? Call 1-866-531-2600. Visit connectsontario.ca. Today explains Sean Ramos from a heads up that the show that's ahead of us will feature some discussion of sexual assault.

But that's not really what it's about. We're here to find out why the whole world seemed to notice a decision made by the Edmonton police in Canada this month. It's because the police used something called DNA phenotyping to try and advance a stalled sexual assault investigation. They created a massive backlash that was centered on questions of racism, but also raised a lot of questions about forensic science and the limitations of DNA. Taylor Lambert is an investigative producer for the CBC in Edmonton. He's been following the story. So in early 2019, a woman was attacked by a stranger on the street and violently sexually assaulted. According to police, it was quite a brutal attack that left her with life-altering injuries. It was winter in Edmonton and so the man was bundled up and his face was covered. Essentially, the only identifying

details that the police had to go on was he was around 5'4", that he had some kind of an accent when he spoke, and that he was black. The investigation goes nowhere, and the police run out of leads. From what the detective told us was that they had so little to go on that they were just grasping at straws. And so they turned to phenotyping as basically their last resort Hail Mary long shot. It's definitely a last resort. And now for those who aren't in the know, what exactly is phenotyping? It's using DNA analysis to produce a computer-generated image of a person, but it's a highly controversial practice. DNA contains the building blocks of an individual, but it's not a blueprint. It's not that precise. So DNA is pretty good at predicting, say, your probability of hair color or skin color. there's just way too many complexities involved. And you can't make a scientific determination

of what a person will look like based on DNA alone. And so, in short, scientists say that it can't be done. But at least one company claims that it can do this behind closed doors. Which company is that? That's an American company called Parabon, which is really the go-to firm for this when it comes to law enforcement. Parabon Nanolabs of Reston, Virginia, has developed sophisticated software capable of unlocking information

contained in DNA left at a crime scene. As the program sifts through billions of pieces of genetic information, it slowly begins to build a suspect's appearance. So obviously, genetic research is valid science. There's experts in that field who do considerable research to advance our understanding of that scientific approach. But where it falls short is making all the countless determinations of probability that you would have to for all of the factors that go into a person's physical appearance in their face. So it's not just a simple thing like your hair color will be red or blonde or brown.

There's so many different complexities involved that scientists say you can't accurately predict what someone's face will look like. So what Parabon does is they say, we are able to do this. DNA is a blueprint for a person. It defines our hair color, our eye color, our skin color. All of that is written in the DNA. And the trick is just to figure out what parts to look at in order to make those predictions

and learn something new about a suspect just based on the DNA. Snapshot is highly accurate and has been successfully evaluated by major metropolitan police departments and federal agencies. But the technology that they use, the algorithms that they use are closed source. They're not peer reviewed. Scientists can't look at the data to determine how accurate they are. They can't look at the failure success rate, if you like, because all of that data is also proprietary.

It sounds at the very least that this is a developing science. How much does a process like this cost the Edmonton Police Department? The police told us that they paid Parabon $1,700 Canadian dollars, which is a little over $1,200 US dollars. Huh, so not that much. No. Does that make this common practice for police departments? site, they've posted a selection of some of these so-called DNA mugshots that they've done for various law enforcement agencies around the world, including many in the U.S. and a few prior cases in Canada. And just based on that catalog, it looks like this is a practice that's been employed

by hundreds of law enforcement agencies worldwide. Some of them do look a pretty good prediction of the person that was eventually caught. Some of them are not even close. A lot of them are not even close. So it seems like it's at best educated guesswork. I want to go look at these photos. Is it DNA nanotech? I'm wondering where to look. Yeah, I think it's snapshot.parabon.com. I got it. Okay, I'm going to this website. I see the Edmonton police's controversial phenotype up here.

Obviously, they haven't caught the suspect, so you don't get to see it before and after. A composite profile they did for Sylvania, Ohio, where their phenotype DNA snapshot looks kind of like the guy. But then I see one in Eugene, Oregon, where the phenotype looks nothing like the guy. So it looks just from this brief sample a little hit or miss. Yeah, for sure. I'm not an expert in police work, but it seems to me that this would be a less reliable tool than, say, a police sketch drawn from an eyewitness account of somebody. So perhaps, presumably, they can't solve the case in Edmonton. They decide to use this phenotype, and they decide that it'll be a good idea to tweet

out the photo because I guess they have to publicize it in order to get the composite out there. Is that what happens? That's right. They put out a press release with the photo. They post it on social media. They hold a press conference announcing their use of this technology. They seem to be fairly proud of it in a way. They're sort of trumpeting the fact that they're using this,

you know, cutting edge, if you like, technology to solve this terrible crime. How do people respond? Not well. The image is immediately excoriated online. People point out that this CGI head of a young black man could look like a lot of people in the community. And those innocent people might have the cops called on them and become suspects or worse.

I mean, we have data that show that in Canada, Black people are more likely to have to interact with the police, more likely to be incarcerated, and more likely to die at the hands of the police. While there was a lot of outrage about the use of this image, there was also a lot of concern expressed about the very real risks that this image would put on racialized communities. How did the Edmonton police respond?

They don't exactly apologize. One guy technically said that he was sorry, kind of, for something, but it's not a real apology. Canadians know a real apology. Two days after they had announced the use of this technology, they announced at another press conference that they are removing the image from their website and social media accounts. The potential that a visual profile can provide far too broad a characterization from within a racialized community, and in this case, Edmonton's Black community, was not something I adequately considered. But they also insisted that they were just trying to find justice for the victim. I prioritized the investigation, which in this case involved the pursuit of justice for the victim, herself a member of the racialized community,

over the potential harm to the black community. They're a little defensive of their use of this technology, and they didn't rule out using DNA phenotyping again in the future. Where does that leave this case, this violent sexual crime from 2019? Well, in the two-day span that it was out in the public, you know, the attention that was on it was certainly focused more on the question of whether this image should have been created and released in the first place more so than on trying to identify a suspect. But if the police did get anything out of it from that perspective, they haven't said. So where this leaves the case is unclear. But I think this case has raised a lot

of questions that resonated quite widely, not just in the city, but across the country and actually even outside of Canada. Questions about how the police view the Black community. Multiple people in the chain of command would have had to have seen and approved this image before its release. And another big question it raises is, what does it mean that the police are turning to unscientific means in their police work? And does the public understand, or for that matter, do the police understand, that forensic DNA phenotyping is not considered valid science? The science of DNA phenotyping, when we return on Today Explained. management software designed to help you save time and put money back in your pocket. Ramp says they give finance teams unprecedented control and insight into company spend.

With Ramp, you're able to issue cards to every employee with limits and restrictions and automate expense reporting so you can stop wasting time at the end of every month. And now you can get $250 when you join Ramp. You can go to ramp.com slash explained ramp.com slash explained r a m p.com slash explained. Cards issued by Sutton Bank. Member FDIC. Terms and conditions apply. Bet MGM, authorized gaming partner of the NBA, has your back all season long. From tip-off to the final buzzer, you're always taken care of with a sports book born in Vegas.

That's a feeling you can only get with bed MGM. And no matter your team, your favorite player or your style, there's something every NBA fan will love about that. MGM download the app today and discover why bed MGM is your basketball home for the season. Raise your game to the next level this year with BetMGM, a sportsbook worth a slam dunk,

an authorized gaming partner of the NBA. BetMGM.com for terms and conditions. Must be 19 years of age or older to wager. Ontario only. Please play responsibly. If you have any questions or concerns about your gambling or someone close to you, please contact Connex Ontario at 1-866-531-2600 to speak to an advisor free of charge. Today Explained, we're back and now we have Professor Dan Crane from Wright State University in Ohio with us. I have testified as an expert witness in cases involving DNA evidence since January of 1991.

Professor Crane knows more about DNA than your average true crime enthusiast. Now over 125 cases of 27 different states, several different countries, Northern Ireland, England, Australia, as well as a number of U.S. federal courts and courts martial. So we asked him what he thought when he saw the Edmonton police using DNA phenotyping to try and track down their suspect. Well, I think it brought to mind a pretty wide variety of thoughts. At one level, I'm thinking this is actually a fairly nondescript representation. It seems like it's a pretty broad target that a pretty wide range of people might be set to match. And yet, I think I also appreciated that that's consistent with what you should expect from that technology, these approaches at this stage in their development.

The concern then is that because it's got DNA associated with it, that people will give it more weight, more credibility than it probably deserves. And we need to be a little bit cautious about how it is that that sort of information is shared with the general public. Let's talk a bit about the backstory of this science. When exactly do we get this technology, DNA phenotyping? When does it make its debut? You know, I think you can go back maybe as long as 100 years. You know, we've learned an awful lot about human genetics through twin studies, you know, the studies of identical twins, especially when they've been separated at birth. And you and I are like, like sisters. Sisters. Hallie, we're like twins. And from that type of work, we've learned that quite a bit about what we are and how we look is determined by our genes.

For instance, about 80% of our height seems to be governed by what it is that we have in our DNA. But there are also some interesting behavioral things that you can glean from these identical twin studies, meaning that there's things in our DNA. For instance, that age-old question of boxers versus briefs, it turns out that there's a strong genetic component to that, that if you find a pair of identical twins that were separated at birth and you tell me one of them has a preference for wearing briefs, I'll put money on the fact that his other twin will also be fond of briefs as opposed to boxers. They're hens. Let's just leave it at that. Almost all of the traits that we might be interested in are what geneticists would call complex

traits. And that means fairly simply that they're determined not by just one gene, but determined by often dozens and sometimes hundreds of different genes, each of which has some say in what the final result is going to look like in an individual. And the sorts of traits, hair color, eye color, height, and then especially things like facial shape or skin color, those are actually things that haven't been studied that much by geneticists. Most of the federal research dollars is dedicated to resolving things associated with life-threatening diseases. And so, for DNA phenotyping, it's that sort of stuff that fell off the back of the truck that people are trying to piece together to get something useful for for this other purpose.

Well, because it isn't as vastly studied a field, I wonder how reliable this science is. You know, as somebody who gets involved in court cases where DNA evidence is involved, to my knowledge, the types of tests that are done for DNA phenotyping haven't even started to appear in the peer-reviewed scientific literature. So it's very difficult to argue that these types of findings would be accepted as evidence in a court of law. And that hasn't stopped police departments, at least the Edmonton Police Department, from trying to use it. Is this a first? Is this an anomaly? Or is this increasingly a method that police departments turn to? Well, for maybe on the order of 20 years now, going back to maybe around 2004,

I'm aware of a case that a United States crime laboratory used DNA phenotyping to assist in an investigation. Today, one of the major players in the investigation into serial killer Derek Todd Lee talked about the case that gripped the Baton Rouge area. She shared the major turning point in the case, a DNA sample that completely refocused the search for the killer. Pretty much up till this point in time it's been used as a method of last resort when it's a crime that's of particular interest to local law enforcement and they've exhausted everything else to generate leads they might then turn to seeing what they could get from this type of DNA phenotyping approach. There have been in the past couple of years a few laboratories that are starting to embrace it.

Edmonton obviously is one, but Australia, their law enforcement has done some kicking of the tires of a number of different approaches to doing DNA phenotyping, and they've settled on one that they now would like to use as part of their routine casework. But, you know, on the other side of the coin, there are countries that have very explicitly and specifically said, you should not do this type of testing. Belgium and Germany, for instance, have laws that very specifically prohibit it. You know, we're still in the early stages of how society wants to deal with this type of evidence. Were you surprised when the Edmonton police backpedaled and deleted this post and said, you know, they were sorry for sharing this composite image.

I'm trying to think what the best way to express this is, because I'm starting to feel like an old guy, right? Because I've seen this play out before. It seems very common, and maybe particularly in the area of forensic science, for there to be very enthusiastic proponents of a new methodology. And they're typically the developers of the new methodology. And they get out in front, and they do a good sales job, and they persuade others to get on their bandwagon. And then the questions start to get asked. And it's kind of like a pendulum swing. We have very enthusiastic proponents that develop these approaches. Everything looks like sunshine and roses. And then you start to hear some of the serious questions. And then you go back to the

other way and you may become overcautious in terms of not utilizing the information that you might be able to get from these new approaches. Do you think it's only a matter of time before DNA phenotyping technology improves and police departments around the world are using this as a more normalized way of investigating crimes they can't solve? I do think it's a matter of time before the work is done that would allow us to use it more reliably. How long that time frame is is a reasonable question. I don't think we're anywhere near there just yet, but maybe in five or ten years, we could start talking seriously about something that would pass muster in American courts at the very least. And until then, what should police departments in America and, I don't know, say Edmonton, Alberta, Canada, be keeping in mind?

Well, until then, I think they should sit patiently and wait. Until it's ready for prime time. Dan Crane is a biology professor at Wright State University in Fairbourn, Ohio. Earlier in the show, you heard from Taylor Lambert with the CBC over in Edmonton. Our program today was produced by Siona Petros. It was edited and fact-checked by Matthew Collette and mixed and mastered by Paul Robert Mounsey. Thanks for listening to Today Explained. Thank you.