Amazing Science

The global pattern of e-mail communication reflects the cultural fault lines thought to determine future conflict, say computational social scientists.

 

Most political scientists consider the Cold War as a conflict between capitalist countries in the west and the Communist Bloc in the East. As such, it was essentially a conflict of ideology. At the end of the Cold War, the question arose of what would drive the next wave of conflicts.

 

In 1992, the Harvard-based political scientist Samuel Hartington suggested that future conflicts would be driven largely by cultural differences. He went on to map out a new world order in which the people of the world are divided into nine culturally distinct civilisations. These include: Western civilisation; Latin American civilisation; the Orthodox world of former Soviet Union countries; the Sinic civilisation including China, the Koreas and Vietnam; the Muslim world of the greater Middle East; Sub-Saharan Africa and so on. His argument was that future conflicts would be based around the fault lines at the edges of these civilisations. He published this view in a now famous article called “The Clash of Civilizations?” in the politcal journal Foreign Affairs.

Today, we get an answer of sorts thanks to the work of Bogdan State at Stanford University in California and a few pals. These guys have analysed a global database of e-mail messages, and their locations, sent by  more than 10 million people over the space of a year. State and co say that the pattern of connections between these people, clearly reflects the civilisations mapped out by Huntington. In other words, the way we send e-mails is a reflection of the mesh of civilisations that is an important driver of future conflict.

“The findings (unsurprisingly) support the idea that geography, transporation and administrative decisions are all important determinant of between-country communication: distance decreases density, as do visas, while direct flights increase it,” say the researchers. And there are surprising results as well. For example, a common border between two countries actually reduces the communication density between them, perhaps because of increased tensions.  “These curious findings do raise the issue of potential problems with European integration, as well as of the higher potential for conflict between countries sharing borders, which may lead to less communication,” say State and pals.

 

There are one or two caveats with this kind of research, the main one being that the process of rescaling the data can introduce artefacts that then influence the observed effects. However, further research  with other data sets should help to iron out these problems. Of course, if our planet is divided by civilisation in the way Huntington suggested, it’s not surprising that this is reflected in the pattern of global communication.

A more interesting question is whether this kind of computational social science can measure the ongoing pulse of global tensions and whether it has any predictive power over in spotting where the next conflicts are likely to arise. That’s beyond the current state of the art but it’s clearly an area worth watching.

The internet connects people all over the world. But could the internet also connect us with dolphins, apes, elephants and other highly intelligent species? In a bold talk in Session 10 of TED2013, four incredible thinkers come together to launch the idea of the interspecies internet. Each takes four minutes to talk, then passes the metaphorical baton, building the narrative in parts.

 

The talk begins with Diana Reiss, a cognitive psychologist who studies intelligence in animals. She shows us a video of an adorable dolphin twirling in the water. But the dolphin isn’t spinning playfully for the camera — the dolphin is watching itself in a two-way mirror.

 

“A dolphin has self-awareness,” says Reiss. “We used to think this was a uniquely human quality, but dolphins aren’t the only non-human animals to show self-recognition in a mirror. Great apes, our closest relatives, also show this ability.” Ditto for elephants and even magpies.

 

Reiss shares her work with dolphins — she’s been teaching them to communicate through an underwater keyboard of symbols that correspond to whistles and playful activities. Through this keyboard, the dolphins learned to perform activities on demand, and also to express their desire for them.

AIST researchers have developed a graphene transistor with a new operating principle. In the developed transistor, two electrodes and two top gates are placed on graphene and graphene between the top gates is irradiated with a helium ion beam to introduce crystalline defects. Gate biases are applied to the two top gates independently, allowing carrier densities in the top-gated graphene regions to be effectively controlled. An electric current on/off ratio of approximately four orders of magnitude was demonstrated at 200 K (approximately −73 °C). In addition, its transistor polarity can be electrically controlled and inverted, which to date has not been possible for transistors. This technology can be used in the conventional production technology of integrated circuits based on silicon, and is expected to contribute to the realization of ultra-low-power-consumption electronics by reducing operation voltage in future.

In the developed transistor, the on state or off state is controlled according to whether the polarities of the voltages applied to the two top gates are the same or different. Therefore, by fixing one gate bias and changing its polarity, it is possible to control whether the transistor operation by sweeping the other gate voltage is n-type or p-type. In the present experiment, voltages of −100 mV and +100 mV were applied to the source and drain terminals, respectively. The relation between the source-drain current and the bias of the source-side gate when the gate voltage of the drain-side, VtgD, is fixed to be positive (Fig. 4(a)), is shown in Fig. 4(b). A logarithmic plot of the same data is shown in Fig. 4 (c). Here, when the gate voltage of the source-side is negative, the transistor is off, and when it is positive, the transistor is on. So it operates as an n-type transistor. Meanwhile, the relation between the source-drain current and the bias of the source-side gate when the gate voltage of the drain-side is negative (Fig. 4(d)), is shown in Figs 4(e) and 4(f). In this case, when the gate voltage of the source-side is negative, the transistor is on, and when positive, the transistor is off. So it operates as a p-type transistor. In other words, it was actually demonstrated that the polarity of a single transistor can be inverted by electrostatic control.

The transistor polarity of conventional silicon transistors is determined by the type of ion for doping, so it is not possible to change the polarity once a circuit is formed. However, because the polarity of the developed transistor can be electrostatically controlled, it is possible to realize an integrated circuit whose circuit structure can be electrically changed.

The researchers are aiming to realize CMOS operation in which transistor polarities can be changed through electrical control. They are also aiming to create a device prototype using a large-scale wafer with graphene synthesized by the CVD method (chemical vapor-phase deposition method). At the same time, efforts to achieve higher-quality graphene will be made in order to improve the on/off ratio of electric current at room temperature and carrier mobility.

Startup Symform says its shredded, distributed cloud is more resistant to natural disasters than traditional computing clouds.

 

The world has embraced the cloud. What’s not to like? Startups can grow rapidly without investing in racks of computers, companies can back up data easily, consumers can travel light and still have access to their huge photo libraries and other personal files.

 

Back in October, however, real clouds clashed with metaphorical clouds when Hurricane Sandy and its aftermath took down some key data centers in New York and New Jersey; a serious problem for businesses who had their main servers in New York and their backup servers in nearby New Jersey.

 

Commercial cloud service providers, for the most part, did pretty well; perhaps because some of the largest data centers, like Amazon’s northern Virginia server farm, were not in the disaster zone. But Sandy certainly reminded cloud service providers that redundant files have to be separated by more than a couple of racks, or even a couple of miles.

 

Startup Symform thinks it can provide better disaster resilience than even data centers hundreds of miles apart. And, says Bassam Tabbara, Symform cofounder and Chief Technical Officer, it can do that in a way that’s extremely cheap—and in some cases free—to its customers.

 

Tabbara describes Symform’s approach as a “decentralized, distributed, virtual, and crowd-sourced” cloud. Living in the San Francisco Bay area, I can visualize that kind of cloud, however, we don’t call it a cloud here, we call it fog.

At the HiPEAC 2013 conference in Berlin, KALRAY demonstrated MPPA256, the world’s first supercomputer-on-a-chip, consisting of 256 computing cores. This innovative processor combines the ultimate in several processor types and will enable a whole new class of embedded and industrial applications in the fields of image processing, signal processing, control, communications and data security. The MPPA256 is completely designed in Europe and is only one of the success stories resulting from the 170 million Euro investment in carefully selected European funded research projects over the past 6 years.

“The MPPA256 combines the benefits of high processing power (500 billion operations per second), low power consumption and high-level programming, and it is the first highly scalable multi-core processor that can also execute real-time applications and is therefore perfectly suited for mixed criticalities systems” stated Benoit Dupont de Dinechin Director of Software Development at KALRAY and one of the main architects of the chip.

KALRAY announced its new product at HiPEAC 2013, a conference that brought together more than 500 delegates from all over the world for three days packed with more than 40 events. "The HiPEAC conference is currently the premier computing systems event in Europe” said Professor Ben Juurlink from the Technical University of Berlin, general chairman of the conference. HiPEAC, the European Network of Excellence on High Performance and Embedded Architecture and Compilation, is geared towards advancing the level of research in this field and stimulating the collaboration between academia and industry, and between computer architects and tool builders. The next HiPEAC conference will take place in Vienna, January 20 to 22, 2014.

Thanks to the strange laws of quantum mechanics, quantum computers would be able to carry out certain computational tasks much faster than conventional computers.  Among the most promising technologies for the construction of a quantum computer are systems of single atoms, confined in so-called ion traps and manipulated with lasers.  In the laboratory, these systems have already been used to test key building blocks of a future quantum computer.  “Currently, we can carry out successful quantum computations with atoms,” explain Andreas Stute and Bernardo Casabone, both PhD students at the University of Innsbruck’s Institute for Experimental Physics. “But we are still missing viable interfaces with which quantum information can be transferred over optical channels from one computer to another.”

What makes the construction of these interfaces especially challenging is that the laws of quantum mechanics don’t allow quantum information to be simply copied. Instead, a future quantum internet – that is, a network of quantum computers linked by optical channels – would have to transfer quantum information onto individual particles of light, known as photons.  These photons would then be transported over an optical-fiber link to a distant computing site.  Now, for the first time, quantum information has been directly transferred from an atom in an ion trap onto a single photon.  The work is reported in the current issue of Nature Photonics by a research team led by Tracy Northup and Rainer Blatt.

The University of Innsbruck physicists first trap a single calcium ion in an ion trap and position it between two highly reflective mirrors.  “We use a laser to write the desired quantum information onto the electronic states of the atom,” explains Stute. “The atom is then excited with a second laser, and as a result, it emits a photon.  At this moment, we write the atom’s quantum information onto the polarization state of the photon, thus mapping it onto the light particle.”  The photon is stored between the mirrors until it eventually flies out through one mirror, which is less reflective than the other.  “The two mirrors steer the photon in a specific direction, effectively guiding it into an optical fiber,” says Casabone.  The quantum information stored in the photon could thus be conveyed over the optical fiber to a distant quantum computer, where the same technique could be applied in reverse to write it back onto an atom.

At IBM’s Watson Research Center in upstate New York, some of the world’s best physicists, chemists, and nanoengineers are trying to create the first high-density, self-assembling carbon nanotube computer chip process. In much the same way that Jack Kilby at Texas Instruments discovered the monolithic VLSI process for making silicon chips in 1958, IBM desperately wants to find the process that enables the creation of carbon nanotube chips. In the next decade — or thereabouts; the goalposts keep shifting — silicon is expected to reach a miniaturization roadblock.

 

At some point, we simply won’t be able to make silicon transistors any smaller. When this happens, there will be a few materials jostling to fill the void, most notably silicon-germanium, galium arsenide, and various forms of carbon (nanotubes, nanowires, graphene). In theory, computer chips made from carbon nanotubes are massively desirable — they would be many times faster than silicon, use less power, and can scale down to just a couple of nanometers. In practice, working with carbon nanotubes — just like graphene — is proving to be rather difficult. It’s sometimes easy to forget that we have decades of experience and billions of R&D dollars plowed into silicon; expertise with new materials won’t come easy.

Stanford Engineering's Center for Turbulence Research (CTR) has set a new record in computational science by successfully using a supercomputer with more than one million computing cores to solve a complex fluid dynamics problem—the prediction of noise generated by a supersonic jet engine.

Joseph Nichols, a research associate in the center, worked on the newly installed Sequoia IBM Bluegene/Q system at Lawrence Livermore National Laboratories (LLNL) funded by the Advanced Simulation and Computing (ASC) Program of the National Nuclear Security Administration (NNSA). Sequoia once topped list of the world's most powerful supercomputers, boasting 1,572,864 compute cores (processors) and 1.6 petabytes of memory connected by a high-speed five-dimensional torus interconnect.

Because of Sequoia's impressive numbers of cores, Nichols was able to show for the first time that million-core fluid dynamics simulations are possible—and also to contribute to research aimed at designing quieter aircraft engines. The exhausts of high-performance aircraft at takeoff and landing are among the most powerful human-made sources of noise. For ground crews, even for those wearing the most advanced hearing protection available, this creates an acoustically hazardous environment. To the communities surrounding airports, such noise is a major annoyance and a drag on property values. Understandably, engineers are keen to design new and better aircraft engines that are quieter than their predecessors. New nozzle shapes, for instance, can reduce jet noise at its source, resulting in quieter aircraft. Predictive simulations—advanced computer models—aid in such designs. These complex simulations allow scientists to peer inside and measure processes occurring within the harsh exhaust environment that is otherwise inaccessible to experimental equipment. The data gleaned from these simulations are driving computation-based scientific discovery as researchers uncover the physics of noise.

The new data-storage molecules are known as graphene fragments, because they largely consist of flat sheets of carbon (which are attached to zinc atoms).

 

An experimental technology called molecular memory could store data in individual molecules has beendeveloped by an international team of researchers led byJagadeesh Moodera, a senior research scientist in the MIT Department of Physics and at MIT’s Francis Bitter Magnet Laboratory.

 

The technology promises a 1,000-fold increase in storage density over hard disks, which are approaching a million megabytes of data per square inch.

Previous schemes for molecular memory have relied on physical systems cooled to near absolute zero. The new molecular-memory scheme works at around the freezing point of water — which in physics parlance counts as “room temperature.”

 

Moreover, where previous schemes required sandwiching the storage molecules between two ferromagnetic electrodes, the new scheme would require only one ferromagnetic electrode. That could greatly simplify manufacture, as could the shape of the storage molecules themselves: because they consist of flat sheets of carbon atoms attached to zinc atoms, they can be deposited in very thin layers with very precise arrangements.

 

DARPA’s new Cyber Targeted-Attack Analyzer program looks at how information is connected and moves to uncover and prevent targeted attacks without a human having to direct it.

 

The Department of Defense (DoD) maintains one of the largest computer networks in the world. The network follows DoD personnel across the globe collecting, transferring and processing information in forms as diverse as data warehouses, in-the-field mobile devices and mission computers on board F-18’s. This network is also constantly changing in size and shape as new missions are undertaken and new technology is deployed. In military terms, that means the cyber terrain of the DoD network is constantly shifting.

 

Traditional approaches to protecting networks involve static cyber firewalls around the network perimeter and patching any discovered holes. DARPA researchers seek a new approach, one that relies on knowing the cyber terrain within the network and understanding how information across the enterprise is connected to find actions associated with an attack buried under or within all the normal data.

 

DARPA’s new Cyber Targeted-Attack Analyzer program will attempt to automatically correlate all of a network’s disparate data sources—even those that are as large and complex as those within the DoD — to understand how information is connected as the network grows, shifts and changes. Once all of the data sources are correlated, the program will attempt to integrate them on a network to allow the defenders to understand the connections—like injecting a contrasting smoke into the air to see how it flows. The third phase of the program also seeks to build tools that use this information for cyber defense of the network.

 

“The Cyber Targeted-Attack Analyzer program relies on a new approach to security, seeking to quickly understand the interconnections of the systems within a network without a human having to direct it,” said Richard Guidorizzi, DARPA program manager. “Cyber defenders should then be capable of more quickly discovering attacks hidden in normal activities.”

Yale University scientists have found a way to observe quantum information while preserving its integrity, an achievement that offers researchers greater control in the volatile realm of quantum mechanics and greatly improves the prospects of quantum computing.

Quantum computers would be exponentially faster than the most powerful computers of today. "Our experiment is a dress rehearsal for a type of process essential for quantum computing," said Michel Devoret, the Frederick William Beinecke Professor of Applied Physics & Physics at Yale and principal investigator of research published Jan. 11 in the journal Science. "What this experiment really allows is an active understanding of quantum mechanics. It's one thing to stare at a theoretical formula and it's another thing to be able to control a real quantum object. "In quantum systems, microscopic units called qubits represent information. Qubits can assume either of two states—"0" or "1"—or both simultaneously. Correctly recognizing, interpreting, and tracking their state is necessary for quantum computing. However, the act of monitoring them usually damages their information content. The Yale physicists successfully devised a new, non-destructive measurement system for observing, tracking and documenting all changes in a qubit's state, thus preserving the qubit's informational value. In principle, the scientists said, this should allow them to monitor the qubit's state in order to correct for random errors.

The innovation by Yale University physicists offers scientists greater control in the volatile realm of quantum mechanics and greatly improves the prospects of quantum computing. Quantum computers would be exponentially faster than the most powerful computers of today.

IBM's Watson—the same machine that beat Ken Jennings at Jeopardy—is now churning through case histories at Memorial Sloan-Kettering, learning to make diagnoses and treatment recommendations. This is one in a series of developments suggesting that technology may be about to disrupt health care in the same way it has disrupted so many other industries. “n Brazil and India, machines are already starting to do primary care, because there’s no labor to do it. They may be better than doctors. Mathematically, they will follow evidence—and they’re much more likely to be right.

 

Harley Lukov didn’t need a miracle. He just needed the right diagnosis. Lukov, a 62-year-old from central New Jersey, had stopped smoking 10 years earlier—fulfilling a promise he’d made to his daughter, after she gave birth to his first grandchild. But decades of cigarettes had taken their toll. Lukov had adenocarcinoma, a common cancer of the lung, and it had spread to his liver. The oncologist ordered a biopsy, testing a surgically removed sample of the tumor to search for particular “driver” mutations. A driver mutation is a specific genetic defect that causes cells to reproduce uncontrollably, interfering with bodily functions and devouring organs. Think of an on/off switch stuck in the “on” direction. With lung cancer, doctors typically test for mutations called EGFR and ALK, in part because those two respond well to specially targeted treatments. But the tests are a long shot: although EGFR and ALK are the two driver mutations doctors typically see with lung cancer, even they are relatively uncommon. When Lukov’s cancer tested negative for both, the oncologist prepared to start a standard chemotherapy regimen—even though it meant the side effects would be worse and the prospects of success slimmer than might be expected using a targeted agent.

 

But Lukov’s true medical condition wasn’t quite so grim. The tumor did have a driver—a third mutation few oncologists test for in this type of case. It’s called KRAS. Researchers have known about KRAS for a long time, but only recently have they realized that it can be the driver mutation in metastatic lung cancer—and that, in those cases, it responds to the same drugs that turn it off in other tumors. A doctor familiar with both Lukov’s specific medical history and the very latest research might know to make the connection—to add one more biomarker test, for KRAS, and then to find a clinical trial testing the efficacy of KRAS treatments on lung cancer. But the national treatment guidelines for lung cancer don’t recommend such action, and few physicians, however conscientious, would think to do these things.

 

Did Lukov ultimately get the right treatment? Did his oncologist make the connection between KRAS and his condition, and order the test? He might have, if Lukov were a real patient and the oncologist were a real doctor. They’re not. They are fictional composites developed by researchers at the Memorial Sloan-Kettering Cancer Center in New York, in order to help train—and demonstrate the skills of—IBM’s Watson supercomputer. Yes, this is the same Watson that famously went on Jeopardy and beat two previous human champions. But IBM didn’t build Watson to win game shows. The company is developing Watson to help professionals with complex decision making, like the kind that occurs in oncologists’ offices—and to point out clinical nuances that health professionals might miss on their own.

 

Watson has gotten some media hype already, including articles in Wired and Fast Company. Still, you probably shouldn’t expect to see it the next time you visit your doctor’s office. Before the computer can make real-life clinical recommendations, it must learn to understand and analyze medical information, just as it once learned to ask the right questions on Jeopardy. That’s where Memorial Sloan-Kettering comes in. The famed cancer institute has signed up to be Watson’s tutor, feeding it clinical information extracted from real cases and then teaching it how to make sense of the data. “The process of pulling out two key facts from aJeopardy clue is totally different from pulling out all the relevant information, and its relationships, from a medical case,” says Ari Caroline, Sloan-Kettering’s director of quantitative analysis and strategic initiatives. “Sometimes there is conflicting information. People phrase things different ways.” But Caroline, who approached IBM about the research collaboration, nonetheless predicts that Watson will prove “very valuable”—particularly in a field like cancer treatment, in which the explosion of knowledge is already overwhelming. “If you’re looking down the road, there are going to be many more clinical options, many more subtleties around biomarkers … There will be nuances not just in interpreting the case but also in treating the case,” Caroline says. “You’re going to need a tool like Watson because the complexity and scale of information will be such that a typical decision tool couldn’t possibly handle it all.”

 

The Cleveland Clinic is also helping to develop Watson, first as a tool for training young physicians and then, possibly, as a tool at the bedside itself. James Young, the executive dean of the Cleveland Clinic medical school, told The Plain Dealer, “If we can get Watson to give us information in the health-care arena like we’ve seen with more-general sorts of knowledge information, I think it’s going to be an extraordinary tool for clinicians and a huge advancement.” And WellPoint, the insurance company, has already begun testing Watson as a support tool for nurses who make treatment-approval decisions.

 

Whether these experiments show real, quantifiable improvements in the quality or efficiency of care remains to be seen. If Watson tells physicians only what they already know, or if they end up ordering many more tests for no good reason, Watson could turn out to be more hindrance than help. But plenty of serious people in the fields of medicine, engineering, and business think Watson will work (IBM says that it could be widely available within a few years). And many of these same people believe that this is only the beginning—that whether or not Watson itself succeeds, it is emblematic of a quantum shift in health care that’s just now getting under way.

 

IBM has created an optical chip that is the first parallel optical transceiver that is able to transfer one trillion bits (or one terabit) of information per second. To put that in perspective, IBM states that 500 high-def movies could be downloaded in one second at that speed, while the entire U.S. Library of Congress web archive could be downloaded in an hour. Stated another way, the Optochip is eight times faster than any other parallel optical components currently available, with a speed that's equivalent to the bandwidth consumed by 100,000 users, if they were using regular 10 Mb/s high-speed internet.

 

One of the unique features of parallel optic chips is the fact that they can simultaneously send and receive data. The Holey Optochip capitalizes on that feature, for its record-setting performance.

 

The "Holey" in the name comes from the fact that the team started with a standard silicon CMOS chip, but bored 48 holes into it. These allow optical access to its inside back surface, where 24 separate receiver and transmitter channels are located - for a total of 48 channels. Each of those channels has its own dedicated VCSEL (vertical cavity surface emitting laser) and photodetector, which are used respectively for sending and receiving data. The chip is designed to be coupled to a multimode fiber array, via a microlens optical system.

This project will develop and test a next generation digital preservation framework including tools for analysing, ingesting, managing, accessing and reusing information objects and data.

The SHAMAN Integrated Project aims at developing a new framework for long-term digital preservation (more than one century) by exploring the potential of recent developments in the areas of GRID computing, federated digital library architectures, multivalent emulation and semantic representation and annotation.

 

The researchers' vision is: "For the longer term, SHAMAN will develop radically new approaches to Digital Preservation, such as those inspired by human capacity to deal with information and knowledge, providing a sound basis and instruments for unleashing the potential of advanced ICT to automatically act on high volumes and dynamic and volatile digital content, guaranteeing its preservation, keeping track of its evolving semantics and usage context and safeguarding its integrity, authenticity and long term accessibility over time."

 

The project plans to deliver a set of integrated tools supporting the various aspects of the preservation process: analysis/characterisation, ingestion, management, access and reuse. Work includes trials and validation of the tools in three application domains dealing with different types of objects: scientific publishing and government archives, industrial design and engineering (e.g. CAD), and e-science resources.

 

SHAMAN's dissemination and exploitation plans aim at actively fostering outreach and take-up of results and will be tailored according to the specific needs of the scientific / academic world and of industry users. SHAMAN's work will be coordinated with other digital preservation projects and initiatives at national and international level.

How’s this for big data: A whole-slide image of a tumor section can be ten billion pixels. There can be thousands of such images in the tumor cohorts maintained by The Cancer Genome Atlas project, which are collected from a large pool of patients.

 

The images are a potential treasure trove for the emerging field of precision medicine. Hidden in those billions of pixels is a story of how tumor cells organize themselves, the molecular networks that influence these structural traits, and what it all means for patients. Unfortunately, culling this information from numerous images is difficult. That’s because no two tumors are alike, and there are myriad technical variations in how samples are prepared.

 

This analysis could soon get much easier. Berkeley Lab scientists have developed an algorithm and a computational pipeline that combs through large sets of images and identifies tumor subtypes. It also identifies heterogeneity, or the extent to which a tumor comprises different organizational structures. The pipeline then uses clinical data to rank cellular signatures that are predictive of patient outcome. It also uses large-scale genomic data to identify molecular correlates of each subtype.

 

The resulting information will help scientists learn more about the genetic and molecular mechanisms that control tumor signatures. It will also shed light on whether tumor subtype can predict the effectiveness of therapies.

 

“Our goals are to identify morphometric and architectural traits that can be predictive of a therapy. We’d also like to learn about the molecular signatures that lead to architectural aberrations,” says Bahram Parvin of Berkeley Lab’s Life Sciences Division. The development of the core computational module and the pipeline were led by Hang Chang and Gerald Fontenay, respectively, in Parvin’s Lab in the Life Sciences Division.

 

The core computational module works by extracting each cell from an image, and then profiling properties of each cell such as size, shape, and organization. In this way, the telltale characteristics of a specific tumor subtype are gleaned from a large cohort of images.

 

As recently reported, the scientists validated their pipeline by applying it to 377 whole-slide images from 146 patients who have an aggressive brain cancer called Glioblastoma Multiforme. The pipeline identified several tumor subtypes based on a range of cellular profiles. It also determined whether each subtype is predictive of a patient’s response to alternative therapy. Although the pipeline was developed in a high-performance computer language, it is compute intensive and required extensive use of the Lawrencium cluster operated by Berkeley Lab’s IT Division.

Salk researchers share a how-to secret for biologists: code for Amazon Cloud that significantly reduces the time necessary to process data-intensive microscopic images

.

The method promises to speed research into the underlying causes of disease by making single-molecule microscopy of practical use for more laboratories.

 

"This is an extremely cost-effective way for labs to process super-resolution images," says Hu Cang, Salk assistant professor in the Waitt Advanced Biophotonics Center and coauthor of the paper. "Depending on the size of the data set, it can save over a week's worth of time."

 

The latest frontier in basic biomedical research is to better understand the "molecular machines" called proteins and enzymes. Determining how they interact is key to discovering cures for diseases. Simply put, finding new therapies is akin to troubleshooting a broken mechanical assembly line-if you know all the steps in the manufacturing process, it's much easier to identify the step where something went wrong. In the case of human cells, some of the parts of the assembly line can be as small as single molecules.

According to the Abbe limit, it is impossible to see the difference between any two objects if they are smaller than half the wavelength of the imaging light. Since the shortest wavelength we can see is around 400 nanometers (nm), that means anything 200 nm or below appears as a blurry spot. The challenge for biologists is that the molecules they want to see are often only a few tens of nanometers in size.

 

"You have no idea how many single molecules are distributed within that blurry spot, so essential features and ideas remain obscure to you," says Jennifer Lippincott-Schwartz, a Salk non-resident fellow and coauthor on the paper.

 

In the early 2000s, several techniques were developed to break through the Abbe Limit, launching the new field of super-resolution microscopy. Among them was a method developed by Lippincott-Schwartz and her colleagues called Photoactivated Localization Microscopy, or PALM.

 

PALM, and its sister techniques, work because mathematics can see what the eye cannot: within the blurry spot, there are concentrations of photons that form bright peaks, which represent single molecules. The downside to these approaches is that it can take several hours to several days to crunch all the numbers required just to produce one usable image.

The 256 bit unified processor and memory system will turn world computer architecture and the Internet upside down in the year 2022.

For the first time in human history, one logical computer will be able to directly address every bit and byte of memory and every device on the Internet. In effect, it will gain direct access to all stored human knowledge.

One major benefit is the lack of duplication required - every computer operating system, every software version, every video, every piece of music, every web site, and bit of information about the world and it's inhabitants is there. No need to distribute videos or music anymore - carefully positioned local cache's across the world will provide enough resilience and high speed access to all information anywhere.

Machine to machine communication will be direct. The 128 bit addresses of IPv6, which allow 4.8×10E28 individual addresses on the Internet, would be a very small subset of the 256 bit address space. Not only would individual devices be online as now, but it will be possible to map device memory and address that directly - in effect writing content directly to a TV screen, an MP3 player, your watch, and maybe even your brain.

With direct access also comes the ability to promote knowledge - medical history, geo data, dietary consumption, personal addictions, communications, and much more can be cross referenced, correlated, and eventually understood. The addressing scheme easily allows storage and access to every human beings genome for example, in both raw and encoded formats.

Having every bit of information in the world allows direct access to whole new universes of knowledge, along with equally massive security concerns. However, we have 10 years to design an architecture that will provide such massive potential pooling of knowledge and removal of duplication that we can afford to spend the resource to ensure it is much more secure than the existing Internet, with it's clouds of "bots" and counter-counter cyber terrorists.

In terms of communications and entertainment, mapping together the video and audio inputs and outputs would allow for example mass video conferences and broadcasts. Millions of people can't speak at the same time, but all can watch an election candidate for example and interact in real time.

The next-generation memory-maker Micron Technology was one of the many innovative companies demonstrating its wares on the Supercomputing Conference (SC12) show floor last November. Micron's General Manager of Hybrid Technology Scott Graham was on hand to discuss the latest developments in their Hybrid Memory Cube (HMC) technology, a multi-chip module (MCM) that aims to address one of the biggest challenges in high performance computing: scaling the memory wall.

 

Memory architectures haven't kept pace with the bandwidth requirements of multicore processors. As microprocessor speeds out-accelerated DRAM memory speeds, a bottleneck developed that is referred to as the memory wall. Stacked memory applications, however, enable higher memory bandwidth.

 

The Hybrid Memory Cube (HMC) is a new memory architecture that combines a high-speed logic layer with a stack of through-silicon-via (TSV) bonded memory die that enables impressive advantages over current technology. According to company figures, a single HMC offers a 15x performance increase and uses 70 percent less energy per bit when compared to DDR3 memory, and takes up 90 percent less space than today's RDIMMs. The Cube is also scalable per application, which is not possible with DDR3 and DDR4. System designers have the option of employing the HMC as near memory for best performance or in a scalable module form factor, as far memory, for optimum power efficiency.