Archives: News

news custom post type.

Unlocking the power of exascale supercomputers

Leading research organizations and computer manufacturers in the U.S. are collaborating on construction of some of the world’s fastest supercomputers. These exascale systems can perform more than a billion billion (a quintillion or 1018) operations per second — about the number of neurons in ten million human brains.

Exascale is about 1,000 times faster and more powerful than the fastest supercomputers today, which solve problems at the lower petascale (more than one quadrillion or 1015 operations per second). The new exascale machines will better enable scientists and engineers to answer difficult questions about the universe, advanced healthcare, national security and more, according to the U.S. Department of Energy’s (DOE) Exascale Computing Project (ECP).

Supercomputer uses in deep learning

Meanwhile, the applications and software that will run on supercomputers are also being developed by ECP developers, which recently published a paper (open-access) highlighting their progress in using supercomputers in deep learning.

“The environment will really allow individual researchers to scale up their use of DOE supercomputers on deep learning in a way that’s never been done before,” said Rick Stevens, Argonne associate laboratory director for Computing, Environment and Life Sciences.

DOE’s Argonne National Laboratory, future home to the Aurora exascale system, is a key partner in the ECP. Its researchers are involved in developing applications and co-designing the software needed to enable applications to run efficiently.

Simulating “virtual universes” with Exasky

One exciting application is simulation of “virtual universes” on demand and at high fidelities to investigate how the universe evolved from its early beginnings. Example: an ECP project known as ExaSky, using cosmological simulation codes.

Researchers are also adding capabilities within their codes that didn’t exist before. “We’re able to include atomic physics, gas dynamics and astrophysical effects in our simulations, making them significantly more realistic,” said Salman Habib, director of Argonne’s Computational Science division. ​

Online data analysis and reduction

Researchers are also co-designing the software needed to efficiently manage the data they create. Today, HPC applications already output huge amounts of data, far too much to efficiently store and analyze in its raw form. So data needs to be reduced or compressed.

One efficient solution to this is to analyze data at the same time simulations are running, a process known as online data analysis or in situ analysis.

An ECP center known as the Co-Design Center for Online Data Analysis and Reduction (CODAR) is developing both online data analysis methods, as well as data reduction and compression techniques for exascale applications. CODAR works closely with a variety of application teams to develop data compression methods, which store the same information but use less space, and reduction methods, which remove data that is not relevant.

Among the solutions the CODAR team has developed is Cheetah, a system that enables researchers to compare their co-design approaches. Another is Z-checker, a system that lets users evaluate the quality of a compression method from multiple perspectives.

Deep learning and precision medicine for cancer treatment

Exascale computing also has important applications in healthcare, and the DOE, National Cancer Institute (NCI) and the National Institutes of Health (NIH) are taking advantage of it to understand cancer and the key drivers impacting outcomes. The Exascale Deep Learning Enabled Precision Medicine for Cancer project is developing a framework called CANDLE (CANcer Distributed Learning Environment) to address key research challenges in cancer and other critical healthcare areas.

CANDLE uses neural networks to find patterns in large datasets. CANDLE is being developed for three pilot projects geared toward understanding key protein interactions, predicting drug response and automating the extraction of patient information to inform treatment strategies.

Scaling up deep neural networks

Unlocking these problems is at different scale — molecular, patient and population levels — but all are supported by the same scalable deep learning environment in CANDLE. The CANDLE software suite includes a collection of deep neural networks that capture and represent the three problems, a library of code adapted for exascale-level computing and a component that orchestrates how work will be distributed across the computing system.

“The environment will allow individual researchers to scale up their use of DOE supercomputers on deep learning in a way that’s never been done before,” said Rick Stevens, Argonne associate laboratory director for Computing, Environment and Life Sciences.

Applications such as these are just the tipping point. Once these systems come online, the potential for new capabilities will be endless.

Citations (open-access): “Exascale applications: skin in the game,” in Philosophical Transactions of the Royal Society A. and Wozniak, Justin M., et al. 2018 and “CANDLE/Supervisor: A Workflow Framework for Machine Learning Applied to Cancer Research.” BMC Bioinformatics 19 (18): 491. https://doi.org/10.1186/s12859-018-2508-4.

Organizations: The U.S. Department of Energy’s Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. Laboratory partners involved in ExaSky include Argonne, Los Alamos and Lawrence Berkeley National Laboratories. Collaborators working on CANDLE include Argonne, Lawrence Livermore, Los Alamos and Oak Ridge National Laboratories, NCI and the NIH.

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Astronomers solve the 60-year mystery of quasars, the most powerful objects in the Universe

Scientists have unlocked one of the biggest mysteries of quasars — the brightest, most powerful objects in the Universe — discovering that they are ignited by galaxies colliding.

Quasars can shine as brightly as a trillion stars packed into a volume the size of our Solar System, but until now it has remained a mystery what could trigger such powerful activity,

Using deep imaging observations from the Isaac Newton Telescope in La Palma, Canary Islands, scientists at the universities of Sheffield and Hertfordshire observed 48 galaxies that host quasars and compared them to more than 100 non-quasar galaxies.

The future of our own Milky Way galaxy

When two galaxies collide, gravitational forces push huge amounts of gas towards supermassive black holes. The ignition of a quasar can have dramatic consequences for entire galaxies. It can drive the rest of the gas out of the galaxy, which prevents it from forming new stars for billions of years into the future. the centre of the remnant galaxy system that results from the collision. Just before the gas is consumed by the black hole, it releases extraordinary amounts of in the form of radiation, resulting in a quasar.

“Quasars are one of the most extreme phenomena in the Universe, and what we see is likely to represent the future of our own Milky Way galaxy when it collides with the Andromeda galaxy in about five billion years,” said Professor Clive Tadhunter, from the University of Sheffield’s Department of Physics and Astronomy.

Beacons to the history (and future) of the universe

“Quasars are important to astrophysicists because, due to their brightness, they stand out at large distances and therefore act as beacons to the earliest epochs in the history of the Universe,” said Dr. Jonny Pierce, Post-Doctoral Research Fellow at the University of Hertfordshire.

“It’s an area that scientists around the world are keen to learn more about. One of the main scientific motivations for NASA’s James Webb Space Telescope was to study the earliest galaxies in the Universe, and Webb is capable of detecting light from even the most distant quasars, emitted nearly 13 billion years ago. Quasars play a key role in our understanding of the history of the Universe, and possibly also the future of the Milky Way.”

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

China tests 1,000 km/h ultra-high-speed maglev passenger train

A Chinese research team has successfully completed the country’s first full-size superconducting test run for a planned ultra-high-speed magnetic levitation (maglev) passenger train, according to the China Aerospace Science and Industry Corporation (CASIC).

The maglev train is expected to be used for commuting between mega-city clusters in the future. A superconducting maglev test line has been built in Datong City, north China’s Shanxi Province.

World’s fastest land travel

Once completed, it will be capable of running at 1,000 kilometers per hour (621 mph) under ideal conditions in a low-vacuum tube.

Speaking at a science popularization exhibition held in Beijing Sunday, Li Ping, a member of the project at CASIC, said that more experiments will be conducted to reach the maximum speed and to verify the entire system’s reliability, according to the South China Morning Post.

The fastest train running on a regular schedule is currently the Beijing to Shanghai line, at 350 km/h (217 mph), according to the High Speed Rail Alliance. It’s also the only operating maglev train in the world at this time. California is building a 220-mph High-Speed Rail.

 

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Nanowire networks learn and remember like a human brain

Neural network-like circuitry and synapse-like signaling suggest intelligence could be physical

An international team led by scientists at the University of Sydney has demonstrated nanowire networks that can exhibit both short- and long-term memory like the human brain.

“In this research we found higher-order cognitive function, which we normally associated with the human brain, can be emulated in non-biological hardware,” said physicist Dr. Alon Loeffler. “This work builds on our previous research, in which we showed how nanotechnology could be used to build a brain-inspired electrical device with neural network-like circuitry and synapse-like signaling.

“Our current work paves the way towards replicating brain-like learning and memory in non-biological hardware systems and suggests that the underlying nature of brain-like intelligence may be physical.”

Nanowire networks

Nanowire network pathways changing and strengthening over time (credit: University of Sydney)

Nanowire networks are a type of nanotechnology typically made from tiny, highly conductive silver wires that are invisible to the naked eye, covered in a plastic material. The wires are scattered across each other like a mesh and mimic aspects of the networked physical structure of a human brain.

Advances in nanowire networks could herald many real-world applications, such as improving robotics or sensor devices that need to make quick decisions in unpredictable environments.

“This nanowire network is like a synthetic neural network because the nanowires act like neurons, and the places where they connect with each other are analogous to synapses,” said senior author Professor Zdenka Kuncic, from the School of Physics.

Testing nanowire network intelligence

To test the capabilities of the nanowire network, the researchers gave it a test similar to a common memory task used in human psychology experiments: the N-Back task.

For a person, the N-Back task might involve remembering a specific picture of a cat from a series of feline images presented in a sequence. An N-Back score of 7, the average for people, indicates the person can recognize the same image that appeared seven steps back. When applied to the nanowire network, the researchers found it could “remember” a desired endpoint in an electric circuit seven steps back, meaning a score of 7 in an N-Back test.

“What we did here is manipulate the voltages of the end electrodes to force the pathways to change, rather than letting the network just do its own thing. We forced the pathways to go where we wanted them to go,” Loeffler said.

“When we implement that, its memory had much higher accuracy and didn’t really decrease over time, suggesting that we’ve found a way to strengthen the pathways to push them towards where we want them, and then the network remembers it. Neuroscientists think this is how the brain works: certain synaptic connections strengthen while others weaken, and that’s thought to be how we preferentially remember some things, how we learn and so on.”

The researchers said that when the nanowire network is constantly reinforced, it reaches a point where that reinforcement is no longer needed because the information is consolidated into memory.

“It’s kind of like the difference between long-term memory and short-term memory in our brains,” Kuncic said. “If we want to remember something for a long period of time, we really need to keep training our brains to consolidate that; otherwise it just kind of fades away over time.

Citation: Loeffler, A., Diaz-Alvarez, A., Zhu, R., Ganesh, N., Shine, J. M., Nakayama, T., & Kuncic, Z. (2023). Neuromorphic learning, working memory, and metaplasticity in nanowire networks. Science Advances. DOI: 10.5281/zenodo.7633957. https://www.science.org/doi/10.1126/sciadv.adg3289 (open-access)

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Wearable ultrasound skin patch lets drugs bypass stomach

MIT researchers have developed a wearable patch that applies painless ultrasonic waves to the skin, creating tiny channels that drugs can pass through. This design could also be adapted to deliver hormones and muscle relaxants, and for wound healing, pain relief, or other medical and cosmetic applications. 

The skin is an appealing route for drug delivery because it allows drugs to go directly to the site where they’re needed — not via the complex stomach route. The conformable ultrasound patch (cUSP) consists of four piezoelectric elements embedded in a poly(dimethylsiloxane (PDMS) substrate, a silicone-based polymer that can adhere to the skin without tape:

Wearable patch elements (credit: MIT)

“The main benefit with skin is that you bypass the whole gastrointestinal tract,” says Aastha Shah, co-lead author of a paper in Advanced Materials. “With oral delivery, you have to deliver a much larger dose in order to account for the loss that you would have in the gastric system. This is a much more targeted, focused modality of drug delivery.”

Reaching the bloodstream directly

With the current version of the device, drugs can penetrate a few millimeters into the skin, making this approach potentially useful for drugs that act locally within the skin. Those could include niacinamide or vitamin C, which is used to treat age spots or other dark spots on the skin, or topical drugs used to heal burns.

With further modifications to increase the penetration depth, this technique could also be used for drugs that need to reach the bloodstream, such as caffeine, fentanyl, or lidocaine, and for delivering hormones such as progesterone. The researchers are now exploring the possibility of implanting similar devices inside the body to deliver drugs to treat cancer or other diseases.

No more sticks

The researchers are also working on further optimizing the wearable patch, in hopes of testing it soon on human volunteers. “After we characterize the drug penetration profiles for much larger drugs, we would then see which candidates, like hormones or insulin, can be delivered using this technology, to provide a painless alternative for those who are currently bound to self-administer injections on a daily basis,” Shah says.

Citation: Chia-Chen Yu et al. A Conformable Ultrasound Patch for Cavitation-Enhanced Transdermal Cosmeceutical Delivery. 19 March 2023 (open-access) https://doi.org/10.1002/adma.202300066

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Brain Images Just Got 64 Million Times Sharper

Duke’s Center for In Vivo Microscopy researchers* have now improved the resolution of MRI, leading to the highest-resolution images ever captured of an entire mouse brain.

The scans are dramatically crisper than a typical clinical MRI for humans — the scientific equivalent of going from a pixelated 8-bit graphic to the hyper-realistic detail of a painting. A single voxel (think of it as a cubic pixel) measures just 5 microns. That’s 64 million times smaller than a clinical MRI voxel.

Visualizing the connectivity of the entire brain

The refined MRI provides an important new way to visualize the connectivity of the entire brain at record-breaking resolution. The researchers say new insights from mouse imaging will in turn lead to a better understanding of conditions in humans, such as how the brain changes with age, diet, or even with neurodegenerative diseases like Alzheimer’s.

The team’s new work, appearing April 17 in the Proceedings of the National Academy of Sciences, is the culmination of nearly 40 years of research at the Duke Center for In Vivo Microscopy.

Revolutionary MRI resolution

Over the four decades, Johnson, his engineering graduate students and his many collaborators at Duke and afar refined many elements that, when all combined, made the revolutionary MRI resolution possible.

Some of the key ingredients include a more powerful magnet (most clinical MRIs rely on a 1.5 to 3 Tesla magnet; Johnson’s team uses a 9.4 Tesla magnet), a special set of gradient coils that are 100 times stronger than those in a clinical MRI, and a high-performance computer equivalent to nearly 800 laptops.

Light sheet microscopy

After Johnson and his team scan the brain, they send off the tissue to be imaged using a different technique called light sheet microscopy. This complementary technique gives them the ability to label specific groups of cells across the brain, such as dopamine-issuing cells to watch the progression of Parkinson’s disease.

The team then maps the light sheet pictures, which give a highly accurate look at brain cells, onto the original MRI scan, which is much more anatomically accurate and provides a vivid view of cells and circuits throughout the entire brain.

One set of MRI images shows how brain-wide connectivity changes as mice age, as well as how specific regions, like the memory-involved subiculum, change more than the rest of the mouse’s brain. Another set showcases a spool of rainbow-colored brain connections that highlight the remarkable deterioration of neural networks in a mouse model of Alzheimer’s disease.

Better understanding of human diseases

The hope is that by making the MRI an even higher-powered microscope, Johnson and others can better understand mouse models of human diseases, such as Huntington’s disease, Alzheimer’s, and others. And that should lead to a better understanding of how similar things function or go awry in people.

“Research supported by the National Institute of Aging uncovered that modest dietary and drug interventions can lead to animals living 25% longer,” Johnson said. “So, the question is, is their brain still intact during this extended lifespan? Could they still do crossword puzzles? Are they going to be able to do Sudoku even though they’re living 25% longer? And we have the capacity now to look at it. And as we do so, we can translate that directly into the human condition.”

Citation: “Merged Magnetic Resonance And Light Sheet Microscopy Of The Whole Mouse Brain,” G. Allan Johnson et al. Proceedings of the National Academy of Sciences, April 17, 2023. DOI: https://www.pnas.org/doi/10.1073/pnas.2218617120

*With colleagues at the University of Tennessee Health Science Center, University of Pennsylvania, University of Pittsburgh and Indiana University

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Ageing studies in five animals suggest how to reverse decline

Ageing seems to affect cellular processes in the same way across five very different kinds of life — humans, fruit flies, rats, mice and worms — according to a study published in Nature on 12 April. The findings could help to explain what drives ageing and offer suggestions for how to reverse it.

“It opens up a really fundamental new area of understanding how and why we age,” says Lindsay Wu, a biochemist at UNSW Sydney in Australia.

As animals age, a variety of molecular processes inside cells become less reliable — gene mutations become more frequent, and the ends of chromosomes snap off, making them shorter. Many studies have explored ageing’s effects on gene expression, but few have investigated how it affects transcription — the process whereby genetic information is copied from a blueprint DNA strand to RNA molecules — says Andreas Beyer, a computational biologist at the University of Cologne in Germany.

Careless copying

To find out, Beyer and his colleagues analysed genome-wide transcription changes in five organisms: nematode worms, fruit flies, mice, rats and humans, at different adult ages. The researchers measured how ageing changed the speed at which the enzyme that drives transcription, RNA polymerase II (Pol II), moved along the DNA strand as it made the RNA copy. They found that, on average, Pol II became faster with age, but less precise and more error-prone across all five groups. “We saw more mismatches between the reads and reference genome,” says Beyer.

Previous research had shown that restricting diet and inhibiting insulin signalling can delay ageing and extend lifespan in many animals, so the researchers then investigated whether these measures had any effect on the speed of Pol II. In worms, mice and fruit flies that carried mutations in insulin signalling genes, Pol II moved at a slower pace. The enzyme also travelled more slowly in mice on a low-calorie diet.

But the ultimate question was whether changes in Pol II speed affected lifespan. Beyer and his team tracked the survival of fruit flies and worms that carried a mutation that slowed Pol II down. These animals lived 10% to 20% longer than their non-mutant counterparts. When the researchers used gene editing to reverse the mutations in worms, the animals’ lifespans shortened. “That really established a causal connection,” says Beyer.

Picking up the pace

The researchers wondered whether Pol II’s acceleration could be explained by structural changes in how DNA is packed inside cells. To minimize the space that they take up, the vast threads of genetic information are tightly wound around proteins called histones into bundles called nucleosomes. By analysing human lung cells and umbilical vein cells, the researchers found that ageing cells contained fewer nucleosomes, smoothing the path for Pol II to travel faster. When the team boosted the expression of histones in the cells, Pol II moved at a slower pace. In fruit flies, the elevated histone levels seemed to increase their lifespans.

The study is a “really exciting piece of work” that demonstrates how ageing mechanisms are consistent across distantly related species, says Colin Selman, who studies ageing in mammals at the University of Glasgow, UK. It also opens the door to exploring how Pol II could be a target for drugs that slow down the ageing process. Changes to Pol II’s transcription process have been implicated in many diseases, including various types of cancer, and a range of drugs have been developed that target Pol II and the molecules that facilitate it. “There may be opportunities to effectively repurpose some of these drugs to investigate their effects on ageing,” says Selman.

Citation: Debès, C. et al.Nature, Ageing-associated changes in transcriptional elongation influence longevity. (2023) https://doi.org/10.1038/s41586-023-05922-y (open-access)

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Can humans endure long-term living in deep space? 

Probably not, suggests a new “Pancosmorio” (world limit) theory of the sustainability of human migration and settlement in space, proposed by scientists at the Carl Sagan Institute at Cornell University and Norfolk Institute.

The first key is gravity, which Earth life needs to function properly, a group that aims to solve problems of human resilience on Earth and in space.

“Humans and all Earth life have evolved within the context of 1G of gravity,” the scientists note in a paper and in “Humans need Earth-like ecosystem for deep-space living,” an article in Cornell Chronicle, April 12, 2023. “Our bodies, our natural ecosystems, all the energy movement and the way we utilize energy are fundamentally based upon 1G of gravity being present, notes Lee Irons, executive director of the Norfolk Institute,

“Gravity induces a gradient in the fluid pressure within the body of the living thing to which the autonomic functions of the life form are attuned. An example of gravity imbalance would be the negative affect on the eyesight of humans in Earth orbit, where they don’t experience the weight necessary to induce the pressure gradient.

“Oxygen is another key factor. Earth’s ecosystem generates oxygen for humans and other life forms. If a technologically advanced primary and a back-up system failed to provide oxygen for the Moon base, for example, it would mean instant doom for the astronauts.”

Citation: Irons, L. G., & Irons, M. A. (2023). Pancosmorio (world limit) theory of the sustainability of human migration and settlement in space. Frontiers in Astronomy and Space Sciences, 10. https://doi.org/10.3389/fspas.2023.1081340

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Is there life on Jupiter’s moons?

On April 13, 2023, the European Space Agency is scheduled to launch a rocket carrying a spacecraft called Jupiter Icy Moons Explorer (JUICE), destined for Jupiter; followed by NASA’s robotic spacecraft Europa Clipper, launching in October 2024.

But the real interest for both: three of Jupiter’s moons — Europa, Ganymede and Callisto. All three almost certainly possess underground oceans of liquid water,” says Purdue University Professor of Planetary Science Dr. Mike Sori in The Conversation.

Top to bottom: Europa, Ganymede and Callisto (credit: NASA, CC BY-ND)

Europa’s ocean alone has more than double the water of all of Earth’s oceans combined

These three moons “have chilly surfaces that are hundreds of degrees below zero,” says Sori. “At these temperatures, ice behaves like solid rock. But just like Earth, the deeper underground you go on these moons, the hotter it gets. Go down far enough and you eventually reach the temperature where ice, warmed by a volcanic sea floor, melts into water.”

How do we know? “When these moons travel through Jupiter’s magnetic field, they generate a secondary, smaller magnetic field that signals to researchers the presence of an underground ocean. Using this technique, planetary scientists have been able to show that the three moons contain underground oceans.”

What about extraterrestrial life?

Liquid water is an important piece of what makes for a habitable world, but far from the only requirement for life. “Life also needs energy and certain chemical compounds in addition to water to flourish. On Europa, for example, the liquid water ocean overlays a rocky interior. That rocky seafloor could provide energy and chemicals through underwater volcanoes that could make Europa’s ocean habitable.

Magnetometers will also be on both missions. These tools will give scientists the opportunity to study the secondary magnetic fields produced by the interaction of conductive oceans with Jupiter’s field in great detail and will hopefully give researchers clues to salinity and volumes of the oceans.

“Scientists will also observe small variations in the moons’ gravitational pulls by tracking subtle movements in both spacecrafts’ orbits. That could help determine if Europa’s seafloor has volcanoes that provide the needed energy and chemistry for the ocean to support life.”

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter. 

Generative modeling tool renders 2D sketches in 3D

Researchers at Carnegie Mellon University’s Robotics Institute have developed a machine learning tool that could potentially allow beginner and professional designers to create 3D virtual models of everything from customized household furniture to video game content.

Pix2pix3d allows anyone to create a realistic, 3D representation of a simple or rough 2D sketch, using generative artificial intelligence tools similar to those powering popular AI photo generation and editing applications.

Pix2pix3d has been trained on data sets including cars, cats and human faces, and the team is working to expand those capabilities. In the future, it could be used to design consumer products, like giving people the power to customize furniture for their homes. Both novice and professional designers could use it to customize items in virtual reality environments or video games, or to add effects to films.

Once pix2pix3d generates a 3D image, the user can modify it in real time by erasing and redrawing the original two-dimensional sketch.

Citation: Deng, K., Yang, G., Ramanan, D., & Zhu, J. (2023). 3D-aware Conditional Image Synthesis. https://arxiv.org/abs/2302.08509

 

Let us know your thoughts! Sign up for a Mindplex account now, join our Telegram, or follow us on Twitter.