NASA Tests the Interplanetary Internet

A system designed to make space communications more robust passes its first test.

A spacecraft 20 million miles away from Earth recently sent the first dozens images back to our planet using a completely new kind of space communications: the Interplanetary Internet.
This system is being developed by NASA's Jet Propulsion Laboratory (JPL) as a more robust way to communicate in space, modelled on the terrerstrial Internet. The project is being led by Google's vice president and chief internet evangelist, Vint Cerf, who designed many of the networking protocols that launched the Internet in the 1970s. Last month, Cerf talked about the Interplanetary Internet with Technology Review and discussed a crucial protocol called Disruption-Tolerant Networking (DTN).
NASA will be testing the new communications link for a month or so to qualify the technology for use on a variety of upcoming space missions.

Untangling Web Information

The Semantic Web organizer Twine offers bookmarking with built-in AI.

Tied together: Twine lets users create or join "twines" devoted to a particular topic in order to collect, share, and discuss information. But Twine’s semantic AI engine also helps users discover related information and people with similar interests.  Credit: Twine

MULTIMEDIA   Watch TR’s reporter use Twine and discuss its features.

The next big stage in the evolution of the Internet, according to many experts and luminaries, will be the advent of the Semantic Web--that is, technologies that let computers process the meaning of Web pages instead of simply downloading or serving them up blindly. Microsoft's acquisition of the semantic search enginePowerset earlier this year shows faith in this vision. But thus far, little Semantic Web technology has been available to the general public. That's why many eyes will be on Twine, a Web organizer based on semantic technology that launches publicly today.

Developed by Radar Networks, based in San Francisco, Twine is part bookmarking tool, part social network, and part recommendation engine, helping users collect, manage, and share online information related to any area of interest. For the novice, it can be tricky figuring out exactly where to start. But for experienced users, Twine can be a powerful way to research a subject collaboratively or find people with common interests, with the usual features of a bookmarking site augmented by Twine's underlying semantic technology.

After creating an account, a user adds a Twine bookmarklet to her browser's bookmarks, then adds items to her Twine page by clicking the bookmarklet as she surfs the Web. Bookmarks, too, can easily be imported from a browser or from another Web bookmarking service.

Twine uses artificial intelligence--machine learning and natural language processing--to parse the contents of Web pages and extract key concepts, such as people, places, and organizations, from the pages that a user saves. The site then uses these concepts to link information and users. For example, creating a twine--a bundle of bookmarks related to a particular topic--devoted to a specialized technique in computer game design quickly led to the discovery of twines (created by other users) devoted to other areas of game design and to twines devoted to a popular game that uses the technique. It also led to other users interested in the subject. Twine is also meant to automatically generate tags, descriptions, and summaries of bookmarked Web pages. In the preview, or beta, version, this feature didn't always work properly, but Nova Spivack, CEO of Radar Networks, says that the functionality has been improved ahead of the public launch. Twines offer a hub for collecting, sharing, and discussing information. For example, users have created twines devoted to twentieth-century music, science and technology, philosophy, and cool things found around the Web.

Highly Efficient Lithium Batteries Could Greatly Extend Battery Life Of Laptop Computers

ScienceDaily (Nov. 29, 2008) — Rechargeable lithium ion batteries provide portable devices that require a lot of energy, such as mobile telephones, digital cameras, and notebook computers, with power. However, their capacity, and thus the running time of the devices, remain somewhat limited. A notebook computer thus usually runs only about two hours.

The reason for this is the relatively small capacity of the graphite anode in these batteries to absorb lithium ions. A team led by Jaephil Cho at Hanyang University in Korea has now developed a new material for anodes, which could clear a path for a new generation of rechargeable batteries. As reported in the journal Angewandte Chemie, their new material involves three-dimensional, highly porous silicon structures.

Lithium ion accumulator batteries produce current by moving lithium ions. The battery usually contains a cathode (positive electrode) made of a mixed metal oxide, such as lithium cobalt oxide, and an anode (negative electrode) made of graphite. While the battery is being charged, lithium ions migrate into the anode, where they are stored between the graphite layers. When the battery is being discharged, these ions migrate back to the cathode.

It would be nice to have an anodic material that could store more lithium ions than graphite. Silicon presents an interesting alternative. The problem: silicon expands a great deal while absorbing lithium ions (charging) and shrinks when giving them up (discharging). After several cycles the required thin silicon layers are pulverized and can no longer be charged.

Cho’s team has now developed a new method for the production of a porous silicon anode that can withstand this strain. They annealed silicon dioxide nanoparticles with silicon particles whose outermost silicon atoms have short hydrocarbon chains attached to them at 900 °C under an argon atmosphere. The silicon dioxide particles were removed from the resulting mass by etching. What remained were carbon-coated silicon crystals in a continuous, three-dimensional, highly porous structure.

Anodes made of this highly porous silicon have a high charge capacity for lithium ions. In addition, the lithium ions are rapidly transported and stored, making rapid charging and discharging possible. A high specific capacity is also attained with high current. The changes in volume that occur upon charging and discharging cause only a small degree of swelling and shrinking of the pore walls, which have a thickness of less than 70 nm.

In addition, the first charging cycle results in an amorphous (noncrystalline) silicon mass around residual nanocrystals in the pore walls. Consequently, even after 100 cycles, the stress in the pore wall is not noticeable in the material.

---

Journal reference:

1. . Three-Dimensional Porous Silicon Particles for Use in High-Performance Lithium Secondary Batteries.Angewandte Chemie International Edition, DOI:10.1002/anie.200804355

Opening the Cloud

Open-source cloud-computing tools could give companies greater flexibility.

Credit: Technology Review

Cloud-computing platforms such as Amazon's Elastic Compute Cloud (EC2), Microsoft's Azure Services Platform, and Google App Engine have given many businesses flexible access to computing resources, ushering in an era in which, among other things, startups can operate with much lower infrastructure costs. Instead of having to buy or rent hardware, users can pay for only the processing power that they actually use and are free to use more or less as their needs change.

However, relying on cloud computing comes with drawbacks, including privacy, security, and reliability concerns. So there is now growing interest in open-source cloud-computing tools, for which the source code is freely available. These tools could let companies build and customize their own computing clouds to work alongside more powerful commercial solutions.

One open-source software-infrastructure project, called Eucalyptus, imitates the experience of using EC2 but lets users run programs on their own resources and provides a detailed view of what would otherwise be the black box of cloud-computing services.

Another open-source cloud-computing project is the University of Chicago's Globus Nimbus, which is widely recognized as having pioneered the field. And a European cloud-computing initiative coordinated by IBM, called RESERVOIR, features several open-source components, including OpenNebula, a tool for managing the virtual machines within a cloud. Even some companies, such as Enomaly and 10gen, are developing open-source cloud-computing tools.

Rich Wolski, a professor in the computer-science department at the University of California, Santa Barbara, who directs the Eucalyptus project, says that his focus is on developing a platform that is easy to use, maintain, and modify. "We actually started from first principles to build something that looks like a cloud," he says. "As a result, we believe that our thing is more malleable. We can modify it, we can see inside it, we can install it and maintain it in a cloud environment in a more natural way."

Reuven Cohen, founder and chief technologist of Enomaly, explains that an open-source cloud provides useful flexibility for academics and large companies. For example, he says, a company might want to run most of its computing in a commercial cloud such as that provided by Amazon but use the same software to process sensitive data on its own machines, for added security. Alternatively, a user might want to run software on his or her own resources most of the time, but have the option to expand to a commercial service in times of high demand. In both cases, an open-source cloud-computing interface can offer that flexibility, serving as a complement to the commercial service rather than a replacement.

Memristors Make Chips Cheaper

The first hybrid memristor-transistor chip could be cheaper and more energy efficient.

Hardware upgrade: Researchers removed transistors from the bottom layer of this silicon-based chip (shown in yellow and blue) and replaced them with fewer memristors in the top layer (shown in red). Memristors can do the work of transistors but require less power and space. According to HP, they could revolutionize integrated circuits and memory technology.  Credit: Qiangfei Xia, HP

MULTIMEDIA   Watch an introduction to memristors.

Entire industries and research fields are devoted to ensuring that, every year, computers continue getting faster. But this trend could begin to slow down as the components used in electronic circuits are shrunk to the size of just a few atoms. Researchers at HP Labs in Palo Alto, CA, are betting that a new fundamental electronic component--the memristor--will keep computer power increasing at this rate for years to come.

Memristors were first predicted in 1971 by Berkeley professor Leon Chua. They are nanoscale devices with unique properties: a variable resistance and the ability to remember the resistance even when the power is off.

After rediscovering Chua's work, researchers at HP Labs built the first working memristor in May of this year. And last week, at the first ever Memristor and Memristor Systems Symposium, in Berkeley, CA, the same team showed how memristors can be integrated into functioning circuits. Their circuits require fewer transistors, allowing more components (and more computing power) to be packed into the same physical space while also using less power to function.

"We're trying to give Moore's Law a boost," says lead researcher Stan Williams, a senior research fellow at HP, referring to a prediction made by Intel founder Gordon Moore that the number of transistors on a computer circuit (and therefore computer performance) should double roughly every two years.

Increasing performance has usually meant shrinking components so that more can be packed onto a circuit. But instead, Williams's team removes some transistors and replaces them with a smaller number of memristors. "We're not trying to crowd more transistors onto a chip or into a particular circuit," Williams says. "Hybrid memristor-transistor chips really have the promise for delivering a lot more performance."

A memristor acts a lot like a resistor but with one big difference: it can change resistance depending on the amount and direction of the voltage applied and can remember its resistance even when the voltage is turned off. These unusual properties make them interesting from both a scientific and an engineering point of view. A single memristor can perform the same logic functions as multiple transistors, making them a promising way to increase computer power. Memristors could also prove to be a faster, smaller, more energy-efficient alternative to flash storage.

Although memristor research is still in its infancy, HP Labs is working on a handful of practical memristor projects. And now Williams's team has demonstrated a working memristor-transistor hybrid chip.

PUTTING PSYCHOLOGY INTO BEHAVIORAL ECONOMICS

Richard Thaler, Sendhil Mullainathan, Daniel Kahneman

RICHARD THALER: Behavioral economics and good psychology, there's a lot of art. There is science and there are well-crafted experiments, but thinking about what the right experiment to run, was art and, there are 80 gazillion experiments, which ones are relevant to getting people to plant the right seed. That's a problem that Sendhil and I have been talking about for, well, since he was born. You're now seeing the results of 15 years of conversations. And there wasn't a scientific way of answering that question.
SENDHIL MULLAINAITHAN: A lot of what makes behavioral economics interesting is psychology, it is about what happens inside the mind. These phenomena are taking things that are happening inside the mind and interfacing them with things happening in the world, the environment, and getting feedback or getting interesting responses from that.
We happen to call the word economics. But it's not economics. You could be talking about crime, you could be talking about many things, in the social domain, the entire spectrum of human behavior. Anyone who is interested in the broader world should be interested in something we currently call "behavioral economics".
DANIEL KAHNEMAN: What we're saying is that there is a technology emerging from behavioral economics. It's not only an abstract thing. You can do things with it. We are just at the beginning. I thought that the input of psychology into behavioral economics was done. But hearing Sendhil was very encouraging because there was a lot of new psychology there. That conversation is continuing and it looks to me as if that conversation is going to go forward. It's pretty intuitive, based on research, good theory, and important.
Sean Parker, Anne Treisman, Paul Romer, Danny Hillis, Jeff Bezos, Salar Kamangar, George Dyson, France LeClerc

Hubble Back In Business: Pair Of Gravitationally Interacting Galaxies In Full View

Perfect "10" due to the chance alignment of two galaxies. The left-most galaxy, or the "one" in this image, is relatively undisturbed, apart from a smooth ring of starlight. It appears nearly edge-on to our line of sight. The right-most galaxy, the "zero" of the pair, exhibits a clumpy, blue ring of intense star formation. (Credit: NASA, ESA, and M. Livio (STScI))

ScienceDaily (Nov. 1, 2008) — The Hubble Space Telescope is back in business with a snapshot of the fascinating galaxy pair Arp 147.

Just a couple of days after the orbiting observatory was brought back online, Hubble aimed its prime working camera, the Wide Field Planetary Camera 2 (WFPC2), at a particularly intriguing target, a pair of gravitationally interacting galaxies called Arp 147.

The image demonstrated that the camera is working exactly as it was before going offline, thereby scoring a "perfect 10" both for performance and beauty.

And literally "10" for appearance too, due to the chance alignment of the two galaxies. The left-most galaxy, or the "one" in this image, is relatively undisturbed, apart from a smooth ring of starlight. It appears nearly edge-on to our line of sight. The right-most galaxy, the "zero" of the pair, exhibits a clumpy, blue ring of intense star formation.

The blue ring was formed after the galaxy on the left passed through the galaxy on the right. Just as a pebble thrown into a pond creates an outwardly moving circular wave, or ripples, an outwardly propagating ring of higher density was generated at the point of impact of the two galaxies. As this excess density collided with outer material that was moving inwards due to the gravitational pull of the two galaxies, shocks and dense gas were produced, stimulating star formation.

The dusty reddish knot at the lower left of the blue ring probably marks the location of the original nucleus of the galaxy that was hit.

Arp 147 appears in the Arp Atlas of Peculiar Galaxies, compiled by Halton Arp in the 1960s and published in 1966. This picture was assembled from WFPC2 images taken with three separate filters. The colours blue, green, and red represent the blue, visible-light, and infrared filters respectively.

The galaxy pair was photographed on 27-28 October 2008. Arp 147 lies in the constellation of Cetus, more than 400 million light-years away from Earth.

---

Adapted from materials provided by ESA/Hubble Information Centre.

Computing with RNA: Devices that self-assemble from biological molecules could represent the future of drug delivery.

Credit: Technology Review

Scientists in California have created molecular computers that are able to self-assemble out of strips of RNA within living cells. Eventually, such computers could be programmed to manipulate biological functions within the cell, executing different tasks under different conditions. One application could be smart drug delivery systems, says Christina Smolke, who carried out the research with Maung Nyan Win and whose results are published in the latest issue of Science.

The use of biomolecules to perform computations was first demonstrated by the University of Southern California'sLeonard Adleman in 1994, and the approach was later developed by Ehud Shapiro of the Weizmann Institute of Science, in Rehovot, Israel. But according to Shapiro, "What this new work shows for the first time is the ability to detect the presence or absence of molecules within the cell."

That opens up the possibility of computing devices that can respond to specific conditions within the cell, he says. For example, it may be possible to develop drug delivery systems that target cancer cells from within by sensing genes used to regulate cell growth and death. "You can program it to release the drug when the conditions are just right, at the right time and in the right place," Shapiro says.

Smolke and Win's biocomputers are built from three main components--sensors, actuators, and transmitters--all of which are made up of RNA. The input sensors are made from aptamers, RNA molecules that behave a bit like antibodies, binding tightly to specific targets. Similarly, the output components, or actuators, are made of ribozymes, complex RNA molecules that have catalytic properties similar to those of enzymes. These two components are joined by yet another RNA molecule that serves as a transmitter, which is activated when a sensor molecule recognizes an input chemical and, in turn, triggers an actuator molecule.

Smolke and Win designed their RNA computers to detect the drugs tetracycline and theophylline within yeast cells, producing a fluorescent protein as an output. By combining the RNA components in certain ways, the researchers showed that they can get them to behave like different types of logic gates--circuit elements common to any computer. For example, an AND gate produces an output only when its inputs detect the presence of both drugs, while a NOR gate produces an output only when neither drug is detected.

But this is just a demonstration, Smolke says. "We're using these modular molecules that have a sort of plug-and-play capability," she says--meaning that they can be combined in a number of different ways. Different kinds of aptamers could potentially detect thousands of different metabolic or protein inputs.

Smolke and Win produce their device by encoding RNA sequences into DNA and introducing it into the cell. "So the cell is making these devices," Smolke says. "RNA is actually a very programmable substrate."

Indeed, the attractive thing about this approach is that the components of the device and the substrate holding them together are all made from RNA, says Friedrich Simmel, a bioelectronics researcher at the Technical University of Munich, in Germany. "This is something that we also would like to do," he says, for not only do such devices self-assemble, but they can also be produced on a single long strand of RNA, all at once.

Smolke and Win have already found collaborators for possible animal studies, to see how biocomputers can be delivered to cells and used once they're there. Smolke also envisions a large-scale collaboration to create a huge library of sensors out of which these devices can be made.

Mimicking Body Biosensors: Artificial antibodies could be used in cheap, field-ready toxin sensors.

Lock and key: An artificial antibody targets a toxic protein called melittin (top). The polymer “antibody” is grown around melittin, shaping it into a binding shell (middle). When the melittin is rinsed away, the polymer retains its shape (bottom) and is ready to bind with the protein again.  Credit: Kenneth Shea

Antibodies aren't just a critical part of the body's defense against disease: they're also the gold standard for biosensing. These natural proteins are widely used in diagnostic tests for cancer and other diseases because they recognize and bind so efficiently to disease markers, which include bacterial and viral proteins and protein fragments. However, protein antibodies are expensive to synthesize in the lab and only last a few months. Now biochemists have developed a method to make artificial antibodies that may be just as effective as the real thing.

Researchers have been working on synthetic antibodies for about twenty years. Artificial antibodies, made from polymers rather than proteins, promise to be cheap and long lasting, says Kenneth Shea, a professor of chemistry at the University of California, Irvine, who led development of the new method. However, no one has been able to make artificial antibodies that bind their targets as tightly and as specifically as natural antibodies do.

To make the new and improved artificial antibodies, Shea and his collaborators at the Tokyo Institute of Technology refined a technique called molecular imprinting. This involves taking a target molecule and placing it in a solution containing the building blocks of a polymer antibody. The polymer then grows around its target, conforming to its shape; once it's done, the target molecule is rinsed away. Then, when the artificial antibody next meets the target molecule, they fit together like a key in a lock. "You can make a mold around almost any molecule," explains Klaus Mosbach, founder of the Center for Molecular Imprinting at the Center for Chemistry and Chemical Engineering, in Lund, Sweden, who pioneered the technique.

Molecular imprinting is "a beautiful idea and a great technology," says Vincent Rotello, a professor of chemistry at the University of Massachusetts Amherst. Plastic antibodies have found some industrial applications, mostly for separating small molecules out of solutions. But they have not been suitable for therapeutic, diagnostic, or biosensing applications because none that have been made work as well as natural antibodies in water-based solutions such as blood. And they could not be made to effectively target large molecules such as proteins.

The researchers developed much better plastic antibodies by enhancing the imprinting process in a novel way. Shea and his collaborators started with a target protein, melittin, and screened libraries of polymer building blocks for those that would bind to it most effectively. "Some effort went into determining the composition that would minimize interaction with [nontarget] garden-variety proteins," says Shea. His artificial antibody has a high affinity for its target protein, comparable in strength to that of a natural antibody, and also works in water solutions. The synthesis is described in a paper in the Journal of the American Chemical Society. 

Shea says that better synthetic antibodies could be used therapeutically in parts of the body where normal antibodies break down quickly, such as the digestive tract. They could also be used in portable devices designed to detect traces of chemical weapons.

The new method for making the artificial antibodies is "a dramatic breakthrough," says Ken Shimizu, a chemist at the University of South Carolina. "People always say their polymers work a lot like antibodies, but in truth, they never did." He adds that Shea's work comes close to living up to this promise. "This is the first time you could imagine these [synthetics] would rival antibodies."

David Spivak, a chemistry professor at Louisiana State University, agrees that Shea's work is "the first example of an imprinted polymer acting like an antibody." Spivak expects that Shea's method will work with other target molecules, including proteins besides melittin. Indeed, Shea says that he is working on targeting about five more complex proteins.

However, some chemists are skeptical that Shea's artificial protein is targeting melittin like a real antibody. They suggest that melittin, which has positively charged regions, may simply be attracted to negative charge on the polymer. But Shea says he selected polymer building blocks based on their ability to bind regions of melittin, and that he has performed controlled experiments to demonstrate that the binding is indeed specific. Further, he says that he has unpublished work showing that his polymer binds to melittin in cell cultures and in small animals, suppressing its toxic activity.

Voyage of the Bacteria Bots: Self-propelled microbots navigate through blood vessels.

Plot and go: Researchers plot a trajectory for their bacteria-powered microrobots, which are guided using an MRI machine.  Credit: The NanoRobotics Laboratory, École Polytechnique de Montréal (EPM)

MULTIMEDIA   Click here to see MRI-controlled bacteria swimming through human blood.

The 1966 science-fiction movie Fantastic Voyagefamously imagined using a tiny ship to combat disease inside the body. With the advent of nanotechnology, researchers are inching closer to creating something almost as fantastic. A microscopic device that could swim through the bloodstream and directly target the site of disease, such as a tumor, could offer radical new treatments. To get to a tumor, however, such a device would have to be small and agile enough to navigate through a labyrinth of tiny blood vessels, some far thinner than a human hair.

Researchers at the École Polytechnique de Montréal, in Canada, led by professor of computer engineeringSylvain Martel, have coupled live, swimming bacteria to microscopic beads to develop a self-propelling device, dubbed a nanobot. While other scientists have previously attached bacteria to microscopic particles to take advantage of their natural propelling motion, Martel's team is the first to show that such hybrids can be steered through the body using magnetic resonance imaging (MRI).

To do this, Martel used bacteria that naturally contain magnetic particles. In nature, these particles help the bacteria navigate toward deeper water, away from oxygen. "Those nanoparticles form a chain a bit like a magnetic compass needle," says Martel. But by changing the surrounding magnetic field using an extended set-up coupled to an MRI machine, Martel and his colleagues were able to make the bacteria propel themselves in any direction they wanted.

The bacteria swim using tiny corkscrewlike tails, or flagella, and these particular bacteria are faster and stronger than most, says Martel. What's more, they are just two microns in diameter--small enough to fit through the smallest blood vessels in the human body. The team treated the polymer beads roughly 150 nanometers in size with antibodies so that the bacteria would attach to them. Ultimately, the researchers plan to modify the beads so that they also carry cancer-killing drugs.

"I think nature has provided an excellent solution to how to make small things swim," says Bradley Nelson, a professor at ETH Zurich, who has researched the use of artificial flagella. "What's interesting about Sylvain's work is that he's actually using nature to do it and not just learning from it."

Last year, Martel and his group published research in the journal Applied Physics Lettersdetailing how they used an MRI machine to maneuver a 1.5-millimeter magnetic bead with a bacteria propeller through the carotid artery of a living pig at 10 centimeters per second. The researchers' latest work, presented at the IEEE 2008 Biorobotics Conference last week, shows that they can track and steer microbeads and bacteria or bacteria alone through a replica of human blood vessels using the same approach. The group has carried out similar experiments in rats and rabbits, according to Martel.

The bacteria bots wouldn't be able to make it in larger blood vessels in on their own, however. The current would be too strong for them to swim against. So the researchers envisage using a larger, magnetically steerable microvehicle to carry the bots close to a tumor. "The vehicle will be a type of polymer, or possibly another type of material," says Martel. "We have a way to release the bacteria while the vehicle stays there and dissolves."

Martel's vehicle contains magnetic nanoparticles and can be moved at about 200 microns per second. He says that he and his team correct the microvehicle's course approximately 30 times a second. While they have developed the microvehicle and bacterial microbots independently, they are now working to combine the two technologies. "We think in two years we'll be able to do that," says Martel.

"This work is promising but, as with any transformative idea, there are a lot of challenges that need to be addressed," says Bahareh Behkam, an assistant professor at Virgina Polytechnic Institute, who has also used bacteria to propel microbeads. She suggests that it could be difficult to maintain normal blood flow and to retrieve the magnetic particles from the body after the procedure is complete.

Some researchers also question whether the body's immune system would attack the bacteria before they could reach a tumor, but Martel defends the approach. "We are very confident from our preliminary tests that this [scenario] will not be an issue," he says. Because the immune system has not encountered these bacteria before, he says, it would not have time to wipe out the microbots before they reach their target.

Quantum Leap: Researchers have controlled the position of a single electron in a silicon circuit.

Fence sitter: A computer model shows an arsenic atom with an electron that’s in two distinct quantum states at the same time. An international team of researchers experimentally confirmed the model’s predictions.  Credit: Insoo Woo and Rajib Rahman, Purdue University

An international team of researchers has shown that it can control the quantum state of a single electron in a silicon transistor--even putting the electron in two places at once. Their discovery could help pave the way toward a practical quantum computer.

Quantum computers take advantage of the strange properties of subatomic particles to perform certain types of calculations much faster than classical computers can. Researchers are exploring a host of different approaches to quantum computing, and some have even built primitive quantum circuits that can perform calculations. But practical quantum computing would require the ability to manufacture devices with millions of quantum circuits--rather than the 12 or 16 achievable now--that can be integrated with more-conventional electronics.

One theoretical approach to practical quantum electronics is to use conventional electronics--tiny semiconductor transistors--to control the state of a quantum system. Researchers led by Sven Rogge, a researcher at Delft University of Technology, in the Netherlands, performed the first practical experiments to verify the approach's theoretical predictions. The team--which also included researchers from Purdue University; the University of Melbourne, in Australia; and IMEC, in Belgium--found that it could control the quantum state of a single electron simply by altering the voltage applied to a transistor. "This represents a nice step towards future devices where performance is determined by manipulation of quantum states of single atoms," says Thomas Schenkel, a scientist at Lawrence Berkeley National Laboratory.

The researchers used prefabricated transistors built for nanotech research, each of which consisted of two crossed silicon nanowires. One of the nanowires--the bottom one--was connected to electrodes that contained arsenic. When that wire was charged, it would sometimes draw arsenic atoms into the transistor. After applying a voltage across about 100 transistors, the team found six that appeared to have individual arsenic atoms embedded in the nanowire. They then found that varying the voltage across the top wire would control the quantum state of one of the atom's electrons. Using an imaging technique called scanning tunneling spectroscopy, they were able to distinguish three states of the atoms in all six devices. One of those states corresponded with the electron's being in two places at the same time--a property necessary for quantum computing.

"But to understand the behavior of one atom, you have to model millions," says Rogge. So he and his colleagues used an off-the-shelf program called NEMO 3D to characterize a large-scale system containing 1.4 million atoms. They found that the measurements of their model system were in good agreement with the spectroscopy results. "It is really cool to see how well their simulations can describe randomly doped transistors, which are all different in detail," Schenkel says.

Altered states: As the electric field induced by a silicon nanowire (gray) increases, an electron in an arsenic atom moves from its ground state (left) to an excited state (right). During this transition, the electron enters a hybridized state (middle) in which it is in both of the other states simultaneously. In theory, such an electron could serve as a “qubit” in a quantum computer.  Credit: Insoo Woo and Rajib Rahman, Purdue University

Depending on the strength of the electric field created by the top nanowire, an electron could be found in one of three states. At low electric fields, the electron remained bound to the arsenic atom. At high electric fields, the electron was pulled away from the atom. But when the electric field was at just the right level, the electron would be in both places at once.

In order for a quantum computer to work, its qubits--the quantum equivalent of a classical computer's bits--need to be "entangled": their quantum states have to be coupled with each other. Pulling an electron away from its atom "might be an interesting way to couple" adjacent qubits, says Schenkel.

"While this result is an important one, the real challenge to making future single-dopant devices is in figuring out how to position the [arsenic atoms] into the silicon host with the required precision," says Bruce Kane, a research scientist at the University of Maryland. The researchers found their six devices by chance; to produce working circuits, they would need to be able to position atoms of arsenic--or some other material--in the transistors more reliably.

While the researchers eventually hope to be able to control the position of the atoms in the transistor, "our next step is to add a second electron and see what happens to the configuration of the electron state," says Gabri Lansbergen, another Delft researcher. "In the far future," Rogge adds, "we would like to experiment with several [materials] and see how they interact."

A Better Network for Outer Space: Why Vint Cerf wants to put Internet-style networking in space.

Networking space: Vint Cerf, cocreator of the Internet and a vice president at Google, is designing protocols for a robust space-communication network, modeled on the terrestrial Internet. Cerf is working with a team at NASA’s Jet Propulsion Laboratory (where he is also a visiting scientist) and at the MITRE Corporation, based in Washington, DC.  Credit: Courtesy of Google

Having designed the networking protocols that launched the Internet, Vint Cerf now wants to put the same kind of robust communications network in outer space. Currently, astronauts and robotic spacecraft communicate with Earth using point-to-point radio links and communications schemes that are tailored to nearly every new mission. This inhibits interoperability and the repurposing of communications equipment, and as the number and complexity of missions increases, it will only become more problematic.

Cerf, who is Google's vice president and chief Internet evangelist, is working with a team at NASA's Jet Propulsion Laboratory (JPL), where he is also a visiting scientist, and at the MITRE Corporation, based in Washington, DC, to design and implement a revolutionary new scheme for space communication. The project, dubbed the Interplanetary Internet, will be tested aboard the International Space Station (ISS) in 2009, and Cerf hopes that by 2010, new space missions will be designed to use the protocols.

Ultimately, the network could interconnect manned and robotic spacecraft, forming the backbone of a communications system that reaches across the solar system.

Technology Review's Brittany Sauser caught up with Cerf to discuss the details of the project.

Technology Review: What's the purpose of the Interplanetary Internet?

Vint Cerf:The project started 10 years ago as an attempt to figure out what kind of technical networking standards would be useful to support interplanetary communication. Bear in mind, we have been flying robotic equipment to the inner and outer planets, asteroids, comets, and such since the 1960s. We have been able to communicate with those robotic devices and with manned missions using point-to-point radio communications. In fact, for many of these missions, we used a dedicated communications system called the Deep Space Network (DSN), built by JPL in 1964.

But one problem with space communication has been the limited use of standards. When we launch a spacecraft with a unique set of sensors onboard, we often end up writing special communication and application software that is adapted to that spacecraft's sensor systems and manipulators. In the Internet world, we use standards called the TCP/IP protocol suite--packet switching and store-and-forward methods--to allow a lot of different devices, billions of things, to interact compatibly with each other. The team set out to develop a suite of protocols that would allow us to have the kind of network flexibility in space that we have on Earth. The Interplanetary Internet project is primarily about developing a set of communication standards and technical specifications to support rich networking in space environments.

TR:What are the challenges of building such a network in space?

VC:We started by working on a set of protocols that could deal with two very important properties of space communication. The first is delay. The distances between the planets are very large. For example, when Earth and Mars are closest together, it still takes 3.5 minutes for a radio signal moving at the speed of light to propagate. If I were on Mars and you were on Earth, it would take seven minutes at best before you heard a response. When Earth and Mars are farthest apart, the round trip takes 40 minutes! The reason we can talk back and forth on Earth so easily is that propagation times are very short by comparison.

The other problem is that the planets and their satellites are in motion, and most are rotating. The rotation of the planets means that if you are talking to something that is on the surface of the planet, it may rotate out of the line of sight so you cannot talk to it anymore, until the device on the surface rotates into view again. The same could be said for some orbiting satellites. You have to develop protocols that will deal with the fact that you cannot always communicate with the other party: the communication is both delayed and potentially disrupted. So that is what we designed: a delay- and disruption-tolerant networking system [DTN]. It will allow us to maintain communications more effectively, getting much more data because we don't have to be in direct line of sight with the ultimate recipient in order to transfer data. The new protocols will be proposed to serve as a potential international standard for space networking.

TR:How does this new protocol, the delay- and disruption-tolerant networking system, work?

VC: We are using store-and-forward methods [routing information through hosts that hold on to it until a communications link can be established] similar to the TCP/IP design in order to service space-communication requirements. But our new bundle protocol is based on DTN principles. We have to cope with the fact that there is a really high potential for delay and disruption in the system. For example, Pluto is a long ways away, on the order of three to five billion miles and about 12 hours round-trip time. Using the DTN bundle protocol allows us to design more-complex mission configurations involving many devices on the surface of planets and in orbit around them. At Mars, for example, there are four orbiters and three landed and operational spacecraft. We expect to be able to use the standard TCP/IP protocols on the surface of planets and inside spacecraft, but we will use the DTN protocols for interplanetary distance communications.

TR: Is this going to require putting new infrastructure in space?

VC: The answer is yes and no. For example, the Deep Impact spacecraft [now called EPOXI] is already in orbit around the sun. It was used to launch a probe into a comet to examine its interior. EPOXI is being temporarily repurposed to test the new DTN protocols. The spacecraft has processing, memory, radio equipment, and solar panels for power so we don't have to put new hardware up. We just have to upload new software. We are lucky to not have to field any new equipment yet, but the DTN protocols eventually have to show up in a fairly significant number of devices in the system to create the kind of network that can serve space-communication needs. Some specialized spacecraft could become store-and-forward routers. Each time a new mission is launched, using the standard bundle protocol, previous mission assets that are still in operation could be used to support the communication requirements of the new mission. In this way, we hope to accrete a kind of interplanetary backbone network.

TR: How are you handling security issues?

VC: There are security concerns, and we have been very careful to build defenses into the basic design. Each bundle-aware node will verify the identity of any other nodes that it is communicating with, and it will refuse to forward data from any nodes that it does not recognize. We will be using strong authentication methods, cryptographic communication methods, to ensure that the parties that are using the resources are authorized to do so.

TR: What is the biggest advantage of building new protocols for space communication?

VC: The important part here is that we have standardized protocols that will allow internetworking of various spacecraft launched by all the spacefaring nations. Over time, as new missions are launched, you start to build up a backbone capability. Every time you put up a new mission, you basically are putting up another potential node in the network. Our hope in the near term is to start putting DTN/bundle-protocol applications up on the Internet terrestrially, and also put them up on the International Space Station for testing. Eventually, we hope to have this capability running all the time, and then, when new deep-space missions are launched using these standard protocols, they will become part of the interplanetary communications system.

Immortalizing a Piece of Yourself

Stem cells derived from volunteers for the Personal Genome Project will be disseminated globally.

Skin biopsy: George Church, a participant in and leader of the Personal Genome Project at Harvard Medical School, undergoes a skin biopsy. Scientists will develop cell lines from Church's skin cells, which will be distributed to scientists around the world for research. 
Credit: Personal Genome Project

Scientists around the globe may soon be studying tiny bits of George Church. The Harvard Medical School professor of genetics will be one of the first people to have stem-cell lines created from his skin cells propagated and distributed worldwide--along with a record of the cells' donor's identity and genetic and medical quirks.

Church and his collaborators hope that the cells will add a new dimension to genetic studies of human disease. Moststudies compare genomes of a group of people with a particular phenotype--say, diabetes or heart disease--with healthy volunteers. But adding stem-cell lines, which can be differentiated into the myriad tissue types affected by disease, should allow scientists to search for molecular changes that are the intermediary between genes and the manifestation of a disease. "For nearly every genetic trait variation, there is a change at the cellular level," says Church.

This new addition to genomics studies comes thanks to induced pluripotent cell reprogramming, a recent breakthrough in stem-cell research that allows scientists to create immortal cell lines from healthy human donors that, like embryonic stem cells, can both replicate themselves and differentiate into many types of tissue. While cultured cells are a ubiquitous part of biomedical research, used to test drugs, study disease, and engineer tissue, most cells cannot form different tissue types, and they come from anonymous donors--meaning that medical and other characteristics of the donor are unknown. In addition, many "immortal" cell lines, which can divide indefinitely, are derived from tumor cells and thus have abnormal chromosomes.

Church won't be alone in distributing his cells. The scientist aims to create hundreds or thousands of cell lines over the next few years as part of the Personal Genome Project, an effort that he launched two years ago to capitalize on advances in gene-sequencing technologies. So far, the project has enrolled 10 volunteers--and garnered headlines, mainly for its genomic-era exhibitionism. Volunteers, including Harvard psychologist Steven Pinker and entrepreneur Ester Dyson, released their medical records and preliminary genetic analyses on the Web earlier this month. But media attention has mostly ignored that fact that they've also given something that may be even more personal. Each has undergone a skin biopsy, which will be used to generate stem-cell lines.

The lines will allow scientists to study cells with a known genetic and clinical profile. Those working with cells derived from Steven Pinker, for example, would know that he suffers from basal cell carcinoma, lichen planus (an inflammatory skin condition), and Reynaud's syndrome (a hypersensitivity of the fingers and toes to extreme temperatures). 

SELECTIVE MEMORY

Altering a protein in the brain of mice can wipe out specific memories as they are recalled

As much as you might want to wipe Uncle Frank’s tasteless joke out of your mind but still remember the flavor of Aunt Fran’s pie, memory researchers have always said “fuhgedabboudit!” Now, a genetically engineered mouse suggests it may be possible to erase certain unwanted memories.

Scientists from the Medical College of Georgia in Augusta and the East ChinaNormal University in Shanghai selectively removed a shocking memory from a mouse’s brain, the team reports in the Oct. 23 Neuron.

Insight from such experiments may one day lead to therapies that can erase traumatic memories for people suffering from post-traumatic stress disorder, or wipe clean drug-associated cues that lead addicts to relapse.

“We should never think of memories as being fixed,” says Howard Eichenbaum, a neuroscientist at Boston University. “They are constantly being renovated and restructured.”

Careful questioning can alter an eyewitness’s recollection during testimony, Eichenbaum says. The new research, which he calls “terrific” and “interesting,” shows that careful use of molecular tools can also manipulate memories.

Joe Tsien, a neuroscientist at the Medical College of Georgia, and his colleagues genetically engineered a mouse to carry an altered version of a protein called alpha-calcium/calmodulin-dependent protein kinase II, or alpha-CaMKII.

A kinase enzyme, alpha-CaMKII is a type of regulatory protein that governs the activity of other proteins. Previous research showed that alpha-CaMKII is involved in learning and memory. Tsien and his colleagues wanted to find out at which stage of memory the kinase enzyme is important. Stages of memory include learning something new and then processing, retrieving and storing the information.

Scientists are beginning to learn more about how memories are made and stored. Memories are likely formed through interactions of brain chemicals and changing connections between neurons. But exactly how that happens and the physical form memory takes remain a mystery.

Researchers can use chemicals to block an enzyme’s activity, but the business end of most kinase enzymes look alike, so most inhibitory chemicals tend to block all kinase activity in the brain. Tsien got around that problem by building a hidden cavity in alpha-CaMKII. A bulky chemical inhibitor fits into the hidden cavity and blocks alpha-CaMKII from doing its job, but doesn’t interfere with the action of other kinases. By manipulating activity of the engineered protein, the researchers learned that alpha-CaMKII is important for recalling memories.

A mouse might not be able to recall a memory for two reasons, Tsien says. “Either you can’t open the door to get the memory, or you can open the door but there’s no memory there.”

Altering alpha-CaMKII’s activity erases memories as they are being retrieved, the researchers found. And the erasure is specific to the memory being recalled.

The researchers placed mice in a chamber and played a sound, then mildly shocked the mice’s feet. The mice learned to associate both the chamber and the sound with a shock and would freeze in anticipation of getting shocked when they entered the chamber or heard the sound.

Once the mouse learned to associate both the chamber and sound with getting shocked, the researchers replayed one of the conditions while altering activity of alpha-CaMKII. If the researchers placed the mouse in the chamber but didn’t play the sound, only the memory of the chamber was erased when alpha-CaMKII’s activity was altered. When tested again later, the mouse forgot to freeze when placed in the chamber, but the mouse would still freeze when it heard the sound. And if conditions were reversed and alpha-CaMKII activity was altered when the mouse was recalling that the sound signals a shock, the sound memory was erased. But the mice still remembered to freeze when entering the chamber. Those results show that erasure is limited only to the portion of the memory being recalled.

Eichenbaum is not convinced that Tsien and his colleagues have erased the mice’s memories. Altering a memory so that it can’t be recalled under certain circumstances might produce similar results, he says. “We never know for sure that it’s really gone,” he says.

But if chemicals can help someone specifically forget painful or traumatic memories, it may be irrelevant whether the memories are entirely erased or are just altered beyond recognition, Eichenbaum says.

Memory-erasing pills are still science fiction, Tsien stresses. This technique will never be used in people as it involves genetically engineering a protein in the brain, he says. But future studies might reveal other ways to selectively forget.

“We’ve only just put our foot on a very tall mountain,” he says.

BIAS, QUAKES AND VIRUSES, OH MY!

Researchers present findings at the annual New Horizons in Science meeting

UCSF’s Joseph DeRisi discussing an expensive machine, called a DNA sequencer, which decodes the genetic information in a large number of samples each day. It costs about $3,000 to use the machine each time. But, DeRisi says that the technology is getting cheaper and faster rapidly. He predicts that within five years scientists will be able to decode a person’s genome within hours.

PALO ALTO, Calif. — Clothing divas have fashion week, film enthusiasts have Sundance and for science journalists there’s the New Horizons in Science — a four-day meeting of presentations and field trips that provide an in-depth look at cutting-edge research in fields from genomics to psychology. Sponsored by the Council for the Advancement of Science Writing, the annual conference is being held in Palo Alto, Calif., this year. Here’s a sampling of research discussed on Day One, October 26:

Powerful, if subconscious, biases

Tall men are more likely to run corporations than shorter men, people with good teeth make more money — these are just some manifestations of the biases that shape decision making in our daily lives. While rational thinking is supposed to be a hallmark of being human, from a young age people develop implicit attitudes that cloud their thinking, says psychologist and neuroscientist Mahzarin Banaji of Harvard University. “To what extent do they leak into our behavior, unbeknownst to us?” Banaji asks. She has recently been investigating this question using conjoint analysis, a technique that asks a volunteer to choose between two scenarios with several variables, such as living in San Francisco while making $35,000 per year with a strict boss named Jennifer versus living in Detroit while making $40,000 per year with a relaxed boss named Jason. Changing the variables in the questionnaire and running several iterations revealed that the majority of participants “chose” to make $3,400 less, on average, to work under a male boss. Yet when the subjects were asked directly if their supervisor’s gender was important the answer was a resounding “no,” Banaji reports.

Putting emphasis on variables that, consciously at least, we do not think are important, can have serious implications on decision making, says Banaji, citing studies showing how race influences people’s perceptions of how “American” someone is, including current presidential candidates. Can we fight these subconscious biases? “Yes, they are elastic,” says Banaji. But are they plastic? Can we consciously remold our thinking to eliminate these biases? She’s not sure. “That work needs to be done,” she says. Try taking a test athttps://implicit.Harvard.edu. —Rachel Ehrenberg

Virus tracking on a chip

Many respiratory infections and diarrheal illnesses are caused by an unknown virus, making them difficult to diagnose and treat.

Joseph DeRisi, a Howard Hughes Medical Institute investigator at the University of California, San Francisco, has developed an identity chip for viruses. The chip contains genetic information for a wide range of known viruses. The scientists can use the chip to screen viral DNA and some RNA in mucus, blood or other samples collected from sick patients — both human and animal. Matching an unknown viral genetic sequence to that of a well-studied virus helps narrow down the identity of the unknown virus. Then the researchers can decode the genetic blueprints of the new virus to learn more about how it causes disease.

DeRisi and his colleagues have already used the chips to track down the cause of a disease that kills parrots. The disease, called proventricular dilation disease or PDD, paralyzes the part of the bird’s digestive system that moves food into the stomach. Food gets stuck and the bird dies of starvation.

A bornavirus related to viruses that cause sad horse disease (also known as Borna disease) is responsible for PDD in parrots, the researchers discovered. Several different bornaviruses infect birds. No one knows where the viruses come from and how they spread to parrots and horses. DeRisi plans to test samples from ducks and other migratory birds to learn whether they may be carrying the viruses.

These viral testing chips are still experimental, but DeRisi hopes to produce them for use in clinical laboratories where they could aid in diagnosing illness and help doctors design better treatment plans for their patients. —Tina Hesman Saey

Quake Ready?

Geologist Tom Brocher has a message for Californians: “Are you ready for this earthquake? Because it’s ready for you.” And he isn’t talking about the San Andreas fault. The Hayward Fault, which also runs through the San Francisco Bay Area, experienced a major quake in 1868, 140 years ago. And it’s due for another big one soon, Brocher said.

The fault has, since 1315, experienced five quakes with magnitudes greater than 6.3. The average number of years between each quake was 138. “It’s not like clockwork, but it’s very constant,” said Brocher, a seismologist with the U.S. Geological Survey and chair of the group 1868 Hayward Earthquake Alliance.

The fault starts at San Pablo Bay and, running east of San Francisco Bay, stretches south to Freemont. “The Hayward fault is probably the most urbanized fault in North America,” Brocher said, adding that rupture of the fault would break both the Bay Bridge and the BART system. About 2.5 million people live within a stone’s throw of the fault, added Brocher, who is the USGS regional coordinator for Northern California Earthquake Hazards Investigations.

Brocher reported USGS estimates of a 63 percent chance that a magnitude-6.7 or higher quake will hit the Bay Area in the next 30 years, and a 67 percent chance for the Los Angeles area.

Even with work already done to retrofit buildings and infrastructure, Brocher shared estimates that $150 billion in damage could result from just the ground shaking of a big Hayward quake, not including damage from fires or landslides. Reports suggest that ground shaking and damage in 1868 was strongest in and around the cities of San Leandro and Hayward. Also, areas built over landfill, such as the San Francisco financial district, are at risk because landfill material essentially acts like a sand-filled liquid when stirred by a quake, Brocher said. —Kristina Bartlett Brody

–

Voices in the car

If Herbie the car could talk, he probably would be less lovable. Robotic voices — like those found on car navigation systems, for example — can trigger strong responses among human users. (Irritation comes to mind.) But Clifford Nass of Stanford University thinks he could engineer these voices to play off people’s dispositions and encourage safer driving.

In a recent study, Nass found that happy drivers get in fewer accidents when a chipper voice talks to them in a vehicle than when a morose voice makes conversation. Unhappy or upset people do better with less-cheery voices.

“Voices in cars manifest more than just content; they manifest emotion,” Nass says. “Upset drivers could benefit from a subdued voice. Happy drivers were screwed up by that subdued voice.”

Software that could detect drowsiness, distraction, emotion and personality of a car’s operator could adjust its voice accordingly, Nass said. It could deliver a male voice for a male driver, or even adjust to imitate its driver’s voice exactly, making it more likeable.

Nass’s studies also reveal that humans may be more willing to cooperate with robots if the machines display paralinguistic skills that more closely mirror human skills. When people disagree with others, those people tend to avoid eye contact and distance themselves from any disapproving statements they deliver. Nass found people are more likely to take advice from a robot when the robot acts in the same way.

“The fact that it didn’t stand right up, and say ‘I disagree with you,’ made people feel better about the robot,” Nass said. The work extends beyond a car criticizing its owner’s driving to any context in which humans interact with robots.

In other affective technology research, Nass has found that male avatars perform better than female avatars on math tests in a virtual world regardless of whether the person actually taking the test is a man or woman, suggesting people play into stereotypes. He also mentioned an upcoming study that reveals how the brain multitasks, which psychologists have long called an impossibility. Such multitasking, he says, has changed the way people think and perform in basic psychological tests. —Elizabeth Quill