Jun 14 2017

What Can be Learned from Non-Coding DNA?


Past ISHI Speaker, Gregory Wray, discusses what we knew about non-coding, “junk” DNA in 2014.



My name is Greg Wray. I’m a professor at Duke University, and for many years I’ve been studying the human genome, and I’m particularly interested in the function of non-coding DNA and the regulatory roles that it plays.

I’ve been very interested through most of my professional career in what the part of the genome that people used to call “junk” does and when I first started working on this, it was sort of a marginal field in biology. Most people didn’t really think there was much to study there. But, both working with the human genome and some other model organisms, I’ve been very interested in understanding what kinds of functions that so-called “junk” does and as we now know is quite a lot that it does. So, I’ve been interested in trying to figure out what sorts of traits are effected by non-coding DNA. I’ve been interested in finding out whether perhaps there are disease impacts from some of that variation, and that’s what got me interested eventually in forensics, because there’s been the presumption for many years that non-coding DNA doesn’t really do much of anything, and that perception is now beginning to change.

We know there are functions occurring outside of protein coding genes, so it’s not junk, but with the ENCODE project set out to do is to tell us exactly where that functional stuff is. Now, it’s not everything we need to know. We have some addresses, now we know this piece does something, maybe that piece doesn’t, but we still don’t know what those non-coding elements (they’re called), what they really do. For very few of them do we know, or there’s such and such trait that’s effected. That’s the challenge for the future. Taking the results of the ENCODE project and turning it into what you’d call actionable information. Information that could be used for diagnosis, for therapy, for drug development, and (potentially) in the realm of forensic science as a way of essentially figuring out a person’s characteristics from a crime sample. That’s something we don’t do very much right now. We usually just do matching. It’s identity matching, and that’s a great tool, but modern DNA technologies are allowing us to move more into a realm where we could actually take a crime scene sample and say, “This person – not only were they definitely male, but they were probably of European decent, and our guess is that they had blue eyes”, and that alone cannot only rule in and out potentially some persons of interest. So that sort of application is what ENCODE and medical genetics is now making possible for forensic science.

We’ve really moved away from thinking of DNA information as a fingerprint to thinking that it’s telling us something about a person. So, the CODIS markers when they first came into general use, they really were like a fingerprint. You couldn’t really say anything about a person. I think one of the take-away lessons from what’s been happening in the past 5-10 years is that we’re on the threshold of saying, “Look, it’s not a fingerprint anymore. It’s a fuzzy set of information. It’s a blurry picture that we’re getting about a person from the CODIS markers and certainly from the rest of the genome, but that picture is going to come into sharper and sharper focus in the coming years.”

So, don’t think of DNA fingerprinting as a fingerprint anymore. In fact, don’t use that word anymore. Think of it as genetic information that sometimes can help with identity matching, paternity, crime scene matching, but it also has the potential to tell you something about a person’s traits.