Sarah Sharman 00:00
Chris, welcome to Tiny Expeditions, Season Five, Episode Eight, the grand finale.
Chris Powell 00:09
It's the last one.
Sarah Sharman 00:10
I'm Dr Sarah Sharman, here to help you understand the science.
Chris Powell 00:13
And my name is Chris Powell. I'm your co-host for this great last episode of the pop cultured season. Tiny expeditions. We spent this whole season looking at how science has been portrayed in some of our favorite TV shows and movies. We've covered a lot of ground.
Sarah Sharman 00:26
We sure have Chris. We looked at the representation of genetic mutations and whether it's plausible that they could cause superpowers.
Chris Powell 00:35
You mean, like the…
Sarah Sharman 00:38
yes, go ahead.
Sarah and Chris 00:39
…the turtles.
Chris Powell 00:40
We also looked at Jurassic Park and the plausibility of resurrecting dinosaurs.
Sarah Sharman 00:45
And we spent two episodes looking at the movie GATTACA and some of the ethical ramifications behind gene editing and humans.
Chris Powell 00:54
So I have a thought now. I feel like we're creative people, Sarah.
Sarah Sharman 00:58
I would think so.
Chris Powell 00:59
I say, let's unleash the creativity.
Sarah Sharman 01:02
Oh, you want to write a movie script.
Chris Powell 01:03
Exactly. So for this last one, let's don't go to a famous movie. Let's make our own famous movie, and then follow the script we did for the other episodes, and we bring in one of our researchers, and we pitch it to them and see if they think this is plausible or not.
Sarah Sharman 01:19
That's great. Let's go to the writers’ room. All right, excellent.
Chris Powell 01:23
So we're in the writers’ room, and I would expect there to be like 10 people here, but apparently, it's just you and I.
Sarah Sharman 01:28
Yeah, sorry, low-budget film.
Chris Powell 01:30
That's fine. We got this. We're gonna do it. We need all of the basic building blocks of a story. We need plot line, we need characterization, we need conflict, we need resolution. Let's start with setting. Let's Where are we going to set this?
Sarah Sharman 01:43
Well, we're in Alabama right now.
Chris Powell 01:46
I see. So let's make it familiar to us. Let's just broadly set it in the United States. Okay, okay, so it's the United States, and let's just go straight for the conflict, right? It's the United States versus someone who's our greatest enemy right now.
Sarah Sharman 01:59
Let's think this year: North Korea, aliens, anybody else?
Chris Powell 02:04
No, you got to think bigger than that. Okay, Canada.
Sarah Sharman 02:09
Isn't Canada really nice, non-confrontational, kind of mind their own business?
Chris Powell 02:14
Yeah, you're talking about Canadians. Those are great people. Yeah, we've got no problem with Canadians at all. But what we're going to deal with is specifically the Canadian maple syrup reserve in Quebec.
Sarah Sharman 02:26
Did you just make that up?
Chris Powell 02:27
No, I promise. It's a real thing. It's a real thing. But for our film, it's going to run out. So here is Canada facing a shortage of maple syrup.
Sarah Sharman 02:37
How will they eat waffles?
Chris Powell 02:39
That's the point. Like, that's how we've got a whole two hour movie right there just with that. I mean, you have no waffles. That's gonna trickle down to the US. No Waffle Houses.
Sarah Sharman 02:48
No Waffle House? Where will college students eat at three o'clock in the morning?
Chris Powell 02:52
Where will I eat at three o'clock in the morning? It's a conundrum of epic proportions, Sarah, that's why we have to solve this. And so, Canada says we can't stand for this. We've got to have more maple syrup, and so they see that there are vast swaths of maple trees in the US. So here's Canada about to invade the United States to take over our maple syrup trees, or the trees that maple syrup comes from.
Sarah Sharman 03:14
So we need a solution. I have a little idea. What if there's a spy on the Canadian lines sending the US messages about their strategy?
Chris Powell 03:26
Exactly. Okay. I like this. The spy in Canada is called Buddy.
Sarah Sharman 03:30
Buddy the Elf?!
Chris Powell 03:31
Shh. He's undercover. Oh, okay, so there's Buddy. How do we communicate with Buddy?
Sarah Sharman 03:35
Text message, obviously.
Chris Powell 03:36
Nope. Can't do that. Phones, out of the question. Beepers, pagers, you've watched by movies. Those are all yeah you know.
Sarah Sharman 03:41
Okay, I have a crazy idea. It brings in a little science. We encode messages in DNA and put it in a carrier pigeon.
Chris Powell 03:52
Perfect. Well, okay, so carrier pigeons is that that's not call that Canadian.
Sarah Sharman 03:59
Um, well, I mean, we could swap it out for, like, what would you say, a Canadian goose?
Chris Powell 04:03
Now we're talking, yep, Canadian goose. So we'd have a message encrypt it in DNA. But the DNA inside of the goose or, like, strapped around his neck.
Sarah Sharman 04:12
Oh yeah, straight in the goose.
Chris Powell 04:14
In the goose, okay. So that means that Buddy, our buddy in Canada, has to know how to sequence DNA.
Sarah Sharman 04:21
Yeah, but that's a minor plot line.
Chris Powell 04:24
Okay, so he gets these messages, and he's able to foil the plan to invade the United States and take over our maple trees once he gets that message from the DNA. Yep, Waffle House saved. Perfect. We need to run this by our faculty investigator, but I think this is flawless.
Sarah Sharman 04:40
Yeah, we're gonna make millions.
Brian Roberts 04:45
I'm Brian Roberts. I'm a researcher at HudsonAlpha Institute. My expertise is in, I guess I'd say functional genomics, which is understanding how the genome actually works and how it accomplishes what it does in terms of expressing genes, making different cell types do different things, and responding to different environmental stimuli. So most of my work is in manipulating cells in a dish, usually those are derived from induced pluripotent stem cells that we turn into other cell types. Most of the time, it's neurons, but we've also played around with other cell types as well. Then, at least half my time, probably more, is in computational work, which is analyzing the very complex data that comes out of those cell-based experiments and trying to figure out what we did and what we might have learned, and also developing new techniques so that we can learn new things that we couldn't before.
Chris Powell 05:51
Before we get into your work specifically and how that kind of coincides with our brilliant screenplay that we have put together, we need to ask you, though, do you have any favorite science fiction movies or TV shows? Anything that sticks out to you where the science is being portrayed?
Brian Roberts 06:10
So in terms of my favorite science fiction, that would probably not be very accurate science fiction. So I'm a big fan of the original Dune, not that I didn't mind the two that came out recently, but the weird David Lynch version that came out in 1984 is one of my favorite science fiction movies. In terms of the accuracy of the science, there was one about, it wasn’t Outbreak, it was one of those pandemic movies that came out, oh, gosh, I can't remember. I think it had Kate Winslet in it.
Chris Powell 06:46
So we've done a little bit of fact-checking, and we believe that this film with Kate Winslet is Titanic.
Sarah Sharman 06:51
Chris, there was not a pandemic on the Titanic.
Chris Powell 06:55
Well, not that they recorded anyway.
Sarah Sharman 06:57
The movie is actually called Contagion.
Chris Powell 06:59
okay, makes more sense.
Brian Roberts 07:00
The science was pretty good in that. So one of the things they were trying to find a cell that they could culture the virus in so they could get enough of it. And so when I saw that part, I was like, that's something that you would actually have to do in this kind of process. So I think they had a pretty good consultant for that one.
Sarah Sharman 07:18
So, does it ruin your movie-watching experience if the science is just wildly inaccurate, or can you find entertainment value in it anyway?
Brian Roberts 07:26
Kind of. If you're setting something far enough in the future, then I think you're in a much better space because then, as long as you're staying sort of consistent within the rules you've set up of like, we have this future science, and it can do this and this and this, and you kind of stay in that framework, then me, as a viewer, I'm like, okay, you know, they're keeping in touch with their rules. But if sometimes it'll be a movie that's only set 50 years in the future, and we've got, you know, spaceships flying at above light speed and all this, I don't think we're gonna be there in 50 years, that seems, and I never understand why you do that. Like, why not just set it 200 years in the future or 500 years in the future? Like, you can do whatever you want.
Chris Powell 08:11
A premise of the movie that we were working on putting together is that data could be stored in DNA and then put inside of a pigeon, which turned into a Canadian goose.
Sarah Sharman 08:24
Oh, that's right, yeah, we changed that to a goose.
Chris Powell 08:26
Let's just start with the first part of that premise, right? That data can be stored in DNA. Is that, is that a legit thing?
Brian Roberts 08:33
Sure, yeah. I mean, I don't know why the four bases of DNA wouldn’t be enough to encode any piece of information that you want. Now, is it an efficient way of storing it? I don't know. You know, I mean, all digital information is stored as a bit, which is a one or a zero, so essentially, that's a two-letter code. So at each position in DNA, you've got more information that's storable because you have four different letters, and in terms of, say, the size, the molecular size of what you're storing, DNA is going to be smaller, most likely than a bit on a microprocessor. I think the challenge is encoding the information into the DNA and then getting it back out. And actually, the getting it back out is now fairly trivial with modern sequencing techniques. So, you know, we can, we can pull DNA sequence out of samples very quickly, in hours now, where it used to take years. Encoding it into DNA is probably going to be the rate-limiting step in that because this is, you know, it's still chemical synthesis of DNA that is fairly time-consuming and slow. And so if you're talking about a very complicated message, which would probably need to be, I mean, I don't know, like, how complicated we're talking here, but if you're like encoding, like blueprints to something, it'd be 1000s of bases, maybe millions of bases, to do that could be even more. That would be very challenging to synthesize.
Chris Powell 10:21
I mean, we'd have to have quite a bit of data because we are talking about Canada's maple syrup reserve here. Yeah, so seems like a lot of data with that.
Sarah Sharman 10:30
I know that I've read that companies have discussed storing data in DNA, but I'm assuming that's just like native, free-floating DNA, and we're talking about putting it in an organism. Is there going to be an additional layer of a problem by putting DNA into an organism?
Brian Roberts 10:49
So yeah, I was thinking about that. So I don't think you would want to put it into the genome of the organism, meaning in every cell, part of its genes, right? So that's a very time-consuming process. I mean, it is done. I mean, this is how we create transgenic mice that are models of disease. In fact, there are mice that have humanized genes. So, instead of having the mouse copy of the gene, they have the human copy of the gene. So that was usually done by taking a mouse embryonic stem cell and then making that change in that stem cell and then re-differentiating that into an early-stage embryo and growing up the organism. And what you're actually hoping for in that case is that the edit appears in the gametes of the organism because then you can just breed that animal and make more and more of that organism. But as you can probably tell that process takes a long time, right? So you know, you're relying upon reproduction of the animal in order to make this line. So usually, making a mouse line from scratch, I'm not a mouse biologist so I don't know for sure, but I think you're talking about better part of a year. Maybe there's specialized labs that can turn around faster than that, so that would probably be inconvenient for sending messages.
Brian Roberts 12:12
I think, though, what you could do is use a retrovirus. So a retrovirus, like lentivirus, is something that we use in the lab every day. It's derived from the HIV virus. But all retroviruses will take their genome (their genomes are RNA) and convert it to DNA, and then they'll insert that into the host genome. And so that's a way of getting a chunk of DNA (because the message in the RNA is going to be preserved in the DNA) into your host cells genome. But you get back to the problem of what cells you want to deliver it to, right? So, just like it's hard to get CRISPR into cell types, and people are looking at using retroviruses to get CRISPR into cell types, you're only going to be able to get that message into a subset of the cells. So, if you're trying to put it into a Canadian goose, you could probably get it into its blood cellsif you injected it with the lentivirus. Now, there's a decent chance it would have an immune reaction, too; I mean, you just gave it a virus. Now, you might be able to engineer your virus to not be as immunogenic as a normal virus would be. So, there might be a way to make that work. And then, you know, if it's in the blood cells, and then it's a person on the other side, could just take a blood sample and then maybe sequence it to find the sequence in there.
Chris Powell 13:47
In our movie, which so far, you haven't disproven, I'm still riding with our movie, you've just made some obstacles. You haven't disproven it. We just have to be ready for the maple syrup shortage about two years in advance.
Brian Roberts 14:04
No, I mean, so the lentivirus you make in cell culture. So, I mean, you can make lentivirus in that day or two.
Chris Powell 14:11
Oh, okay, that's not hard, yeah. Side note, I've never heard of this virus before, so you're teaching me today.
Brian Roberts 14:17
It’s a commonly used lab technique when you just want to knock in a gene. So when we're doing it with CRISPR, we're knocking in the CRISPR genes, which are bacterial genes, so they're not normally in human cells, so we're just knocking those in, or we're knocking something that will express a trans gene, so a gene that maybe is already in there, but we want to make more of it. So, we'll use lentivirus to knock it in that way. Very commonly done. Pretty easy to actually make lentivirus.
Sarah Sharman 14:46
So now that we have our mechanism for getting the DNA into the goose, I think the real question now is, how is Buddy going to read it? Are we going to need AI for that?
Brian Roberts 14:57
Yeah, so I think you would want the recipient to have a sequence of DNA to look for. So, the tricky thing about it is when it integrates into the host cells genome, it integrates randomly, so you don't know where it's going to land. So, finding it without knowing a sequence to look for would be really tough. So that actually might help with the sort of encryption. Part of it is that even if the opposition knows to look in the blood cells of a goose, they wouldn't know what sequences to look for. Like you'd probably want what I would call a preamble, like a chunk of DNA that says, ‘This is the start of the message,’ and then anything after this is the message. If you had that as a recipient, it would be trivial to find the DNA code. Be very easy to do. You would just do, I mean, you'd probably just do a whole genome sequence, and most things would align to the goose genome, and then you would just find any reads that had that preamble in it, and you could actually then assemble the message based on that. That would be pretty easy to do. I don't think you would need AI to do it. You could just do it with existing tools.
Chris Powell 16:15
Complete side note, but do we know how complex the goose genome is?
Brian Roberts 16:20
I don't, I don't have much experience in avian genomes.
Chris Powell 16:25
I was just curious if it was, like, 10 times more complex.
Brian Roberts 16:29
It's a hard thing to speculate on because there are ones you're like, I don't know, it's probably like most other genomes. And then, if I say that, then it turns out the Canadian goose genome is incredibly complex, or something like that.
Sarah Sharman 16:39
We'll look it up and have it as a side note.
Sarah Sharman 16:44
We did some deep fact-checking. *whispers* (It's just Google), and we found that the goose genome is right over a billion bases; comparing that to the human genome, which is 3 billion, the goose genome is significantly smaller than the human genome.
Chris Powell 16:59
Are we talking about Canadian Geese here? Gray geese?
Sarah Sharman 17:02
Canadian, yes. Okay, good.
Brian Roberts 17:04
But I think also the advantage of this random integration is it would make it hard on people who are trying, like opposition, trying to decode it, because then they wouldn't know where to look, where to start, and you know they're gonna align to the goose genome, and most of the reads are going to align. So they're gonna have to pick the ones that don't align and sift through those. The ones that don't align would not align for a lot of different reasons. So it could be just bad reads. It could be actually other viral genomes that are naturally in. So, not necessarily genomes that that goose was sick with and got during its lifetime, but genomes it inherited. Because these viral genomes stay in your DNA, and you can pass them on to your progeny. Those they would have to filter through. I think it could be done, though, eventually. So I think what would give away your sequence is it would be this fairly large contiguous set of exogenous sequence. And so there might be an approach where you could just start trying to assemble like we used to do to make genomes. We just take small chunks of DNA and figure out which one fits with which, and then you can stretch them out and get the entire sequence. They could probably do that with those reads that don't align and be like, well, most of these assemble into these fairly small chunks of DNA, but one of them assembled into this, you know, 10 kilobase region that looks suspicious to me. That might be the secret message.
Sarah Sharman 18:49
Interesting. Well, so as we're Oh, go ahead.
Chris Powell 18:52
I was just gonna say now we know how to go on defense, also, just in case this scenario ever pops up.
Brian Roberts 18:58
Yeah. I mean, if you want to fragment your message and integrate it in separate chunks that could be another that could make it harder to decode. Okay, yeah, wow.
Sarah Sharman 19:10
So as we're thinking through the set, do we have to have a lab that these people are working in, or would it be too outlandish that they might have like a handheld DNA sequencer?
Brian Roberts 19:23
Um, I don't think it's outlandish. There are, you know, briefcase-size sequencers right now, and there's a company named Oxford Nanopore, which makes a sequencer that can plug into your USB drive on your computer. I don't know how well it can do a whole goose genome, but you could make it at work if you really wanted to. So it could be done out in the field. What you need a lab for is you have to take the blood, and you'd have to process it. DNA and isolate it and purify it, and things like that that that you probably still need a lab for. Okay, although I don't know there are automated DNA prep machines; I don't think they would fit in your hand, though; probably fit on a table.
Sarah Sharman 20:13
Okay, so that would be a scientific red flag for you if we had a handheld DNA sequencer in this video.
Brian Roberts 20:21
Just set it 100 years in the future, and then I could probably buy it.
Chris Powell 20:26
Would that handheld device be loaded with the informatics software for interpreting or would that have to be separate?
Brian Roberts 20:35
Sure it probably could.
Sarah Sharman 20:37
Yeah, I'm wanting, like, a one stop shop, like we extract DNA, we sequence it, we get our message out, basically, like a Game Boy.
Chris Powell 20:45
Yeah, a Game Boy or a, what do you call it? The blood sugar meter. Oh, yeah, yeah, right. So you just prick, done.
Brian Roberts 20:56
I think you would need more blood than you did off of a finger stick.
Chris Powell 21:02
just chop the finger off.
Brian Roberts 21:03
I think you just, I don't know if the goose is gonna make it. You need, you need a lot of his blood.
Chris Powell 21:10
That's fine.
Sarah Sharman 21:11
We can't kill geese in this movie.
Chris Powell 21:13
Sure, we can.
Chris Powell 21:16
Well, and one side note is that we're gonna make the plot thicken, because it turns out that this was actually an American goose impersonating a Canadian goose to get across the border.
Sarah Sharman 21:33
This has been such a fun discussion, but I think, as our science consultant, we just wanted to know if you're green-lighting our film or not.
Brian Roberts 21:41
Oh, do I have that kind of power? Yeah, sure.
Chris Powell 21:45
It depends on if you green light it.
Brian Roberts 21:48
Sounds good to me.
Chris Powell 21:50
Then you have that power.
Sarah Sharman 21:50
Yeah? Awesome. We'll check back with you about the whole lentivirus once we start getting going.
Chris Powell 21:57
Okay, so you green-lit it. Does that mean you also are funding?
Brian Roberts 22:02
I can give you like 10 bucks or something.
Chris Powell 22:04
Perfect, low budget.
Sarah Sharman 22:09
All right, thanks so much.
Chris Powell 22:19
I think that was a glowing endorsement from our faculty investigator. Our screenplay is greenlit. It's going to be coming to a blockbuster store near you, in the VHS aisle.
Sarah Sharman 22:28
Assuming that someone wants to buy our script. So, if you're interested, we're here. We could use a couple milli.
Chris Powell 22:35
Yeah, we'll let you know where to send it. So thank you for joining us on this season of Tiny Expeditions, we hope you've enjoyed getting pop cultured with us. Stay tuned early next year for an announcement about our next season.
Sarah Sharman 22:50
Tiny Expeditions is a podcast about genetics, DNA and inheritance from the HudsonAlpha Institute for Biotechnology. We're a nonprofit research institution in Huntsville, Alabama, with a unique mission.
Chris Powell 23:02
We bring together scientists and companies to develop and apply genomic advances to make a better world, that's everything from cancer research to agriculture for changing climate.
Sarah Sharman 23:10
If you enjoyed this episode, swing by your favorite podcast app and hit that subscribe button while you're there, consider leaving us a review. It really helps us spread the knowledge. Thanks for joining us.
