Staging a Canine Crime Scene to Teach Forensics Concepts

Sara Katsanis explains how she and her colleagues developed a mock wrongful convictions canine crime scene to teach forensics concepts to a middle school student.





I’m Sara Katsanis. I’m in the faculty at Duke University at the Initiative for Science in Society and I research the applications of DNA and genetic information in social contexts like medical testing and also DNA forensics.

So, I was approached by a middle school student last year who was interested in researching for her eighth grade final project wrongful convictions. We came up with a mock dog crime scene where we could use DNA from dogs rather than from humans to set up a mock crime scene where an individual would be accused of a crime and then we’d give this information to the student and see if the student could figure out whether that dog could be excluded from being a contributor of the crime sample evidence.

So the middle schooler was interested in wrongful convictions, so what we did was take DNA from multiple dogs in her community. So, the student got the experience of swabbing saliva from dogs (multiple dogs). Some of them she had to instruct the owner of the dog how to do the DNA collection, and some of those came back not so good – they had food in the swab, so there was some learning experience there in trying to understand proper sample handling and collection.

We took those samples and then we typed them for a lot of different markers. This laboratory we used at Duke happened to have sequencing technology rather than STR capillary electrophoresis, so we used the sequencing technology to look at the particular SNPs that were useful for traits. So, once we had typed those dogs that we had access to independently, we then selected which markers would be most useful and informative, and then we selected a dog out of that (we being the instructors) selected the dog, and the student didn’t know which dog. In fact, the other instructor didn’t know which dog either, so I was the only one that knew which dog would be accused of the crime.

So, we got this other dog and staged a rip up of a toy, of an item of evidence, and we had the dog who eventually was the contributor of the sample tear up this toy, or mock tear it up, and put the saliva all over the toy, and then we created a police report that mirrored what might happen at a crime scene.

In this particular case, we used a dog that looked fierce. The crime scene was at a dog park, and the dog that looked fierce and barked a lot and growled a lot at the dog park was accused of tearing up the toy. Not because she was seen tearing up the toy, but because she seemed scary. Maybe because the color of her fur is black. So we accused that dog of this crime, or the police report accused this dog, and the dog was banned from the dog park.

That’s the scenario we gave the student. So then the student had the opportunity to test the evidence. One day she extracted the DNA from the reference sample. Another day, she extracted the DNA from the evidence from the crime scene. And she was able to work out the amologen markers very similar to humans, but a larger difference from humans, so she could see it on an agorous gel, which was an advantage, so she could immediately exclude some of the contributors of the sample. Then she typed for multiple different traits using the SNP technology with the help of this laboratory that routinely does human sequencing analysis.


(Forensics) isn't as simple as it seems to be on TV. Click To Tweet


In the end, she was able to exclude all of the contributors, including the accused, except for this dog, Brisket. She reported how difficult it was to do the extractions, that it wasn’t as simple as it seemed to be on TV, that it was hard to keep track of the samples. The use of controls was something new for her. It was quite nerve-wracking when she got a pellet with a negative control and she was nervous that this was going to be a problem in the analysis. Some of the markers didn’t type the first time around. When she came up with the last suspect, and she said “Ah! Brisket did it!” I said, “How do you know Brisket did it? Were you there? Did you see it?”

So it was a good educational tool to reinforce how DNA forensics really works in crime scenes. That it can only exclude, and cannot ever convict an individual. Well, not ever, but in this case it couldn’t convict, because we don’t know if Brisket slobbered all over this dog before it was ripped up. We have no idea.