Nature Podcast

This is a transcript of the 03rd April edition of the weekly Nature Podcast. Audio files for the current show and archive episodes can be accessed from the Nature Podcast index page (, which also contains details on how to subscribe to the Nature Podcast for FREE, and has troubleshooting top-tips. Send us your feedback to


The Nature Podcast, in association with Bio-Rad. Working with primary cells? Increase transfection efficiencies by optimizing in the Gene Pulser MXcell. On the web at

End Advertisement

Adam Rutherford: Coming up, will a new piece of telescope kit allow us to seek out Earth-like planets in the darkest reaches of space?

Ronald L. Walsworth: I think yes, the astro-comb together with very good spectrographs and telescopes will allow those to be discovered.

Kerri Smith: And our podium presenter takes climate change scientists to task.

Roger Pielke, Jr.: The IPCC is too optimistic in assuming that without any action by policymakers, new technology will dramatically reduce the growth of future emissions.

Kerri Smith: This is the Nature Podcast, I'm Kerri Smith.

Adam Rutherford: And I am Adam Rutherford. First this week, the evidence is pretty solid that people who smoke are at a bigger risk from lung cancer. That link makes smoking a leading cause of preventable death in the world with around 5 million deaths every year. In a trio of new papers in Nature and Nature Genetics, three different groups have zoomed in on a spot on the genome that gives the best correlation yet between a genetic variant and lung cancer. Kari Stefansson and his team from deCODE Genetics a biotech company based in Iceland they found a link between a specific gene locus and the tendency to become addicted to nicotine. Stefansson anecdotally sees this correlation in his own family. Nature 452, 638–642 (3 April 2008), Nature 452, 537–538 (3 April 2008) , Nature 452, 633–637 (3 April 2008); Nature Genetics , (2 April 2008);

Kari Stefansson: You inherit the tendency to become addicted to nicotine. There is always a contribution both by the environment and by the genetics. My father, my late father was an avid smoker. When the big propaganda against cigarette smoking began, he said I'm not going to let this people tell me how to behave and he threw away the pipe and started to smoke cigarettes and he died from lung cancer at the age of 67, unfortunately. I tried to smoke as a teenager, I failed completely and I just became nauseated and threw up, but then I have a daughter who is an avid smoker and she smokes, even holds the cigarettes a little bit like her late grandfather.

Adam Rutherford: The three papers report the same result. Using a genome-wide study of genetic variation, the risk of lung cancer is closely associated with a region on the long arm of chromosome 15. This spot is known to harbour a gene of particular interest to smoking researchers. Here's Stephen Chanock from the National Cancer Institute in Maryland.

Stephen J. Chanock: So now this is the excitement we see, aha! This particular nicotine acetylcholine receptor, vow! That's a great gene. People have been interested in that for nicotine addiction for quite sometime and there have been some very exciting reports. So the genome-wide study has pointed directly to a place. It's been a place of interest for an extended period of time. All we know right now is that smoking and lung cancer are showing up in this particular region here. And since we know smoking and lung cancer are so closely related to each other and in the studies that are reported almost all the lung cancer patients have been smokers, we can't fair that out yet.

Adam Rutherford: Chanock along with David Hunter of Harvard University has co-written a news and views article on the results. The studies, one led by Stefansson, the others by Paul Brennan at the International Agency for Research and Cancer in France and Christopher Amos at the MD Anderson Cancer Centre in Texas, all point to the same spot on the genome, but they have drawn different conclusions about what's causing that association with cancer.

David J. Hunter: The fascinating thing here is that the three papers, two in Nature and one in Nature Genetics have all found robust associations of the same chromosomal loci or area of the chromosome with lung cancer, but they have reached rather different conclusions on whether or not that is mediated through a propensity to smoke more cigarettes or be addicted to nicotine. So two of the papers suggest that the association is really with lung cancer and not with smoking and the other paper suggests that the association is really with smoking and that's the reason that lung cancer risk is increased.

Adam Rutherford: David Hunter, despite the disagreement the new data unveils a potentially powerful weapon in the antismoking arsenal. The dangers of smoking are not controversial. It is a killer, but Hunter thinks we should express caution in formulating public health policy.

David J. Hunter: All people ask whether they should go and get tested for these and other variants and I think right now, we are not ready to recommend testing for such a large number of gene variants across so many conditions, where the data is still in evolution, we are going to find out more. The biggest problem is this is a deluge of information and we've got no experience in communicating that information to individuals, in understanding what people's reaction to it is, you know understanding whether it might expose them to in some societies health insurance discrimination or life insurance discrimination, so I think if people wish to go and get tested there are some companies setting up to do it and that's people's own choice, but I don't think we are ready to recommend that as a clinical activity at this point.

Adam Rutherford: But Kari Stefansson shares his father's defiance.

Kari Stefansson: What is important to keep in mind here is that it should be the decision of the individual whether or not they want to learn about the risk, people should at least be given an opportunity for learning their risk and then deciding on their own, whether or not it is enough for them to change their behaviour and I resent the patronizing view that people are not ready to learn about their risks. A very substantial part of who we are is the risk that we have of the common diseases that kill most people. So I think that we are always ready to learn more about ourselves.

Adam Rutherford: Kari Stefansson from deCODE Genetics. His team's paper along with the study by Paul Brennan's group and Hunter and Chanock's news and views article all available on the Nature web site along with the link to the third study from Christopher Amos' team, that paper is in Nature Genetics.

Kerri Smith: Coming up in just a moment what monkeys in virtual reality can tell us about how the brain sees in three dimensions, but first the podium. In Nature this week Roger Pielke Junior, Professor of Environmental Studies at the University of Colorado argues that the researcher's task with providing objective information about climate change, the IPCC have too rosy a view of how to deal with the problem.  452, 531–532 (3 April 2008)

Roger Pielke, Jr.: If you think that the challenge of climate policies is daunting, I may have some bad news. Reducing the emissions of carbon dioxide over the coming century is going to be more challenging than society has been led to believe. My colleagues and I believe that the technological challenges of reducing carbon dioxide emissions have been significantly underestimated by the Intergovernmental Panel on Climate Change. The IPCC is too optimistic in assuming that without any action by policymakers, new technologies will dramatically reduce the growth of future emissions. Around the world, the amount of carbon dioxide emissions per unit of energy consumption is already higher than the IPCC predicts because of rapid economic development, particularly in Asia where economies are currently becoming more energy intensive. Their energy to man is being met by conventional fossil fuel technologies. Consider that every year for the rest of the current decade, China is expected to add emissions equal to all of Germany or the UK to its total annual carbon dioxide output. Such intensive development is expected to continue for decades throughout Asia and eventually in Africa. The IPCC scenarios for future emissions growth separates those that will occur spontaneously or in other words in the absence of any climate policies and those that are the policy-driven as responses to climate change. Our attention is naturally drawn away from those emissions reductions that are expected to occur automatically and we focus our attention instead only on those covered by climate policies. This separation hides the full magnitude of the technology challenge of stabilizing the carbon dioxide in the atmosphere. Projections should not assume automatic emissions reductions, they should instead place the full scope of the challenge before us. The most recent IPCC assessment report assumes that two-thirds or more of the emissions reductions required to stabilize carbon dioxide concentrations at an acceptable level will occur automatically. This is unlikely to happen under current policies. Indeed right now, we are moving in the opposite direction. Technological innovation is necessary in climate policy, but to what degree should policy focus on directly motivating such innovation. The IPCC plays a risky game in assuming that automatic advances will achieve most future emissions reductions. We should focus instead on those conditions necessary and sufficient for such innovations to occur. So perhaps the news that we bring is not so bad after all, because it is only with a clear eyed view of the full scope of the mitigation challenge that we can hope to adopt effective policies. We hope that our analysis is one step towards such a clear eyed view.

Kerri Smith: Roger Pielke Junior of the University of Colorado in Boulder. You can read his commentary at

Adam Rutherford: The human effects on modern climate change are well known and publicized, but a record of climate past also received some attention in Nature this week, the evidence: dust. Here's Mike Hopkin. Nature 452, 616–619 (3 April 2008)

Michael Hopkin: Dust, might not sound like the most fascinating thing to study, but it has the potential to tell us a lot about the behaviour of Earth's past climate. That's why climatologists have just finished a painstaking study of dust trapped in Antarctic ice over the past 800,000 years. The ice is from the famous Dome C Ice Core carefully drilled by geologists in East Antarctica a few years ago, which has already given us valuable information about pre-historic green house gas concentrations. Examining the dust could provide yet more information about the effects of temperature changes that have occurred as Earth has undergone repeated cycles of ice ages. Jean Robert Petit of the French National Scientific Research Centre explains.

Jean Robert Petit: The dust is of importance in the climate system. It may have an effect on the temperature by absorbing the incoming solar radiation or playing a role in the nutriment supply to the ocean and then they play with the carbon cycle and the atmospheric CO2. The other properties of measuring the dust size, we have an access to the strength of the atmospheric circulation and when measuring the chemical composition, it tells us about the origin of the dust. So typically the dust in Antarctica is located in the geographical sources and we know that it is in the Southern America. Also the record we have is the climatic record which span the last 800,000 years. So it's a marvellous record on the atmospheric circulation and it allows us to give a link between the different modes of climate during the past.

Michael Hopkin: And what did it appear that the dust actually does to the climate? It is not obvious how dust would actually affect things like temperature.

Jean Robert Petit: The purpose of our papers is not looking at the climatic effect, but using the dust as a tracer of the atmospheric circulations, typically it is the flux of dust which reaches the polar region will depend on the source and therefore on the environment over the continent. We know from a previous study that during the glacial period, as the temperature was about 5 degrees below the arid area and the desert were bigger and the atmosphere at this time was also dustier just because the atmospheric water was reduced and this may explain why we get more dust in polar area during glacial period.

Michael Hopkin: The EPICA study has taught us a lot already about Antarctica and its climate. How does that influence how we think about the whole world, I mean what's it got the potential to tell us?

Jean Robert Petit: Yes, we get the wealth of information from the ice core from Antarctica. Here from the dust we obtained a record comparable to other record elsewhere in the world and a comparison with the list we got in China was done over the 800,000 years and this confirmed that global character of the phenomenon of increase of dust during the glacial period.

Michael Hopkin: How long did it take, you and your colleagues, to sift through the ice core and take all the measurements you needed?

Jean Robert Petit: It takes a long time and we have to acknowledge who were the people who were on the field. Typically a part of the ice core was cut into, one-fourth of the ice core, which was continuously melted on the field or in a laboratory with 3.2 kilometres of ice which was melted, millimetre by millimetre and the water was measured for chemistry from one side and measured also for dust with laser system, so it takes about three years to get all the data and almost three more years to reduce the data and provide some interpretation which is consistent with our knowledge on the climate system.

Michael Hopkin: So does that mean that your work could help climate scientists improve their forecast of the climate in the future by taking this dust into account?

Jean Robert Petit: Yes, at least on the mode of operating of the climate, not directly the climate effect on the dust, this is a secondary effect but at least on the dynamic point of view the dust and this notion that is a new concept of climate coupling through different periods and likely through thresholds in the climate system will help for testing the GCM and to make them more close to the reality.

Kerri Smith: Jean Robert Petit there. Now next time you're staring out of the window of a train your brain will be busy calculating how far away that tree or that house is even though you're not moving towards them. Our brains use two types of information to allow us to perceive depth. One, the actual images that our eye processes and two, a cue called motion parallax: the apparent movement of an object caused by the observer changing place like when the train moves. How the brain manages the first type of depth perception cue is quite well understood, but a paper published in Nature this week is the first insight into how the second cue works. Greg DeAngelis and his team from the University of Rochester in New York used an intriguing set up involving sliding monkeys and virtual reality to find out. Here he is telling me more about the experiment. Nature 452, 642–645 (3 April 2008)

Gregory C. DeAngelis: In essence, we've shown what is evidence for a second major neuromechanism of depth perception in the brain. Most people are familiar I think with the idea that having two eyes allows us to have depth perception and we know quite a lot about how depth perception from two eyes happens in a brain, but it turns out that we actually have fairly good depth perception with using only one eye provided that we are also moving through our environment. When we move a single eye obtains different views of the scene over time and those different vantage points over time provide another very powerful depth cue called motion parallax and although we've known for a long time that humans are very good at perceiving depth from motion parallax, we haven't known anything about how that happens in the brain and this study has really provided the first evidence for a neuromechanism.

Kerri Smith: You have been looking at how the brain deals with this phenomenon of motion parallax. So what have you added to the picture of how the brain processes depth that we didn't know before?

Gregory C. DeAngelis: So what's really new here is the first evidence for neuromechanism for depth perception that can operate on images from the single eye over time as we move. It turns out that the critically important thing there is that in addition to having the images from that eye, the brain also needs to know, get some information about the actual movement of the head or the eye and what we found is that neurons in the middle temporal area of the brain combine this visual motion with a neural signal related to the movement of the subject and that by combining those two things together, these neurons are able to recover the sign of depth that is whether an object is near or far from this motion parallax cue.

Kerri Smith: So it is not only what we're seeing but while we're moving that's helping to perceive depth. I wonder if you can tell us a bit about your methods, how did you actually do this? You worked with macaques, is that right?

Gregory C. DeAngelis: Right we worked with macaque monkeys and trained them to sit in an apparatus. It is a virtual reality apparatus, kind of like a little slide simulator in which the animal subject gets moved side to side and while they are doing that they have to keep their eyes fixated on a particular point in, sort of, a virtual three-dimensional environment and by doing that we are able to have the control over moving the animal and providing these non-visual cues about the animal's movement as well as providing visual information that simulates objects at different distances. And it's really that combination of being able to do those two things together which allowed us to ask this question for the first time.

Kerri Smith: I have to say that's actually one of the cooler experimental setups we've come across, monkeys on trays in virtual reality environment. So it was neurons in this area MT that reacted more strongly when the monkeys were moving.

Gregory C. DeAngelis: It's not just that they reacted more strongly, it's that they respond to objects moving on the retina regardless, but whether an object is near or far, it can produce very similar motion on the retina and so neurons that just respond to the motion on the retina alone are not enough to solve the problem to determine whether the object is near or far. In addition they've got to combine this other source of information about the movement of the animal and so, these cells don't necessarily respond more strongly, it's that they now exhibit a form of selectivity that is selectivity for whether the object is far or near that they don't have otherwise, when the animal is not being moved.

Kerri Smith: And does this fit in quite well with our understanding of what this brain region generally does?

Gregory C. DeAngelis: Yes and no, in some sense these results were a surprise, because previous work had suggested that this particular part of the brain was largely purely visual and didn't receive much other information about the movement of the subject either from the vestibular system or from the movements of the eyes themselves. So it's a bit surprising that we do find such a strong interaction in this area between the retinal image motion and this non-visual signals. On the other hand, it actually makes a lot of sense that this part of the brain would be doing this because we also know that neurons in this part of the brain are selective for depth from binocular disparity which is the powerful binocular cue to depth and computationally the problem of computing depth from binocular disparity and motion parallax are quite similar, so it makes sense for a lot of reasons that a single area would be involved in doing both.

Kerri Smith: And finally then what can we do with this information now that we have it. Are there any implications of this work?

Gregory C. DeAngelis: This is a basic science result and so short term the most direct implications have to do with our understanding of how the brain combines different depth cues together in order to achieve more robust perception. Longer term, I think one possibility is that understanding the neural system for motion parallax may actually help us to train people who suffer from strabismus or being cross-eyed while they were children, to recover some of their binocular depth perception. So one interesting possibility is that developing training techniques, where subjects would judge depth for motion parallax and disparity cues together, might be able to sort of retrain the brain to make use of these binocular neurons again by instead training the subjects to make use of the monocular depth information and coupling the two in the right ways.

Kerri Smith: Gregory DeAngelis there. Finally this week, a new use for an established technology could make it easier to spy Earth-like planets that so far have been out of reach. Here's Geoff Brumfiel.

Geoff Brumfiel: Stars give off a unique spectrum of light. If you graph it, it looks a little like a mountain range. An orbiting planet causes the peaks of the range to periodically wobble. Those wobbles can be seen with a very precise spectrometer and this is the way astronomers spot extrasolar planets, but there is a problem, not with the spectrometers but with their calibration. To find rocky Earth-like planets requires a very precise evenly spaced set of little peaks that can be compared to a star spectrum. That hasn't been possible until now. A group has developed a sort of new laser ruler that can be stable for years at a time. I called Ron Walsworth at the Smithsonian Astrophysical Observatory at Harvard to find out how the new system might help astronomers spot far off Earths. Nature 452, 610–612 (3 April 2008)

Ronald L. Walsworth: Right now there are two techniques which are commonly used, one involves a lamp known as a thorium-argon lamp which emits a whole bright set of lines at particular wavelengths and that is used as a standard ruler, if you will. You look at it that light, into the telescope and with the spectrograph and you see where those known lines imprint a pattern on the spectrum and you say okay you can repeat that calibration periodically and compare measurements we take from a star to that. That's one technique. The second one is kind of the inverse of that which is not to imprint lines coming from an additional source, but to put something in the way of the light that's coming from the star and that imprints on that incoming light, a series of absorption lines, decreases in intensity and that is done typically with what's called an iodine cell, molecular iodine gas that can absorb light at specific wavelengths and that puts little notches, little decreases of the incoming light at some specific wavelength markers and since those lines in the two cases iodine and thorium-argon, those emission or absorption lines are well known from molecular and atomic physics in studies done in laboratories, you have a ruler or a calibrator.

Geoff Brumfiel: So, tell me a little about the frequency comb that you've developed here and how it gets around these limitations?

Ronald L. Walsworth: Okay, well what we did is, we took an existing technology that had been developed over the last 10 years known as the laser frequency comb, which is a wonderful technology that produces very regular spacing of laser line, spacing in terms of their separation and wavelength, so they are very equally spaced across the visible and to the near infrared and we tweaked and augmented and adapted laser frequency comb to make the spacing between those laser lines in spacing, in terms of wavelength appropriate for use with astrophysical spectrographs. It turns out that the pre-existing laser frequency combs which were honoured with the Nobel Prize in 2005 had the spacing of these teeth - the spacing between them in terms of wavelength much too tight for use with astrophysical spectrographs. The spectrographs that are associated with telescopes are very large objects they have to deal with the light that comes from these very large telescopes spread out over a very broad spectrum and they want their calibrator to provide individual bright lines that are these comb-like markings on a ruler that are spread apart in wavelength much larger than you would get from existing laser frequency combs. So we conceived a technique that would allow us to greatly increase that spacing and make it tuneable so that you could set the spacing to whatever you need for a particular astrophysical spectrograph.

Geoff Brumfiel: So I guess the real bottom line question is how precise is this going to allow these radio velocity measurements to get, I mean, will we be able to see Earth mass planets now?

Ronald L. Walsworth: So I think the prospects are good. In the laboratory, we've seen that this astro-comb, as we call it, has the stability and the properties needed to calibrate spectrographs of telescopes sufficient to see 1 centimetre per second not meter per second or kilometre per second now, but 1 centimetre per second changes in the velocity of stars. Now that's in the lab and that performance has not yet been realized at telescopes - that's the next step. Take the astro-comb use it at a telescope with the spectrograph there and see improved calibration and more stable calibration over long periods of time really allows us to realize sensitivity of very light planets like Earth-like planets.

Geoff Brumfiel: Do you think it's going to work? Do you think you'll see an Earth like planet with this thing there?

Ronald L. Walsworth: Well, let's assume that there are Earth like planets out there. Then I think most astronomers believe that there is a whole host of Earth and Mercury and Venus and mars type planets around the stars, so let's assume that's there, there is something to discover. What do I think? I think, yes, the astro-comb together with very good spectrographs and telescopes will allow those to be discovered. They will probably be a couple little technical bumps along the way learned over the next couple of years, learning how to mate this astro-comb with the spectrographs, but we don't foresee any really major problems.

Kerri Smith: Ron Walsworth talking to Geoff. That's all from us this week.

Adam Rutherford: Our Sound of Science is taken from a quite brilliant video from YouTube it's a spoof hip-hop tune featuring animations of all the major players who strive against the rise of creationism. The maker of the video has so far remained anonymous and online there's even some dispute about whose side they're on. A few weeks ago, we had MC Hawking. Now make some noise for MC Dawkins featuring PC Myres and Big Daddy Dennett. I'm Adam Rutherford.

Kerri Smith: And I am Kerri Smith

[Sound of Science Plays]


The Nature Podcast is sponsored by Bio-Rad, at the centre of scientific discovery for over 50 years, on the web at

End Advertisement