Why DNA Doesn't Always Solve Crimes

Added: 2026-06-04

[00:00:00]After Carla's clothing was sent to that lab in California, we then received a full profile off of Carla's bra.

[00:00:06]>> This profile was uploaded into Cotus and there were no hits.

[00:00:11]>> I hit what I like to call the Cotus wall. We were stuck and I was deflated and I thought, man, you know, all of this work and we still can't get there.

[00:00:19]This new technology and we still can't get there.

[00:00:22]In a lot of cases, you'll hear about DNA evidence being recovered and a profile has been built and searched through Kotus. And so then the question is why can't you take that profile and then bring it to a forensic genetic genealogy database and look for distant relatives?

[00:00:37]The reason in the Kotus database you're using DNA markers that are called short tandem repeats, STRs, and there's about 20 of them. With those 20 markers, you can detect a parent child relationship and perhaps a relationship between siblings. But you're not going to have enough markers to make those distant connections to distant relatives. In genetic genealogy, you're using a different kind of marker called a single nucleotide polymorphism uh or snip SNP.

[00:01:04]And in those kind of scenarios, you're collecting more like 2 3 400,000 different snip markers to have the amount of data necessary to detect those distant relationships.

[00:01:15]>> It's like the number of pixels on a picture. If you get a few thousand pixels, you see something blurry and maybe you can make out what's there. But if you get all of the pixels, then you're going to definitely make out what's there.

[00:01:29]>> Many of the people that watch this YouTube channel uh have probably already done DNA testing with a consumer DNA testing company that then returns a profile and a list of people you might be related to. And so you may be wondering why why can't we use that kind of approach for crime scenes? And the answer lies in the condition of the DNA.

[00:01:47]In consumer testing, the pamphlet that accompanies the kit says you can't eat or drink prior to testing. So, it's fresh DNA. It's high quality and it's from one person. Well, if you can't eat any food before taking this test, you can imagine when you do testing at crime scenes, you're facing the complete opposite set of uh parameters. We're talking about DNA that is so chemically damaged that the DNA is not recognizable in these areas as DNA. When remains are found out in the woods, they're not kept in the sterile environment the way you'd expect in a research lab or a medical testing facility. You've got very little DNA, sometimes just trace amounts of DNA. You've got DNA that's very old and degraded. It could be chemically damaged either because of exposure to the elements or treatment that was applied when the crime scene was preserved or they're overrun with DNA from other kinds of sources like bacteria. So forensic DNA is just so much more difficult to work with. in the kotus database the DNA is really degraded if it's chemically damaged sometimes you might get say 13 of the 20 markers or eight of the 20 markers and in that case scientists will call this a partial profile even if you get a partial profile using STR testing uh it does not mean that you can't get a better profile with this more modern kind of testing that relies on snip markers snip profiles can be built on very degraded DNA DNA that's been chemically damaged and so there's hope when investigators were trying to find Carlo Walker's rapist and murderer. They had originally recovered a small amount of DNA and sent it to testing. They built the Kotus profile with the 20 STR markers. That search revealed no matches. We're really stressed out as investigators. When you get excited, feeling like, hey, we've got a lead and it doesn't pan out, it really does affect you more than people realize. Then they had to go back to the original evidence to do a different kind of profile work, the snip profile. We were recommended a company to try and do genetic genealogy and the company that this was sent to pretty much consumed everything that we had and no results. So even after investigators have turned to forensic genetic genealogy, they continue to struggle to find the right process, the right lab, the right methodology to uh translate that original physical biological evidence into a digital profile that can be used to do genetic genealogy. And the problem is anytime you test DNA, you're destroying it. So you can only test DNA so many times before there's no DNA left. It's a big concern, especially in our field, because if we don't have answers, somebody down the road might be able to get them if they still have DNA left.

[00:04:23]>> Fortunately, the original lab that had found the DNA had located a secondary spot on the bra strap, but much less DNA, about 10 times less than they had found in the original sample.

[00:04:33]>> He assured us that the new processes that he had only used a minimal amount.

[00:04:38]>> We asked him to give it a shot. Uh it was really all that could be done. What else were they going to do to work this investigation? We don't have any other evidence that we feel like we can get DNA from, but we know this case is 46 years old. We thought this is our last shot, so we're going to do this.

[00:04:54]Fortunately, when they brought it to our lab, we used forensic techniques to basically handle these kind of challenging samples. And in the end, we were able to build a profile with hundreds of thousands of markers and almost immediately it led to the identification of the man responsible for Carla's death. That was a goosebump moment because I knew I was hearing something that detectives had wanted to hear for 46 years. We were extremely excited. We knew we were on the trail of our guy that had killed Carla. In the Tracy Sue Walker case, there had been so much work to try to piece together the clues as to who this young girl was. And in the end, none of those clues were able to lead investigators to an identity. After traditional methods were attempted back in 1985, lots of great investigators in those early days continued to try to carry the ball down the field and see if they could get this identification, but it just sort of stailed. Not because of lack of law enforcement work, but because of simply lack of technology. And it just grew cold. Then with the advent of COTUS, investigators sent the remains to build a DNA profile, which they were able to build and search Cotus, but there was no hit. And so the next step was to do forensic genetic genealogy. And in this case, the investigators sent this DNA off to a lab that has a decent reputation for doing STR and CODUS testing, but really hadn't yet fully onboarded the capabilities for forensic genetic genealogy.

[00:06:20]>> I thought this is maybe a way to be able to figure out who this person is. And so we submitted those remains to a private lab. And a few months later, I got the dreaded email of unfortunately due to the um degrading of the remains, there's no way we can get a profile to work with.

[00:06:36]>> There was a point where I thought, we've got to stop because we got to wait until the technology gets better. I was frustrated.

[00:06:43]>> Fortunately, special agent Elkins refused to give up. Eventually, we made contact and the rest is history. We were able to build a DNA profile that very quickly led to the identity of Tracy Sue Walker. To be able to get Tracy's name back and to be able to get the answers for the family is huge. Most rewarding thing I've ever done in my career.

[00:07:05]>> Even though you may have DNA available at a crime scene, it really matters how you build the DNA profile and not all the labs are equipped yet with the technology necessary to do that. So, one key part of our mission here at Oram is to make sure that we can upgrade all public crime labs so they can start taking on forensic genetics genealogy in their own labs.