Making whole genomes from eDNA

Phil Sansom and Sally Le Page spoke with Eske Willerslev from the University of Cambridge and Beth Clare from the York University, Toronto, about combining these tiny fragments of eDNA into entire genomes...

Phil - Eske, part of what you've been doing is moving from fragments of DNA from each animal to sequencing entire genomes, all the DNA from one creature. How do you get an entire genome, a lot of DNA, from these samples?

Eske - For many years, it was what is known as metabarcoding, l ike what you just experienced here, where you are pulling out specific regions of the genome that you know are quite informative in terms of doing some species identification, but obviously just look at the human genome, right? Most of the information that is being used to understand, I mean, not only what species it is, but also this particular individual, how is that related to other individuals? And also we can start looking at selection, for example, what genetic regions are important for survival. But for all these things you need other parts of the DNA. So what we started doing was to go from the so-called metabarcoding to simply shotgun sequencing. That's where you just sequence randomly all the pieces of DNA in your environmental samples. I mean, the challenge is you need a proper reference for comparison, right? And in some situations they are not present and you have to go out and create those references yourself. But what we did this year was that we managed through these many pieces of DNA to assemble it together into genome-scale sequences. And we did that for two bear species that lived 12,000 years ago.

Sally - That DNA must be very old and broken apart. The pieces must be tiny. This is like the hardest jigsaw puzzle in the world. You've got these tiny, tiny pieces and you're sticking them together and you don't even know what the overall picture looks like. How'd you do it?

Beth - Metaphorically using the puzzle metaphor, you've got a picture of what you think it should look like. And you've got all these pieces and you assemble it on the picture to try and fill in the gaps to see what's there. And it's enough to do things like population genetics to figure out the diversity of a population in an area, that's a relatively easy challenge for eDNA. So, you may not get the complete picture, but you get enough of it to be really scientifically useful for understanding how fragile a population is or how important a landscape is, in terms of what is living on it. And that's where the real application of eDNA comes in. It's monitoring populations effectively in real time when you can't and don't have the manpower to go out and sample them by looking for the individuals. You just sample the environment they're in to learn a whole lot about who they are, where they came from, how many there are, what their diversity is, what their origins are. I think we can do that with modern day live animals that are still breathing in their environment, the same way that Eske can with historical records.

Sally - How would this tie in with DNA from the air? Do you think being able to get whole genomes from the air; is that going to be possible?

Beth - Oh, I think so. There's different challenges with air. So, things like UV light are probably going to be damaging DNA in a way in air that you don't have in soil, but it's not insurmountable.