What a trove of potato genomes reveals about the humble spud
32:41 Share

Instacart is on a mission to have you not leave the couch this basketball season. Because between the pregame rituals and the postgame interviews, it can be difficult to find time for everything else. So let Instacart take care of your game day snacks or weekly restocks and get delivery in as fast as 30 minutes.

Because we hear it's bad luck to be hungry on game day. So download the Instacart app today and enjoy $0 delivery fees on your first three orders. Service fees apply for three orders in 14 days, excludes restaurants. And now, a next-level moment from AT&T business. Say you've sent out a gigantic shipment of pillows, and they need to be there in time for International Sleep Day. You've got AT&T 5G, so you're fully confident, but...

The vendor isn't responding. An international sleep day is tomorrow. Luckily, AT&T 5G lets you deal with any issues with ease, so the pillows will get delivered and everyone can sleep soundly, especially you. AT&T 5G requires a compatible plan and device. Coverage not available everywhere. Learn more at att.com slash 5G network.

Welcome back to the Nature Podcast. This week, a potato pangenome and the most cited research papers of the 21st century. I'm Benjamin Thompson and I'm Nick Pertscher-Chowe.

First up on the show, researchers have been creating a collected group of genomes known as a pan genome to understand the genetic diversity of the humble potato. This effort could help in the breeding of new potato varieties such as those adapted to disease or climate change. Now, potatoes are incredibly flexible. You can boil them, you can mash them and you can even feed over 1 billion people with them.

But if you want to make a new variety of them, that isn't so easy. So compared to other cultivars, potatoes are particularly difficult because of the tetraploid nature. This is Sergio Tussauds, a geneticist who's very familiar with the difficulties potatoes and the breeding of their different varieties or cultivars pose.

You see, the potato is tetraploid, meaning that instead of having two sets of chromosomes, like you and I do, potatoes have four. So this makes it particularly difficult when it's related to breeding programs, right? So you want to introduce new traits or characteristics, then it's very difficult to put it in the four copies of the genome. So that means that

The breeding programs in potatoes have not been very effective and we have mostly been using the same varieties in the last, let's say, 200 years. When you cross two different parent varieties of potatoes together, it's hard to control which genes you're going to get in the offspring, making it difficult to know what traits you might end up with.

And that matters, because if it's hard to breed new types, it could make it difficult to adapt potatoes to new challenges, like new diseases or climate change. And that is important because… I guess the first thing is that it's tasty, right? I think everyone likes potato.

But probably also very important that is the food for 1 billion people in the world. So it's one of the most important crops. So to help this tasty and important crop become easier to breed, Sergio and a team of researchers are writing in Nature this week about a pan genome of different European potato varieties.

essentially a combination of multiple genome sequences that allows researchers to get a sense of where they're similar and where they're different. And from this we can sort of predict what is the diversity

that we find in most of the cultivars that we consume today. As it turned out, potato varieties are very genetically similar. So the team were able to capture an estimated 85% of the genetic diversity that exists in these European potatoes by looking at just 10 genomes. Those genomes came from old potato varieties that are essentially the parents of many present-day potatoes.

That's in part because of the potato's particular history. Because while they appear on plates the world over now, for thousands of years previously, potatoes were only grown in the Americas, before being brought to Europe by the Spanish in the 16th century. This handful of varieties were bred and bred and bred to create many of the huge variety of tasty tubers enjoyed around the world today.

And this gave the team an advantage when it comes to capturing most of their genetic diversity. There's something like 1,700 different varieties of potato grown in Europe today, and that's too many to assemble. This is Craig Dent, another member of the Potato Pan Genome team.

To overcome this, they went back to some of those original varieties brought over to Europe. We thought if we looked further back in time, we might be able to find the different sets of DNA that have then spread out amongst this population, but get them where they're a bit more concentrated, up near the top of the family tree. And whilst that may sound straightforward, it was quite a challenge.

as that complicated genetic tetraploid nature has not only thwarted breeders, it's made things hard for geneticists too. Here's Sergio. You can picture assembling a genome as a puzzle, right? So you have many different pieces and then you put them together. But in this case, instead of having one, you will have four layers of the puzzle.

in which the figure of the puzzle is quite similar between them and it's very easy to mix between them. The complexity of having to solve four very similar looking jigsaw puzzles has meant that it's only been in the last few years that a potato genome has been assembled at all.

In this instance, the team needed to combine the genomes of multiple varieties in order to make a pangenome. But with improvements in DNA sequencing and some clever techniques, the team managed it. And so, what does the potato pangenome tell us? The first thing that we found is that when you look at the sequences of DNA, they're actually super, super different, right? So you have extremely high diversity at the sequence level.

but actually these diversities contrast by the number of unique sequences, which is very, very low. Now, that may sound a bit counterintuitive, but basically it means that there are very few unique bits of DNA between the different varieties. However, when they are unique, they are very different from one another.

Sergio thinks that this indicates that potato varieties lost a lot of their uniqueness when they were grown in the Americas, possibly from the indigenous peoples in the Americas breeding them for desirable traits, or it could just be a quirk of how this species intermixes.

Previously, it was thought that because only a handful of potato varieties were brought to Europe, this acted as a genetic bottleneck, removing uniqueness and making subsequent varieties more similar. And whilst the team do see signs of this as well in their research, it looks like potato varieties lost a lot of genetic diversity before the trip to Europe, although more work will have to be done to understand the history of the potato fully.

This lack of uniqueness between potato genomes could mean it's even harder to breed new ones with desirable traits, as there's not many new genes that could be introduced, at least through conventional crossing of cultivated types. But because potato varieties are so similar, the team hopes their pan genome can speed up the sequencing of varieties not included in their work.

something they tried out with a very popular potato. And the cultivar that we took for that is this cultivar rice at Burbank that's used by McDonald's for their French fries. You might be familiar. And so we took that cultivar and just using some relatively cheap sequencing techniques

tried to reconstruct this genome and we found we could put together a large part of it. Assembling genomes of potatoes very quickly and cheaply is still a way off, but that's the direction the team are moving in. Once that is possible, it could also be possible to identify specific traits that you'd want to breed or even genetically engineer into new or existing varieties.

This could be key to making sure we have potatoes resistant to new diseases and the future warmer world they'll find themselves in. One outstanding question is the genetic diversity of potatoes that the team didn't capture. They estimate they got around 85% of the genetic diversity of European potatoes, meaning there's about 15% this work doesn't cover.

This may include a lot of DNA that's been brought into modern potatoes in the last 50 years or so. During that time, people went back to the wild potatoes in the Americas to try and find useful traits like disease resistance and bred them into the potatoes we eat today. The team are looking to fill that gap, and it's unclear what might be found in those remaining potato genetics.

Greg for one is excited to find out, and about a whole new world of potato genetics in the future. We're kind of riding a wave of potato genomics right now. Two years ago the first potato genomes for these tetraploid cultivars were released and I think in this year we're expecting 100 to be released.

So it's really ramping up. And what we're going to find from all that information about what makes a potato, I honestly can't say yet. That was Craig Dent from the Max Planck Institute for Plant Breeding Research in Germany. You also heard from Sergio Tuso from Ludwig Maximilian University, also in Germany.

If that story has made you hungry for more, check out the show notes for some links. Coming up, a team here at Nature has been investigating the most highly cited papers of the past 25 years. If you want to hear what number one is, stick around. Right now though, it's time for the Research Highlights with Dan Fox. Medieval manuscripts are an important source of historical knowledge. And not just for what's written on the pages.

Researchers have used a range of techniques, including analysis of ancient DNA, to examine the intriguingly hairy outer bindings of some 12th and 13th century books from France, Belgium and the UK. The analysis revealed that the bindings were seal skin, with four covers identified as coming from harbour seals, another traced to a harp seal and one to a bearded seal.

DNA analysis of five of these skins suggested they probably originated in the seas around Scandinavia and Denmark. Their presence as far inland as Champagne in France attests to an important medieval trade in seal skins. Pelts for book coverings could have been purchased from Norse traders or merchants from the Hanseatic League, with both groups active in the region at the time.

You won't need DNA analysis to find the origin of the name. It's published in Royal Society Open Science. Physicists at the Atlas detector of the Large Hadron Collider have spotted top quarks – the heaviest of all known elementary particles.

Shortly before its winter recess each November, the LHC switches from accelerating and colliding beams of protons to smashing together heavy atomic nuclei such as those from lead. When the nuclei collide, they produce a quark-gluon plasma - a soup of particles thought to be similar to the dense hot state of the matter in the universe just instants after the Big Bang.

This state of matter is unusual in that quarks, which normally exist only inside a composite particle such as a proton or neutron, can roam free within it. ATLAS has now confirmed that these lead nucleus collisions produce top quarks, together with their antiparticle counterparts. The top quarks' fleeting interactions with the quark-luon plasma could help researchers to understand how the plasma evolves in time.

You can detect that research in physical review letters. This week, Nature is publishing a feature article looking at the most cited papers of the 21st century and why these articles are breaking records.

One of the team behind the analysis is Nature's Richard Van Norden, who joins me now. Richard, thank you as always for being here. Pleasure as always, Ben. So everyone loves when things are ranked, right? Like discussions can be had, that sort of thing. Why is Nature doing this? Yeah, well, we are a quarter of a way through the century. And I thought it would be fun to see what the highest cited modern papers were.

And not only that, but also to have a look at where they featured in the top cited papers of all time. We did a very popular feature more than a decade ago now where we looked at the top cited papers ever.

And actually, that list has changed quite a lot now. There's quite a lot of modern 21st century papers that are now reaching the top of those charts. And this isn't like a subjective list of which is best. Right. So citations are the way that authors acknowledge their previous sources in the literature. So a highly cited work is one that's very influential, but it's not necessarily the best or the highest quality. In fact,

readers might be surprised to know that big discoveries of the 21st century, like gravitational waves or the Higgs boson or CRISPR gene editing, none of those things

None of those papers are the highest cited of all. Of course, they are highly cited. And that's because researchers tend to cite methods and software the most. And that's because these are kind of workhorses of science that researchers depend upon. And it's important to cite what you use. And how have you gone about doing this Herculean task, I have to say? When we did this a decade ago, it was quite hard. And we asked the owners of a particular database to send us their list.

and that was the Web of Science, which is now owned by the US company Clarivate. But nowadays, there's many more databases that rank

papers by their citations and all these databases have different journals within them and they have different citation numbers so it's a total mess so what we did was we looked at five databases and we took the median rankings across them so we said well look the numbers of citations are actually different like google scholar the search engine will give hundreds of thousands of citations to a paper whereas another database which is not picking up as many works that might only give you know fewer than a hundred thousand for example

So we said, OK, yeah, but is the paper top ranked in Google Scholar and top ranked in this other database? Well, then it's number one. And that's given us a fairly robust overall ranking for these papers. Let's cut to the chase then. So you've looked at the most cited papers of the 21st century and you've said they're most of the big...

discoveries that we will be familiar with and covered on the podcast aren't in there. So Richard Van Odden, what is top of the shop? Well, the number one most cited paper is a paper from Microsoft researchers. And it's got, depending on the database you look at,

between 100,000 and up to 250,000 citations. And it's about a particular kind of artificial neural network called a ResNet, a deep residual learning network. Now, if people remember, artificial neural networks are the algorithms that underpin deep learning and all the AI advances that are built on deep learning.

And this sort of underpinned the deep learning revolution. Now, was this paper, you know, the best in AI? There are other AI papers on this list, like a paper called Attention is All You Need. That's number seven on our list. And that was the genesis of the Transformer, the architecture that's behind ChatGPT, where GPT stands for Generative Pre-trained Transformer.

So I think perhaps more people have heard of that paper in the field. But the Microsoft paper has accrued more citations than its number one. But, you know, they're both incredibly highly cited. And there are many other...

AI papers on this list. As you say, AI does dominate the list for a technology that kind of hasn't really existed for that long. Only the last couple of years, I suppose, has it exploded. Is there something about this field which has given it maybe an advantage, I suppose, in the list? Well, I think AI is just massively relevant to a huge number of different fields. They're all citing it. And in the 21st century, AI is the field where we've seen huge progress and loads of papers. And

it's almost kind of unfair if you're in an area where there are lots of papers you're going to get more citations so one thing we did check is that we asked the bibliometrician to do a bit of a correction thing we say okay well what about if you correct for like when a paper was published obviously the paper was published last year it's not making it on this list it hasn't had much time

But also, what about correcting for papers field? And even whether it was a journal paper or conference paper. So we did all of that. And to be honest, the same papers came up top, although some papers about COVID did make it in there. So these are sort of papers published in 2020, 2021, which have been cited very quickly about the pandemic.

but don't quite make this list otherwise because they've only been around for four or five years. And when researchers are writing papers, certainly in the medical or biomedical field, the first paragraph will often cite a statistic, like how many cases of a disease have happened globally each year, that sort of thing. And these are often citation number one on a paper. And

Unsurprisingly, in some fields, this happens a lot more than others. And those papers are making it into your list. Absolutely. So two of the papers in the top 10 are literally a report on global cancer statistics. And these reports are brought out every one or two years and everybody cites them. Also, what's sometimes called psychiatry's Bible, the Diagnostic and Statistical Manual of Mental Disorders, or DSM,

That is just a big book, essentially, that describes what are the criteria you should use for categorising addiction, depression, other mental disorders. And again, that gets referenced all the time by researchers. And so that's in there as well. Did anybody you speak to not realise that their work has been cited quite so much? I'm sure they must have Google Alerts or something that pings when a paper references them. Basically, everyone we contacted absolutely knew that they'd been cited a lot. There was one researcher who'd written a review paper about

cancer. And he said, students came up to me to say, I'm studying cancer research because of this review. And he said, I feel like a rock star. So people feel really proud. Of course they should, right? When they have a heavily cited work. And some people feel really surprised. There were two researchers who created an article about what's called thematic analysis, which is when you've done a lot of interviews and you're trying to pull out themes from those interviews. And

And these researchers, Virginia Brown and Victoria Clark, who are psychologists, they wrote a paper about thematic analysis. How do you do it well? And everyone started citing it.

And Clark said it was completely life-changing and that they've received invitations to meetings across the world as a result of this paper. I found it quite remarkable that this paper, which I personally never heard of, more fool me, was number three in this list. And do you think there's a through line between these papers? They're from very disparate fields, of course. Is there something intangible maybe that links them all that you've come across when you've been stacking them all up? I think it's all about that they're very useful methods. I think that's the key through line. One thing that

turned up which also surprised me was that if I looked at the database OpenAlex which is a free online database the number one was actually a mistake by that database but the number one was the coding software R literally the letter R

And Open Alex thinks that more than 300,000 articles have cited the use of RR, which doesn't surprise me. I'm sure millions of scientists have used RR. It was actually a mistake because what's being cited is not a journal paper, as Open Alex mistakenly recorded it. It's actually a website, literally the website for the RR project.

That's kind of intriguing and kind of shows you how capricious citation culture is, because if only the greatest of that software had written a paper about their software, they would have been absolutely top of this list. But at least more than 300,000 mentions of their website...

shows you just how important and influential the development of new methods, which nowadays in the 21st century generally means software, is to scientists. Speaking of the 21st century then, how do these papers from the last 25 years stack up then in the all-time pantheon? You said that Nature did a top 100, I think it was what, 2014, and you've updated it. What's going on there? Well, the Microsoft paper is also in the top 10 of all time.

But there are still three big hitters, and they're all biological methods. And the top paper is still a 1951 paper about determining the amount of protein in a solution, a kind of assay. And that's been cited more than 350,000 times, according to the Web of Science. And I don't think people still need to cite that paper. But for whatever reason, it's become a kind of symbolic thing that

you cite it because it's just become a norm to acknowledge in some fields. It shows other researchers the intellectual heritage that you're working with and so forth. Now, according to my calculations, if the Microsoft paper keeps on getting cited that it currently is, it could become number one before 2030. Well, I hope you'll join me again in five years, Richard, and we can find out. But for the time being, Richard Van Norden, thank you, as always, for joining me today. Thanks very much, Ben. Nature's Richard Van Norden there.

To read more about these highly cited papers, look out for links in the show notes. Finally on the show, it's time for the briefing chat, where we discuss a couple of articles that have been highlighted in the Nature Briefing. Ben, what have you been reading this week? Yeah, quite an intriguing one, this one, Nick. This is a story I read about in Science, and it's based on a paper in the Astrophysical Journal. And it's an update on a story I covered on the podcast back in 2023, and it centres on

on a star eating a planet. You might remember it. It was a fun story. But there's been some new evidence from the team behind that work, some new detective work, if you will, which maybe puts a new spin on how things played out. Right. I do remember this story, but I don't remember exactly the details. So maybe you can give me a recap of what the original finding was and how this differs from that. Of course. So the story was that a star in a dense region of the Milky Way, maybe 10,

12,000 light years from Earth or so, suddenly got a lot brighter in the sky. And the researchers pointed some different telescopes at it. And to cut a long story short, the conclusion they came to to explain this brightening was that this star had eaten a planet, a planet about the size of Jupiter, they said. Now, this was the first time this sort of act happened.

had been caught. And in particular, they thought that what was going on was that the star was coming towards the end of its life. So it was expanding, swelling up,

And in doing so, gobbled down this Jupiter-sized planet as it got larger. That makes sense. So star gets bigger, planet gets eaten, star gets a bit brighter, gets picked up by telescopes. But that may not be what's happened. Right. And this event was first detected in 2020, I think. And since then, the James Webb Space Telescope, the JWST, came online. And so, of course, the team were itching to point this new instrument at this star. I guess it's been described as a post-mortem examination. And interestingly...

In doing so, as I say, they've potentially changed the narrative here. So the JWST measurements suggest that this star actually wasn't bright enough to be in this kind of bloated end-of-life state where it expanded to eat the star. It was more of a regular, in inverted commas, star. So what they think happened, Nick, is that...

Rather than the star eating the planet, it may be that the planet essentially threw itself into the star. I mean, that is quite a different way of this happening. And that raises a question in of itself. Like, how might that have happened? Like, planets normally have pretty stable orbits around stars. What might have gone on here? Right. So this planet was very close together.

I think estimates suggest it was closer than Mercury is to our sun. So big gravitational pull on it. And what may have happened is that this constant pull deformed the planet, a bit like how tides are made here on Earth. And this constant deformation ultimately made the planet lose some of its orbital energy, making it slowly, slowly edge-shape.

towards the star before ultimately getting so close that it did smash into it, heating up and firing out stellar gas, which may explain this brightening that was seen. So more like what they saw originally was the end of a very slow car crash, I guess, if I can put it like that. And does this research tell us anything more about planets and stars and how they sort of interact with each other? Well, that's a really good question. But I think what has to be said is

This is kind of a narrative. This is a good demonstration of science at work. With new technology, with new instruments, comes new influences. And so, really, we've got these two ideas at play from the same group of researchers, and more work will need to be done to unpick what isn't.

happen. In the article they say that perhaps some of the stuff kicked out by the star masked how bright it really was so actually potentially it looks younger than it was if you see what I mean so maybe the original narrative is still applicable. So I'm sure more measurements will be taken but if it is this kind of second idea that the planet spun in

into the star it adds more insight into what's going on in the universe I think generally people assumed that this expansion followed by Goblin Up was what happened to planets as I say hard to catch in the act this was the first reported example of it happening

But maybe there's another way that planets do end up getting swallowed by stars, which is this inexorable slow circling of a stellar drain, if you will, before they cease to be. Wow. I mean, it is endlessly fascinating how much we still have to learn about the universe and how everything works.

Because as you say, I would think like, yeah, star expands, swallows up planets, makes good sense. But there are different ways that this could happen. So thanks, Ben. And I guess keeping with the destruction theme, I've got a story about trees being struck by lightning. But it's actually kind of good for them. OK, right. Lightning, as I've learned also from the podcast, is one of the most complicated things.

physics phenomenon. So I'm not going to ask you how lightning works, but as you say, when lightning does strike a tree, usually that signals the end of the road for that tree. Yeah. I mean, the thing about lightning is it's very powerful. Without going too much into the physics, as you say, a lightning strike has like 300,000 amps of electricity. So that's quite a thing, and that can generate a lot of heat, and that can destroy trees.

But in this article, which I was reading in Life Science, based on a paper in the New Phytologist, some researchers have identified a tree called the Tonka bean tree that actually benefits from this process and seems to come out from lightning basically unscathed. Right. Okay. So what's going on here? Presumably the lightning's not giving the trees superhero-like powers. No.

No, this isn't a 1950s science fiction story, unfortunately. So what is happening here, the researchers believe, is that when lightning strikes the tree, basically it's not just the tree that's being hit. So this tonka bean tree may be struck by lightning, but many of its neighbours will also be caught in that impact.

and they will be killed by that strike, which could actually be of a benefit to this tree because then it's got less competition. And one of the things this tree really needs is light, and it struggles a lot with competition for that.

And the other thing that lightning may help with is that there are parasitic vines that basically climb up this tree and can cause it some stress. And they also get cleared by the lightning strike, while the tree itself, as I say, comes out pretty much unscathed. Right, so the competition's getting electrocuted then, leaving the forest...

playing field a little bit clearer for the Tonka bean tree. But surely the tree itself is toast as a result of these strikes? No, as I say, it's quite unscathed from it. Now, I will say the researchers aren't exactly sure why, but one of the leading ideas is that the tree may have great conductance. So what that means is essentially the lightning is just

Wow, I mean that's something, right? Lightning, as the old saying goes, rarely strikes twice.

but is this phenomenon seen in other places? Well, one fun thing is actually these trees live for centuries. So they estimated that some of these trees have probably been struck like 10 times or so. So lightning has very much struck more than twice in this particular case.

But yeah, one researcher who wasn't part of this work, who was interviewed for this article, said that this could apply to other trees as well. Researchers have seen in the past that trees can survive multiple lightning strikes and survive for centuries afterwards. So what's happening with this tree could also be applicable to other trees as well and

As I say, it will take a bit more research to figure out exactly how it is that these trees are surviving. But for very big and tall trees like this tree, this could very well be part of their evolutionary adaptation because the thought is here that this tree is adapted to lightning strikes because it benefits from them. It doesn't take any damage, but its neighbours...

They certainly do. That is super neat. But I have to ask, the story that I covered about lightning a long time ago was based on a communications tower in the Alps that researchers knew would be hit multiple times. Presumably, if you're in a forest, you can't just watch a tree for an amount of time and hoping that it will be hit by lightning.

by lightning. So how do they go about it? Yeah, I mean, it's not an easy thing actually to figure out where exactly lightning will strike. And they actually had a very complicated setup to look at this. So they had an entire array of antennas. And again, not going too much into the physics of how lightning works, but when lightning strikes, there are radio waves that can be detected by these antennas. And because they had an array, they could pinpoint exactly where

where it was and then they could look at these patterns detected by their antenna array and then figure out exactly what tree was hit and look at the trees over time so they then monitored the trees over time using drones and things like that and then they showed that yeah these trees are basically fine whereas all these competitors um are not doing so well there's actually a picture in the article that listeners can go look at of a tree before and after a strike and the

The tree before the strike is covered in vines and it's quite crowded. And then the tree after the strike is on its own, looking strong, looking very cleaned up, as it were. Right. Well, we will put a link to that story in the show notes so you can go and see the lightning-stripped Tonka Bean tree. We'll also put a link to the other story as well. But if you want more stories like this, listeners, delivered directly to your inbox, then look for another link which will let you sign up for the Nature Briefing, which will do just that.

That's all for this week. But just before we go, we just want to mention once again that we have been nominated for several Webby Awards and we need your help. Absolutely right. So there's a People's Voice Award that can be voted for by people.

people like you. And it'll take you about five minutes to vote, ideally for The Nature Podcast, and we'll put links on where to do that in the show notes. In the meantime, if you want to keep in touch, you can follow us on X or on Blue Sky, or you can even send an email to podcast at nature.com. I'm Nick Pettrichow. And I'm Benjamin Thompson. Thanks for listening.

And now, a next-level moment from AT&T business. Say you've sent out a gigantic shipment of pillows, and they need to be there in time for International Sleep Day. You've got AT&T 5G, so you're fully confident. But the vendor isn't responding, and International Sleep Day is tomorrow. Luckily, AT&T 5G lets you deal with any issues with ease, so the pillows will get delivered and everyone can sleep soundly, especially you. AT&T 5G requires a compatible plan and device. Coverage not available everywhere. Learn more at att.com slash 5G network.

Download the Redfin app to get started.

Equal Housing Opportunity, CADRE, number 01521930.