Interviewer: Adam Levy
This week: storing data in the tiniest of spaces.
Interviewee: Fabian Natterer
Current technology is of the order of a million atoms and we have now reached the very end which is a single atom.
Interviewer: Kerri Smith
And, a study of microorganisms in Neanderthal mouths paints a graphic picture of their encounters with humans.
Interviewee: Laura Weyrich
And so it actually suggests that there might have been some sort of spit-swapping between humans and Neanderthals.
Interviewer: Adam Levy
Plus, we chat to the scientists who uncovered the world’s oldest fossils. This is the Nature Podcastfor March the 9th2017. I’m Adam Levy.
Interviewer: Kerri Smith
And I’m Kerri Smith.
[Jingle]
Interviewer: Kerri Smith
Adam, open wide please. I need to look at your dental calculus.
Interviewer: Adam Levy
My what?
Interviewer: Kerri Smith
Your dental calculus: calcified dental plaque. Basically, bits of food and microbes that live among your pearly whites.
Interviewer: Adam Levy
No, I’d really rather you didn’t. My teeth are spotless and you’re definitely not a dentist.
Interviewer: Kerri Smith
Correct! But I am very interested in what dental plaque can reveal about diet and health. A new study in Naturetakes a look at some dental calculus from Neanderthal teeth: one individual from Belgium and one from Spain. They used DNA sequencing on the material furring up and solidifying on the Neanderthal teeth. It’s given them a lot of information about food, bacteria, even medicinal compounds – really anything these guys put in their mouths.
Interviewer: Adam Levy
Sounds fascinating. Why not get the authors to open up about that?
Interviewer: Kerri Smith
Well, handily for you, that’s what Ewen Callaway has done. He and author Laura Weyrich, who’s at the University of Adelaide, had a wide open discussion about mouth health and the big differences between Neanderthal diets in different places. Here’s Ewen talking to Laura.
Interviewer: Ewen Callaway
Your team found really striking differences between Neanderthals living in two different parts of Europe, with regards to their diet, right?
Interviewee: Laura Weyrich
Yeah, absolutely. So we looked at Neanderthals in both Belgium as well as Spain, and we also looked at another Neanderthal in Italy but it didn’t happen to result in good DNA sequences. But the Neanderthals in Belgium are what we would consider the more classic Neanderthals. These guys were eating a lot of meat so we found evidence of DNA in their teeth for woolly rhinoceros as well as mouflon sheep which are a wild type of sheep that live in Europe at that time. But that’s very, very different from what we see in Spain. The Spanish Neanderthals are almost vegetarian, so we don’t see any evidence of meat consumption present in their calculus. We only see things like pine nuts, moss, even tree bark. They were eating basically what they could find in their very forest-like ecology that they were living in. These would really have been the true Palaeo Diet Neanderthals. We joke about that a lot, that we should write the true Palaeo Diet cookbook and include moss and tree bark and pine nuts.
Interviewer: Ewen Callaway
So we found, in Neanderthal plaque, evidence that they were foragers in one area and meat-eaters in another area: true ‘locavores’. What else did we learn from their plaque?
Interviewee: Laura Weyrich
So, when we were looking for dietary information we incidentally found some evidence that provided insight into the medicine or the medicinal practices that these Neanderthals may have had. There was one Neanderthal in particular from El Sidrón Cave in Spain that was suffering, probably, some really nasty side effects from a dental abscess, as well as a gastro-intestinal pathogen microsporidia. The dental abscess is evident in the skeletal material. You can see where that abscess was eroding its jaw but the microsporidia was something we were able to pick up in the dental calculus. So it’s likely that he was suffering from a diarrhoea or some kind of gastro-intestinal type bug that would have made him noticeably sick.
So we were actually able to go in and look for, not only pathogens that would have made him sick, but then also how he might be treating those pathogens. And so we found evidence for him consuming poplar bark and poplar bark contains salicyclic acid which is the active ingredient in aspirin. It would have treated his pain from this disease. We also found evidence for him consuming moulded herbaceous material, and this moulded herbaceous material contained Penicillium which is a natural source of antibiotic penicillin. So we probably don’t think Neanderthals thought to seek out this particular mould and consume it as a medicinal act but more so that they may have known that eating mouldy material might make you feel better.
Interviewer: Ewen Callaway
Are you able also to get a snapshot of not just a couple of pathogens but of the overall microbiome of the teeth?
Interviewee: Laura Weyrich
We certainly are and for me that’s the most exciting part about this paper. This is actually the first microbiome of an extinct species as well as of an extinct hominid and it really was a lot of work to try to reconstruct these healthy, beneficial microorganisms that are present in the plaque. In this particular study we’re able to show that chimpanzees and vegetarian Neanderthals and gatherer humans used to share a microbiome, and it’s really the introduction of meat into the diet in both Neanderthals and ancient humans that causes the first change away from that shared hominid microbiome a long time ago. So it really affords us a bit of a window into our own health as well as Neanderthals.
Interviewer: Ewen Callaway
So what more can we learn from looking at ancient plaque?
Interviewee: Laura Weyrich
So I think that there’s a lot of stories that are present in each individual bacterial species that we find in the plaque and in this paper we really talk about one bacterial species in particular and it’s not really a bacteria, it’s an archaea. We can actually use the genomic mutations present in that species we find in Neanderthals and compare that to the same species that we find in modern humans and that tells us when one of the strains could have been introduced into one of those species. So here we actually identify that this archaea, Methanobrevibacter oralis, was actually introduced into Neanderthals about 120,000 years ago. And this is around the same time that interbreeding began occurring between Neanderthals and humans, and so it actually suggests that there might have been some sort of spit-swapping between humans and Neanderthals while these interbreeding interactions were going on. And that’s quite a unique thing to think about because many of these interbreeding interactions have been described as very rough, very quick encounters, not something that would have been very sensual or very intimate, but certainly if you’re swapping spit between species, there’s kissing going on or at least there’s food sharing, direct food sharing, which would suggest that these interactions were much more friendly and much more intimate than anybody ever possibly imagined.
Interviewer: Ewen Callaway
I didn’t catch that from your paper that you could imagine how Neanderthals might have kissed; that’s quite an insight!
Interviewee: Laura Weyrich
Yeah, it is. [Laughs]
Interviewer: Ewen Callaway
Wow, yeah, that’s really fascinating. Do any other aspects of your results help you imagine Neanderthals as individuals, as real people?
Interviewee: Laura Weyrich
It really paints a different picture almost of the personalities, of really who they are. And this technique allows us to go in a step further and really look in their medicine cabinets, look in their food cabinets, figure out where they were going and how they were viewing the world. It’s a very different take on ancient DNA and how it’s being used today.
Interviewer: Kerri Smith
That was Laura Weyrich talking to Ewen Callaway. Next, Laura wants to sequence food encrusted teeth from any ancient human relative she can find so her team can compare the mouthy microbiomes and prandial preferences of our family tree. Find this paper of our two Neanderthals at nature.com/nature. Ewen has also written a News story which you can read for free at nature.com/news.
Interviewer: Adam Levy
We’ve got tiny data storage facilities and tiny ancient fossils coming up in the show. Plus the Highlights cover woolly mammoths and clumsy drones.
Interviewer: Kerri Smith
First, an update on a story from last week. You might remember that food scientist Kassem Alsayed Mahmoud made it from his home country, Syria, to Belgium and is working at the Free University of Brussels. But the way the story ended was with his family in Turkey waiting to hear if they can join him. Well, last week Kassem emailed with an update: his family have been granted a visa to join him in Belgium and they should be arriving before the end of this month. Here’s wishing them a smooth trip.
Interviewer: Adam Levy
If you missed Kassem on last week’s show about migration you can find it and the wholeNature Podcast archive at nature.com/nature/podcast.
[Jingle]
Interviewer: Adam Levy
To a computer, your precious holiday photos, spreadsheets, or all those archived episodes of the Nature Podcastare no more than strings of ones and zeros. Within the computer’s hard drive each one or zero, known as a ‘bit’ of data, is stored by a tiny magnet. This can point up or down and thankfully stays that way even when power is off. In order to cram more and more data into the same space, computers use ever smaller magnets to represent each bit. Fabian Natterer At the Swiss Federal Institute of Technology in Lausanne, spoke to reporter Lizzie Gibney about his team’s efforts to shrink magnetic storage down to the ultimate limit.
If you think about how small you can make something, there’s a natural end to it and the smallest possibility is the single atom. So, we wanted to try to see whether a single atom can be a stable magnet and whether we can use the stable magnet to also code information – the north pole pointing upwards and pointing downwards like a normal hard drive.
Interviewer: Lizzie Gibney
And just for comparison, what kind of size is each magnetic ‘bit’ in a current hard drive?
Current technology is of the order of a million atoms for one single ‘bit’. Compared to the atomic world, one million is a large number and we have now reached the very end which is a single atom.
Interviewer: Lizzie Gibney
So what did you use then as the little tiny magnet in your experiment?
So it’s a magnetic atom from the so-called rare-earth elements. In this case it was holmium and we used it because there was a report that showed that it had some kind of mechanic stability.
Interviewer: Lizzie Gibney
Because when you code that information you want it to stay there.
Exactly. So a hard drive has no use if the information that you code into it is lost after some time.
Interviewer: Lizzie Gibney
How long in your experiments were you able to maintain that magnetic state of your single atom?
You can only look at one atom at a time and so far all the atoms that we were looking at were stable for hours. And our instrument can only run for so long so we cannot really characterise the real lifetime of the state but for our world it is almost as if it were infinite.
Interviewer: Lizzie Gibney
So at that kind of level, when we’re talking about single atoms, how do you go about actually writing the information, encoding your ‘bits’ and then reading it out again.
So we work in the atomic world, so all of our tools are atomic-sized. We use a so-called scanning tunnelling microscope which is essentially a very sharp metal needle. The very apex is one single and this is our microscope that we use to look at the atoms on the surface so we can move along the surface and if there’s an atom we can sense this atom and also look at the electronic properties they have by having a current run through them.
Interviewer: Lizzie Gibney
And then you can use that current to make the atom either a one or a zero.
Exactly, and we can do this in a controlled way. We can use another tool that we have; it’s an atomic-sized MRI machine and essentially it measures magnetic fields and this little atom that we can have on the surface can measure these magnetic properties of our single atoms. So we have two independent ways.
Interviewer: Lizzie Gibney
Well it sounds very promising anyway for someday actually being able to use these single atoms in some kind of storage device. What have you made so far with them?
First of all we were showing that they were very stable and then we thought, okay, let’s build a prototypical atomic memory – so we have two sitting next to each other and this is, for us, a proto-typical two-bit atomic memory and then we were coding this memory. So we were putting information into it and every time it remained stable in this kind of state.
Interviewer: Lizzie Gibney
And do you think it will be easy to scale this up? Obviously, a two bit memory is very impressive but it’s still a pretty tiny amount of information that you can store.
For us, it’s a proof of concept. In principle it should be possible to make a larger array of these single atom magnets.
Interviewer: Lizzie Gibney
And if you were able to do that, what kind of level of data storage could you achieve?
You could reach something of the order of 1,000 terra-bits per square inch. So, the recent technology is at a magnitude of about one terra-bit per square inch. So it would be a thousand times less than what you can reach if you go to the very atomic level.
Interviewer: Lizzie Gibney
And does having these single atom magnets allow you to do any new kinds of research?
Yes, so we are very excited because for us it’s like playing Lego; it’ magnetic Lego. We now have stable magnets and we have ways of moving them around so we can position them on the surface with our microscope. And now we can play with this atomic Lego; we can put them together and have them interact with each other and observe what kind of new properties emerge from this interaction.
Interviewer: Lizzie Gibney
So it’s a bit like we did in school playing with iron filings and a magnet, but at the atomic level.
Exactly, and these kind of new properties are very exciting for us scientists to study. So we are very excited about this as a future goal.
Interviewer: Adam Levy
That was Fabian Natterer who was on the line from Lausanne, Switzerland. Check out his paper at nature.com/nature.
Interviewer: Kerri Smith
Coming up in the News, artificial intelligence takes on poker and IBM release a cloud quantum computer. But now for a short, sweet shot of science; it’s the Research Highlights with Corie Lok.
[Jingle]
Interviewer: Corie Lok
Drones are often built with stiff, bulky frames for protection, but they tend to break during high speed collisions. To make these devices more crash proof, researchers looked to the insect world for inspiration. Many flying insects such as wasps have wings that deform when they crash into plants so researchers in Switzerland made a 50 gram drone with wings that behave in a similar way. The wings are made of flexible fibreglass and during flight are locked into place with magnetic joints. During a collision, the joints release, allowing the wings to bend, absorb the shock, and then snap back into place. This allows the drone to regain its shape after a crash landing. You can learn more about the work in the journal IEEE Robotics and Automation Letters.
[Jingle]
A genomic melt down might have led to the demise of the last woolly mammoths. These giant animals died off around ten thousand years ago in North America and Siberia, but a tiny population on an island in the Arctic Ocean managed to hang on for another 6,000 years. Researchers compared the genome sequences of an island mammoth with that of a mainland mammoth. They found that the island animal had many more harmful genetic mutations such as deleted genes. These mutations may have affected the animal’s social behaviour and sense of smell and could have even made their coat look silky and translucent. The researchers conclude that conserving small, isolated populations of endangered animals may no save them from genetic harm. You can find the study in the journal Lost Genetics.
[Jingle]
Interviewer: Kerri Smith
The origin of life provides some of the biggest questions in Biology. How and when did the first cells evolve and what did they look like? The usual method of finding out is to look for traces of these life forms in ancient rocks but rocks rarely survive for billions of years without being squished, melted or otherwise metamorphosed. That doesn’t stop intrepid scientists from looking. One such scientist was Dominic Papineau. He was excited to examine a very old vein of rock in Quebec in his native Canada and he brought some samples back to his lab at University College London. Inside these rocks were microscopic fossils. He described these micro-fossils in a paper last week and he’s convinced they’re amongst the oldest traces of life on earth. We sent Shamini Bundell along to his lab to examine the ancient rocks.
So, this is called, typically, a Jasper. It is very red in colour and this red here, this is where we find the micro-fossils.
Interviewer: Shamini Bundell
And the really special thing about these rocks is the age. So how old is this piece of rock that I’m holding now?
So this Jasper has a minimum age of 3,770 million years old. So this rock really formed within a few hundred million years of the formation of the earth.
Interviewer: Shamini Bundell
So this Jasper, you found in Quebec, you brought it back to your lab, and then what was the next step in investigating it?
So the first step is to look at it in the optical microscope. That’s when Matt, my PhD student, started to identify specific targets that had potential biological signatures.
Interviewer: Shamini Bundell
And what kind of things did he see that gave you the idea that you might have something exciting here?
So, Matt, at some point, he said, hey I found a lot of filaments in this particular thin section.
Interviewer: Shamini Bundell
So you have a piece of the Jasper on a slide here in the microscope.
Yes, the feature that we see here is akin to the terminal knobs of iron-oxidising bacteria.
Interviewer: Shamini Bundell
One of the things that makes us identify it as a terminal knob is that there’s two filaments shooting off from it and both of these have these twisted structures composed of haematite, these spiral structures. You can see when I change the focus here on the microscope…
Interviewer: Shamini Bundell
Oh yeah, so the red is sort of spiralling round, out from that central blob.
Yes.
Interviewer: Shamini Bundell
And how exactly are these blobs and filaments created by bacteria?
If they are identical or very similar to the modern iron-oxidising bacteria, these twisted filaments would be excreted product from cells that were living here terminal knob.
Interviewer: Shamini Bundell
So, this is bacteria poo?
A combination of excreted, mineralised, undesired material, yes.
Interviewer: Shamini Bundell
So you found all these amazing structures that look very similar to structures produced by modern organisms but what do you know about the microbes that might have produced them?
So they’re very similar to these filament’s iron-oxidising bacteria that live in a wide range of environments but those that are most relevant to our study come from hydro-thermal vent systems.
Interviewer: Shamini Bundell
And what was so exciting about discovering that?
These observations imply that there are organisms similar to those living today around hydro-thermal vent systems that existed back then – all the way back to the beginning of the sedimentary rock record.
Interviewer: Shamini Bundell
Before this paper, what ideas did we have about how long ago life first evolved?
So most biology textbooks would put evidence for the oldest life at 3.5 billion years ago. There are rocks that have been suggested to contain evidence of early life but there’s been a lot of debate in that field because it’s a difficult problem. We need to have several independent lines of evidence to support a claim like this.
Interviewer: Shamini Bundell
There was a paper by Allen Nutman from last year.
This paper by Allen Nutman reported the newly discovered stromatolites from south-west Greenland.
Interviewer: Shamini Bundell
And what is a stromatolite?
A stromatolite is a rock built by micro-organisms. They grow in layers and as they grow they precipitate minerals and when they die their bodies themselves become mineralised… very beautiful structures. We find them in places like Shark Bay in Western Australia.
Interviewer: Shamini Bundell
And how old was that find?
So the minimum age of the stromatolites in Greenland is 3.71 billion years.
Interviewer: Shamini Bundell
So that’s very similar to the piece of Jasper that you showed me earlier and the microbes that you think made your micro-fossils – would they have been similar to the microbes that were making the stromatolites that Nutman found?
They were not similar. Their discovery is for life living in shallow marine environments, whereas we discovered micro-fossils living in deep-sea hydro-thermal vents. So, both our discoveries are more significant when combined together because they implicate that life had diversified significantly more rapidly than biologists thought.
Interviewer: Shamini Bundell
Are we ever going to find traces of life from long enough ago to answer the really hard questions about how the very first cell evolved and things like that?
We don’t have the samples on earth to answer that particular question but some people have suggested that because there was a period of heavy bombardment on the early earth, that potentially some meteorites from earth have landed on the moon, and that may have these pieces of the earliest pre-biotic maybe, even, evolution of the carbon cycle.
Interviewer: Kerri Smith
That was Dominic Papineau of University College London showing off his ancient micro-fossils to Shamini Bundell. You can read the full paper in the March 2ndissue of Nature, still available at nature.com/nature.
Finally this week, it’s the News and I have not one but two of Nature’s finest joining me in the studio. Lizzie Gibney is here. We’re going to talk about artificial intelligence and then later on, Davide Castelvecchi and I will have a quick chat about quantum computing, so some classic News Chat topics coming up. Now, Lizzie, artificial intelligences have mastered games like chess and like Go, and even a few video games, but what’s different about your report this week?
Interviewee: Lizzie Gibney
So these are A.I.s from two different groups and what they’ve done is they’ve beaten humans but this time at Poker, or at least a particular kind of Poker which is called Texas Holdem. It’s a game where you have some public cards and you have some private cards and you make up your best hand out of those. So these two groups have been rivals for about the past ten years and one of them, from the University of Alberta, in 2015 managed to crack one version where you have limits on the betting, and this time they’ve managed to both beat professional humans at the version where there’s no limits which is a lot harder game.
Interviewer: Kerri Smith
And how does Poker in general differ from these games like chess and Go?
Interviewee: Lizzie Gibney
So, in a game like Go, you have all the information available to you and the same with something like chess: you have a board and you have all the pieces on it. In something like Poker, it’s what they call an imperfect information game or an incomplete information game because you are making your decisions based on not only what you can see in front of you, but what you think your opponent has. So, you’re having to think about what you think they have, what you think they think you have, and all of this is based on your previous betting and your strategies about that.
Interviewer: Kerri Smith
That sounds a lot more like real life decision making, frankly.
Interviewee: Lizzie Gibney
Exactly, so there are a lot of real life analogies with things like auctions or financial negotiation, and even medical diagnosis, so that’s where people are really interested in trying to solve these kinds of imperfect information games.
Interviewer: Kerri Smith
And they’ve been at it for years. What sense do you get about how the two teams interact? They must go to the same meetings and discuss these algorithms a little bit.
Interviewee: Lizzie Gibney
Yeah, I think they probably do. They are certainly the two big names in this field and in fact, the person who leads the Alberta team used to work for the Carnegie Mellon team, who’s the other group, so yes, I’m sure there’s quite a lot of rivalry going on between them.
Interviewer: Kerri Smith
And in terms of how the A.I. actually works for something like Poker versus something like Go or chess. Are there differences in how they’ve had to get this algorithm…?
Interviewee: Lizzie Gibney
Yeah, so the added complexity of it being an imperfect information game really does make a difference. So, historically when trying to play Poker, AIs have had to, before the game, work out a whole decision tree, but clearly that’s impossible so they do a much, much smaller one. And that they map onto the game that they’re actually playing so it won’t actually be exact and that’s why they haven’t been able to play it all that well. Both of the two AIs now are both actually able to compute their solutions live, during the game, which is a big, big difference.
Interviewer: Kerri Smith
One of the classic features of the game is just that humans bluff and they lie to each other. Are these computers able to do anything like that?
Interviewee: Lizzie Gibney
Yeah, well they absolutely bluff but it’s not… we think of it as something really special and to do with psychology and reading your opponent, but really it’s just a mathematical strategy, right. All it’s doing is trying to ensure that your opponent doesn’t know what cards you’re holding because of how you’ve played in the past so you’ve got to through them off by just changing your betting strategy and sometimes you bet on cards that are rubbish. And so that’s all it is for a computer.
Interviewer: Kerri Smith
No facial tics or anything to give it away.
Interviewee: Lizzie Gibney
That’s true. I suppose maybe they’re less likely to give things away.
Interviewer: Kerri Smith
Do you happen know whether any of the human professional players who play Poker I suppose to win money but also for enjoyment, are any more or less bored by playing an AI that’s just kind of playing not to lose?
Interviewee: Lizzie Gibney
Some of them are starting to train against AIs. Only now are the AIs getting good enough that it’s worth doing really, but they have started to train against them, a bit like again with Go, to learn strategies they might not have otherwise applied. But generally you’re not going to be playing against, or you shouldn’t be playing against a Poker player AI because they’re not actually allowed in online casinos. So I don’t think anyone is yet thinking that this is going to really, dramatically change their professional careers.
Interviewer: Kerri Smith
Alright, well thank you Lizzie for briefing us on that bit of the future happening today. Davide, the other bit of the future that’s happening today – this is how these stories feel to me anyway – is that IBM have made an announcement about the next wave of quantum computing, haven’t they?
Interviewer: Davide Castelvecchi
Yes, so this is something that has been in the making for a while. IBM has sort of kept its cards hidden…
Interviewer: Kerri Smith
Very nice.
Interviewer: Davide Castelvecchi
But only in part, and they have unveiled this quantum computing service that will be available on the cloud: sort of like any company can book computing facilities from Amazon or Microsoft. But my favourite thing about it is that it’s called IBM Q which is reminiscent of the James Bond king of technology and gadgets.
Interviewer: Kerri Smith
The difference being that Q’s gadgets are – at least in the film world – real and accessible and they work whereas IBM Q doesn’t appear to be able to do any of those magical things that we think quantum computers ought to be able to do, quite yet.
Interviewer: Davide Castelvecchi
Well, it will be doing magical things; it will just be doing very simple magical things. For a quantum computer to be better than a traditional computer, classical computer, it will need at least fifty quantum bits – these qubits, the units of quantum computation. At this stage it’s not clear yet – this IBM Q – it’s not clear how many qubits it will have. It will maybe be seven or nine or around that range, at least in the first iteration, and then IBM says that they will plan to ramp it up.
Interviewer: Kerri Smith
What will IBM Q be doing for people once it goes live then? And who do you anticipate might sign on to use its services?
Interviewer: Davide Castelvecchi
Well in part it could be researchers who do research on quantum algorithms and quantum programming who want to practice on an actual machine. For example, computer scientists who develop an algorithm and they want to test it in real life. But also – and this is the primary reason why IBM is doing it – there could be a lot of companies that want to start experimenting with these machines and to see if they can actually be useful to them because until now quantum computers have been kind of an answer in search of a question. There isn’t a lot of demand for a quantum computing facility like this yet because companies don’t quite know what to do with it. And so IBM hopes to have a learning process together with other commercial partners where they will find ways to put it to use.
Interviewer: Kerri Smith
As you put it in the top of the story: if you build it, they will come.
Interviewer: Davide Castelvecchi
Yes.
Interviewer: Kerri Smith
This could lead I suppose, one day, to this new-sounding field – at least new to me anyway – about quantum coding, right? I mean, you can’t program these things in the way that you would program a traditional computer.
Interviewer: Davide Castelvecchi
That’s right, it’s completely different. And also, one thing that is interesting to me is that it’s one thing to have algorithms – quantum algorithms – that work in theory… it’s another thing to make them work on an actual machine.
Interviewer: Kerri Smith
So the IBM Q announcement was made on Monday but has anything like this ever existed before?
Interviewer: Davide Castelvecchi
Yeah, so in fact this builds on an existing initiative by IBM called Quantum Experience where a very, very basic quantum computer was made available to anyone who wanted to try and program it.
Interviewer: Kerri Smith
What’s the best thing is that the MIT students, having taken their course in quantum computing, were applying their skills to online Poker in the evenings. Davide and Lizzy, thank you very much for joining me.
Interviewer: Adam Levy
That’s all we’ve got time for this week. For more on IBM Q check out the website resech.ibm.com/ibmq where you can also give Quantum Experience a go. Davide’s News story, and Lizzie’s on Poker-playing-AIs are both free to read at nature.com/news where you’ll also find a Comment piece about making quantum computers commercially viable.
Interviewer: Kerri Smith
Next time: if planes ran on biofuels. I’m Kerri Smith.
Interviewer: Adam Levy
And I’m Adam Levy.
[Jingle]
Interviewer: Adam Levy
This week: storing data in the tiniest of spaces.
Interviewee: Fabian Natterer
Current technology is of the order of a million atoms and we have now reached the very end which is a single atom.
Interviewer: Kerri Smith
And, a study of microorganisms in Neanderthal mouths paints a graphic picture of their encounters with humans.
Interviewee: Laura Weyrich
And so it actually suggests that there might have been some sort of spit-swapping between humans and Neanderthals.
Interviewer: Adam Levy
Plus, we chat to the scientists who uncovered the world’s oldest fossils. This is the Nature Podcastfor March the 9th2017. I’m Adam Levy.
Interviewer: Kerri Smith
And I’m Kerri Smith.
[Jingle]
Interviewer: Kerri Smith
Adam, open wide please. I need to look at your dental calculus.
Interviewer: Adam Levy
My what?
Interviewer: Kerri Smith
Your dental calculus: calcified dental plaque. Basically, bits of food and microbes that live among your pearly whites.
Interviewer: Adam Levy
No, I’d really rather you didn’t. My teeth are spotless and you’re definitely not a dentist.
Interviewer: Kerri Smith
Correct! But I am very interested in what dental plaque can reveal about diet and health. A new study in Naturetakes a look at some dental calculus from Neanderthal teeth: one individual from Belgium and one from Spain. They used DNA sequencing on the material furring up and solidifying on the Neanderthal teeth. It’s given them a lot of information about food, bacteria, even medicinal compounds – really anything these guys put in their mouths.
Interviewer: Adam Levy
Sounds fascinating. Why not get the authors to open up about that?
Interviewer: Kerri Smith
Well, handily for you, that’s what Ewen Callaway has done. He and author Laura Weyrich, who’s at the University of Adelaide, had a wide open discussion about mouth health and the big differences between Neanderthal diets in different places. Here’s Ewen talking to Laura.
Interviewer: Ewen Callaway
Your team found really striking differences between Neanderthals living in two different parts of Europe, with regards to their diet, right?
Interviewee: Laura Weyrich
Yeah, absolutely. So we looked at Neanderthals in both Belgium as well as Spain, and we also looked at another Neanderthal in Italy but it didn’t happen to result in good DNA sequences. But the Neanderthals in Belgium are what we would consider the more classic Neanderthals. These guys were eating a lot of meat so we found evidence of DNA in their teeth for woolly rhinoceros as well as mouflon sheep which are a wild type of sheep that live in Europe at that time. But that’s very, very different from what we see in Spain. The Spanish Neanderthals are almost vegetarian, so we don’t see any evidence of meat consumption present in their calculus. We only see things like pine nuts, moss, even tree bark. They were eating basically what they could find in their very forest-like ecology that they were living in. These would really have been the true Palaeo Diet Neanderthals. We joke about that a lot, that we should write the true Palaeo Diet cookbook and include moss and tree bark and pine nuts.
Interviewer: Ewen Callaway
So we found, in Neanderthal plaque, evidence that they were foragers in one area and meat-eaters in another area: true ‘locavores’. What else did we learn from their plaque?
Interviewee: Laura Weyrich
So, when we were looking for dietary information we incidentally found some evidence that provided insight into the medicine or the medicinal practices that these Neanderthals may have had. There was one Neanderthal in particular from El Sidrón Cave in Spain that was suffering, probably, some really nasty side effects from a dental abscess, as well as a gastro-intestinal pathogen microsporidia. The dental abscess is evident in the skeletal material. You can see where that abscess was eroding its jaw but the microsporidia was something we were able to pick up in the dental calculus. So it’s likely that he was suffering from a diarrhoea or some kind of gastro-intestinal type bug that would have made him noticeably sick.
So we were actually able to go in and look for, not only pathogens that would have made him sick, but then also how he might be treating those pathogens. And so we found evidence for him consuming poplar bark and poplar bark contains salicyclic acid which is the active ingredient in aspirin. It would have treated his pain from this disease. We also found evidence for him consuming moulded herbaceous material, and this moulded herbaceous material contained Penicillium which is a natural source of antibiotic penicillin. So we probably don’t think Neanderthals thought to seek out this particular mould and consume it as a medicinal act but more so that they may have known that eating mouldy material might make you feel better.
Interviewer: Ewen Callaway
Are you able also to get a snapshot of not just a couple of pathogens but of the overall microbiome of the teeth?
Interviewee: Laura Weyrich
We certainly are and for me that’s the most exciting part about this paper. This is actually the first microbiome of an extinct species as well as of an extinct hominid and it really was a lot of work to try to reconstruct these healthy, beneficial microorganisms that are present in the plaque. In this particular study we’re able to show that chimpanzees and vegetarian Neanderthals and gatherer humans used to share a microbiome, and it’s really the introduction of meat into the diet in both Neanderthals and ancient humans that causes the first change away from that shared hominid microbiome a long time ago. So it really affords us a bit of a window into our own health as well as Neanderthals.
Interviewer: Ewen Callaway
So what more can we learn from looking at ancient plaque?
Interviewee: Laura Weyrich
So I think that there’s a lot of stories that are present in each individual bacterial species that we find in the plaque and in this paper we really talk about one bacterial species in particular and it’s not really a bacteria, it’s an archaea. We can actually use the genomic mutations present in that species we find in Neanderthals and compare that to the same species that we find in modern humans and that tells us when one of the strains could have been introduced into one of those species. So here we actually identify that this archaea, Methanobrevibacter oralis, was actually introduced into Neanderthals about 120,000 years ago. And this is around the same time that interbreeding began occurring between Neanderthals and humans, and so it actually suggests that there might have been some sort of spit-swapping between humans and Neanderthals while these interbreeding interactions were going on. And that’s quite a unique thing to think about because many of these interbreeding interactions have been described as very rough, very quick encounters, not something that would have been very sensual or very intimate, but certainly if you’re swapping spit between species, there’s kissing going on or at least there’s food sharing, direct food sharing, which would suggest that these interactions were much more friendly and much more intimate than anybody ever possibly imagined.
Interviewer: Ewen Callaway
I didn’t catch that from your paper that you could imagine how Neanderthals might have kissed; that’s quite an insight!
Interviewee: Laura Weyrich
Yeah, it is. [Laughs]
Interviewer: Ewen Callaway
Wow, yeah, that’s really fascinating. Do any other aspects of your results help you imagine Neanderthals as individuals, as real people?
Interviewee: Laura Weyrich
It really paints a different picture almost of the personalities, of really who they are. And this technique allows us to go in a step further and really look in their medicine cabinets, look in their food cabinets, figure out where they were going and how they were viewing the world. It’s a very different take on ancient DNA and how it’s being used today.
Interviewer: Kerri Smith
That was Laura Weyrich talking to Ewen Callaway. Next, Laura wants to sequence food encrusted teeth from any ancient human relative she can find so her team can compare the mouthy microbiomes and prandial preferences of our family tree. Find this paper of our two Neanderthals at nature.com/nature. Ewen has also written a News story which you can read for free at nature.com/news.
Interviewer: Adam Levy
We’ve got tiny data storage facilities and tiny ancient fossils coming up in the show. Plus the Highlights cover woolly mammoths and clumsy drones.
Interviewer: Kerri Smith
First, an update on a story from last week. You might remember that food scientist Kassem Alsayed Mahmoud made it from his home country, Syria, to Belgium and is working at the Free University of Brussels. But the way the story ended was with his family in Turkey waiting to hear if they can join him. Well, last week Kassem emailed with an update: his family have been granted a visa to join him in Belgium and they should be arriving before the end of this month. Here’s wishing them a smooth trip.
Interviewer: Adam Levy
If you missed Kassem on last week’s show about migration you can find it and the wholeNature Podcast archive at nature.com/nature/podcast.
[Jingle]
Interviewer: Adam Levy
To a computer, your precious holiday photos, spreadsheets, or all those archived episodes of the Nature Podcastare no more than strings of ones and zeros. Within the computer’s hard drive each one or zero, known as a ‘bit’ of data, is stored by a tiny magnet. This can point up or down and thankfully stays that way even when power is off. In order to cram more and more data into the same space, computers use ever smaller magnets to represent each bit. Fabian Natterer At the Swiss Federal Institute of Technology in Lausanne, spoke to reporter Lizzie Gibney about his team’s efforts to shrink magnetic storage down to the ultimate limit.
If you think about how small you can make something, there’s a natural end to it and the smallest possibility is the single atom. So, we wanted to try to see whether a single atom can be a stable magnet and whether we can use the stable magnet to also code information – the north pole pointing upwards and pointing downwards like a normal hard drive.
Interviewer: Lizzie Gibney
And just for comparison, what kind of size is each magnetic ‘bit’ in a current hard drive?
Current technology is of the order of a million atoms for one single ‘bit’. Compared to the atomic world, one million is a large number and we have now reached the very end which is a single atom.
Interviewer: Lizzie Gibney
So what did you use then as the little tiny magnet in your experiment?
So it’s a magnetic atom from the so-called rare-earth elements. In this case it was holmium and we used it because there was a report that showed that it had some kind of mechanic stability.
Interviewer: Lizzie Gibney
Because when you code that information you want it to stay there.
Exactly. So a hard drive has no use if the information that you code into it is lost after some time.
Interviewer: Lizzie Gibney
How long in your experiments were you able to maintain that magnetic state of your single atom?
You can only look at one atom at a time and so far all the atoms that we were looking at were stable for hours. And our instrument can only run for so long so we cannot really characterise the real lifetime of the state but for our world it is almost as if it were infinite.
Interviewer: Lizzie Gibney
So at that kind of level, when we’re talking about single atoms, how do you go about actually writing the information, encoding your ‘bits’ and then reading it out again.
So we work in the atomic world, so all of our tools are atomic-sized. We use a so-called scanning tunnelling microscope which is essentially a very sharp metal needle. The very apex is one single and this is our microscope that we use to look at the atoms on the surface so we can move along the surface and if there’s an atom we can sense this atom and also look at the electronic properties they have by having a current run through them.
Interviewer: Lizzie Gibney
And then you can use that current to make the atom either a one or a zero.
Exactly, and we can do this in a controlled way. We can use another tool that we have; it’s an atomic-sized MRI machine and essentially it measures magnetic fields and this little atom that we can have on the surface can measure these magnetic properties of our single atoms. So we have two independent ways.
Interviewer: Lizzie Gibney
Well it sounds very promising anyway for someday actually being able to use these single atoms in some kind of storage device. What have you made so far with them?
First of all we were showing that they were very stable and then we thought, okay, let’s build a prototypical atomic memory – so we have two sitting next to each other and this is, for us, a proto-typical two-bit atomic memory and then we were coding this memory. So we were putting information into it and every time it remained stable in this kind of state.
Interviewer: Lizzie Gibney
And do you think it will be easy to scale this up? Obviously, a two bit memory is very impressive but it’s still a pretty tiny amount of information that you can store.
For us, it’s a proof of concept. In principle it should be possible to make a larger array of these single atom magnets.
Interviewer: Lizzie Gibney
And if you were able to do that, what kind of level of data storage could you achieve?
You could reach something of the order of 1,000 terra-bits per square inch. So, the recent technology is at a magnitude of about one terra-bit per square inch. So it would be a thousand times less than what you can reach if you go to the very atomic level.
Interviewer: Lizzie Gibney
And does having these single atom magnets allow you to do any new kinds of research?
Yes, so we are very excited because for us it’s like playing Lego; it’ magnetic Lego. We now have stable magnets and we have ways of moving them around so we can position them on the surface with our microscope. And now we can play with this atomic Lego; we can put them together and have them interact with each other and observe what kind of new properties emerge from this interaction.
Interviewer: Lizzie Gibney
So it’s a bit like we did in school playing with iron filings and a magnet, but at the atomic level.
Exactly, and these kind of new properties are very exciting for us scientists to study. So we are very excited about this as a future goal.
Interviewer: Adam Levy
That was Fabian Natterer who was on the line from Lausanne, Switzerland. Check out his paper at nature.com/nature.
Interviewer: Kerri Smith
Coming up in the News, artificial intelligence takes on poker and IBM release a cloud quantum computer. But now for a short, sweet shot of science; it’s the Research Highlights with Corie Lok.
[Jingle]
Interviewer: Corie Lok
Drones are often built with stiff, bulky frames for protection, but they tend to break during high speed collisions. To make these devices more crash proof, researchers looked to the insect world for inspiration. Many flying insects such as wasps have wings that deform when they crash into plants so researchers in Switzerland made a 50 gram drone with wings that behave in a similar way. The wings are made of flexible fibreglass and during flight are locked into place with magnetic joints. During a collision, the joints release, allowing the wings to bend, absorb the shock, and then snap back into place. This allows the drone to regain its shape after a crash landing. You can learn more about the work in the journal IEEE Robotics and Automation Letters.
[Jingle]
A genomic melt down might have led to the demise of the last woolly mammoths. These giant animals died off around ten thousand years ago in North America and Siberia, but a tiny population on an island in the Arctic Ocean managed to hang on for another 6,000 years. Researchers compared the genome sequences of an island mammoth with that of a mainland mammoth. They found that the island animal had many more harmful genetic mutations such as deleted genes. These mutations may have affected the animal’s social behaviour and sense of smell and could have even made their coat look silky and translucent. The researchers conclude that conserving small, isolated populations of endangered animals may no save them from genetic harm. You can find the study in the journal Lost Genetics.
[Jingle]
Interviewer: Kerri Smith
The origin of life provides some of the biggest questions in Biology. How and when did the first cells evolve and what did they look like? The usual method of finding out is to look for traces of these life forms in ancient rocks but rocks rarely survive for billions of years without being squished, melted or otherwise metamorphosed. That doesn’t stop intrepid scientists from looking. One such scientist was Dominic Papineau. He was excited to examine a very old vein of rock in Quebec in his native Canada and he brought some samples back to his lab at University College London. Inside these rocks were microscopic fossils. He described these micro-fossils in a paper last week and he’s convinced they’re amongst the oldest traces of life on earth. We sent Shamini Bundell along to his lab to examine the ancient rocks.
So, this is called, typically, a Jasper. It is very red in colour and this red here, this is where we find the micro-fossils.
Interviewer: Shamini Bundell
And the really special thing about these rocks is the age. So how old is this piece of rock that I’m holding now?
So this Jasper has a minimum age of 3,770 million years old. So this rock really formed within a few hundred million years of the formation of the earth.
Interviewer: Shamini Bundell
So this Jasper, you found in Quebec, you brought it back to your lab, and then what was the next step in investigating it?
So the first step is to look at it in the optical microscope. That’s when Matt, my PhD student, started to identify specific targets that had potential biological signatures.
Interviewer: Shamini Bundell
And what kind of things did he see that gave you the idea that you might have something exciting here?
So, Matt, at some point, he said, hey I found a lot of filaments in this particular thin section.
Interviewer: Shamini Bundell
So you have a piece of the Jasper on a slide here in the microscope.
Yes, the feature that we see here is akin to the terminal knobs of iron-oxidising bacteria.
Interviewer: Shamini Bundell
One of the things that makes us identify it as a terminal knob is that there’s two filaments shooting off from it and both of these have these twisted structures composed of haematite, these spiral structures. You can see when I change the focus here on the microscope…
Interviewer: Shamini Bundell
Oh yeah, so the red is sort of spiralling round, out from that central blob.
Yes.
Interviewer: Shamini Bundell
And how exactly are these blobs and filaments created by bacteria?
If they are identical or very similar to the modern iron-oxidising bacteria, these twisted filaments would be excreted product from cells that were living here terminal knob.
Interviewer: Shamini Bundell
So, this is bacteria poo?
A combination of excreted, mineralised, undesired material, yes.
Interviewer: Shamini Bundell
So you found all these amazing structures that look very similar to structures produced by modern organisms but what do you know about the microbes that might have produced them?
So they’re very similar to these filament’s iron-oxidising bacteria that live in a wide range of environments but those that are most relevant to our study come from hydro-thermal vent systems.
Interviewer: Shamini Bundell
And what was so exciting about discovering that?
These observations imply that there are organisms similar to those living today around hydro-thermal vent systems that existed back then – all the way back to the beginning of the sedimentary rock record.
Interviewer: Shamini Bundell
Before this paper, what ideas did we have about how long ago life first evolved?
So most biology textbooks would put evidence for the oldest life at 3.5 billion years ago. There are rocks that have been suggested to contain evidence of early life but there’s been a lot of debate in that field because it’s a difficult problem. We need to have several independent lines of evidence to support a claim like this.
Interviewer: Shamini Bundell
There was a paper by Allen Nutman from last year.
This paper by Allen Nutman reported the newly discovered stromatolites from south-west Greenland.
Interviewer: Shamini Bundell
And what is a stromatolite?
A stromatolite is a rock built by micro-organisms. They grow in layers and as they grow they precipitate minerals and when they die their bodies themselves become mineralised… very beautiful structures. We find them in places like Shark Bay in Western Australia.
Interviewer: Shamini Bundell
And how old was that find?
So the minimum age of the stromatolites in Greenland is 3.71 billion years.
Interviewer: Shamini Bundell
So that’s very similar to the piece of Jasper that you showed me earlier and the microbes that you think made your micro-fossils – would they have been similar to the microbes that were making the stromatolites that Nutman found?
They were not similar. Their discovery is for life living in shallow marine environments, whereas we discovered micro-fossils living in deep-sea hydro-thermal vents. So, both our discoveries are more significant when combined together because they implicate that life had diversified significantly more rapidly than biologists thought.
Interviewer: Shamini Bundell
Are we ever going to find traces of life from long enough ago to answer the really hard questions about how the very first cell evolved and things like that?
We don’t have the samples on earth to answer that particular question but some people have suggested that because there was a period of heavy bombardment on the early earth, that potentially some meteorites from earth have landed on the moon, and that may have these pieces of the earliest pre-biotic maybe, even, evolution of the carbon cycle.
Interviewer: Kerri Smith
That was Dominic Papineau of University College London showing off his ancient micro-fossils to Shamini Bundell. You can read the full paper in the March 2ndissue of Nature, still available at nature.com/nature.
Finally this week, it’s the News and I have not one but two of Nature’s finest joining me in the studio. Lizzie Gibney is here. We’re going to talk about artificial intelligence and then later on, Davide Castelvecchi and I will have a quick chat about quantum computing, so some classic News Chat topics coming up. Now, Lizzie, artificial intelligences have mastered games like chess and like Go, and even a few video games, but what’s different about your report this week?
Interviewee: Lizzie Gibney
So these are A.I.s from two different groups and what they’ve done is they’ve beaten humans but this time at Poker, or at least a particular kind of Poker which is called Texas Holdem. It’s a game where you have some public cards and you have some private cards and you make up your best hand out of those. So these two groups have been rivals for about the past ten years and one of them, from the University of Alberta, in 2015 managed to crack one version where you have limits on the betting, and this time they’ve managed to both beat professional humans at the version where there’s no limits which is a lot harder game.
Interviewer: Kerri Smith
And how does Poker in general differ from these games like chess and Go?
Interviewee: Lizzie Gibney
So, in a game like Go, you have all the information available to you and the same with something like chess: you have a board and you have all the pieces on it. In something like Poker, it’s what they call an imperfect information game or an incomplete information game because you are making your decisions based on not only what you can see in front of you, but what you think your opponent has. So, you’re having to think about what you think they have, what you think they think you have, and all of this is based on your previous betting and your strategies about that.
Interviewer: Kerri Smith
That sounds a lot more like real life decision making, frankly.
Interviewee: Lizzie Gibney
Exactly, so there are a lot of real life analogies with things like auctions or financial negotiation, and even medical diagnosis, so that’s where people are really interested in trying to solve these kinds of imperfect information games.
Interviewer: Kerri Smith
And they’ve been at it for years. What sense do you get about how the two teams interact? They must go to the same meetings and discuss these algorithms a little bit.
Interviewee: Lizzie Gibney
Yeah, I think they probably do. They are certainly the two big names in this field and in fact, the person who leads the Alberta team used to work for the Carnegie Mellon team, who’s the other group, so yes, I’m sure there’s quite a lot of rivalry going on between them.
Interviewer: Kerri Smith
And in terms of how the A.I. actually works for something like Poker versus something like Go or chess. Are there differences in how they’ve had to get this algorithm…?
Interviewee: Lizzie Gibney
Yeah, so the added complexity of it being an imperfect information game really does make a difference. So, historically when trying to play Poker, AIs have had to, before the game, work out a whole decision tree, but clearly that’s impossible so they do a much, much smaller one. And that they map onto the game that they’re actually playing so it won’t actually be exact and that’s why they haven’t been able to play it all that well. Both of the two AIs now are both actually able to compute their solutions live, during the game, which is a big, big difference.
Interviewer: Kerri Smith
One of the classic features of the game is just that humans bluff and they lie to each other. Are these computers able to do anything like that?
Interviewee: Lizzie Gibney
Yeah, well they absolutely bluff but it’s not… we think of it as something really special and to do with psychology and reading your opponent, but really it’s just a mathematical strategy, right. All it’s doing is trying to ensure that your opponent doesn’t know what cards you’re holding because of how you’ve played in the past so you’ve got to through them off by just changing your betting strategy and sometimes you bet on cards that are rubbish. And so that’s all it is for a computer.
Interviewer: Kerri Smith
No facial tics or anything to give it away.
Interviewee: Lizzie Gibney
That’s true. I suppose maybe they’re less likely to give things away.
Interviewer: Kerri Smith
Do you happen know whether any of the human professional players who play Poker I suppose to win money but also for enjoyment, are any more or less bored by playing an AI that’s just kind of playing not to lose?
Interviewee: Lizzie Gibney
Some of them are starting to train against AIs. Only now are the AIs getting good enough that it’s worth doing really, but they have started to train against them, a bit like again with Go, to learn strategies they might not have otherwise applied. But generally you’re not going to be playing against, or you shouldn’t be playing against a Poker player AI because they’re not actually allowed in online casinos. So I don’t think anyone is yet thinking that this is going to really, dramatically change their professional careers.
Interviewer: Kerri Smith
Alright, well thank you Lizzie for briefing us on that bit of the future happening today. Davide, the other bit of the future that’s happening today – this is how these stories feel to me anyway – is that IBM have made an announcement about the next wave of quantum computing, haven’t they?
Interviewer: Davide Castelvecchi
Yes, so this is something that has been in the making for a while. IBM has sort of kept its cards hidden…
Interviewer: Kerri Smith
Very nice.
Interviewer: Davide Castelvecchi
But only in part, and they have unveiled this quantum computing service that will be available on the cloud: sort of like any company can book computing facilities from Amazon or Microsoft. But my favourite thing about it is that it’s called IBM Q which is reminiscent of the James Bond king of technology and gadgets.
Interviewer: Kerri Smith
The difference being that Q’s gadgets are – at least in the film world – real and accessible and they work whereas IBM Q doesn’t appear to be able to do any of those magical things that we think quantum computers ought to be able to do, quite yet.
Interviewer: Davide Castelvecchi
Well, it will be doing magical things; it will just be doing very simple magical things. For a quantum computer to be better than a traditional computer, classical computer, it will need at least fifty quantum bits – these qubits, the units of quantum computation. At this stage it’s not clear yet – this IBM Q – it’s not clear how many qubits it will have. It will maybe be seven or nine or around that range, at least in the first iteration, and then IBM says that they will plan to ramp it up.
Interviewer: Kerri Smith
What will IBM Q be doing for people once it goes live then? And who do you anticipate might sign on to use its services?
Interviewer: Davide Castelvecchi
Well in part it could be researchers who do research on quantum algorithms and quantum programming who want to practice on an actual machine. For example, computer scientists who develop an algorithm and they want to test it in real life. But also – and this is the primary reason why IBM is doing it – there could be a lot of companies that want to start experimenting with these machines and to see if they can actually be useful to them because until now quantum computers have been kind of an answer in search of a question. There isn’t a lot of demand for a quantum computing facility like this yet because companies don’t quite know what to do with it. And so IBM hopes to have a learning process together with other commercial partners where they will find ways to put it to use.
Interviewer: Kerri Smith
As you put it in the top of the story: if you build it, they will come.
Interviewer: Davide Castelvecchi
Yes.
Interviewer: Kerri Smith
This could lead I suppose, one day, to this new-sounding field – at least new to me anyway – about quantum coding, right? I mean, you can’t program these things in the way that you would program a traditional computer.
Interviewer: Davide Castelvecchi
That’s right, it’s completely different. And also, one thing that is interesting to me is that it’s one thing to have algorithms – quantum algorithms – that work in theory… it’s another thing to make them work on an actual machine.
Interviewer: Kerri Smith
So the IBM Q announcement was made on Monday but has anything like this ever existed before?
Interviewer: Davide Castelvecchi
Yeah, so in fact this builds on an existing initiative by IBM called Quantum Experience where a very, very basic quantum computer was made available to anyone who wanted to try and program it.
Interviewer: Kerri Smith
What’s the best thing is that the MIT students, having taken their course in quantum computing, were applying their skills to online Poker in the evenings. Davide and Lizzy, thank you very much for joining me.
Interviewer: Adam Levy
That’s all we’ve got time for this week. For more on IBM Q check out the website resech.ibm.com/ibmq where you can also give Quantum Experience a go. Davide’s News story, and Lizzie’s on Poker-playing-AIs are both free to read at nature.com/news where you’ll also find a Comment piece about making quantum computers commercially viable.
Interviewer: Kerri Smith
Next time: if planes ran on biofuels. I’m Kerri Smith.
Interviewer: Adam Levy
And I’m Adam Levy.
[Jingle]