Culture

Most new achievements in artificial intelligence (AI) require very large neural networks. They consist of hundreds of millions of neurons arranged in several hundred layers, i.e. they have very "deep" network structures. These large, deep neural networks consume a lot of energy in the computer. Those neural networks that are used in image classification (e.g. face and object recognition) are particularly energy-intensive, since they have to send very many numerical values from one neuron layer to the next with great accuracy in each time cycle.

Computer scientist Wolfgang Maass, together with his PhD student Christoph Stöckl, has now found a design method for artificial neural networks that paves the way for energy-efficient high-performance AI hardware (e.g. chips for driver assistance systems, smartphones and other mobile devices). The two researchers from the Institute of Theoretical Computer Science at Graz University of Technology (TU Graz) have optimized artificial neuronal networks in computer simulations for image classification in such a way that the neurons - similar to neurons in the brain - only need to send out signals relatively rarely and those that they do are very simple. The proven classification accuracy of images with this design is nevertheless very close to the current state of the art of current image classification tools.

Information processing in the human brain as a paradigm

Maass and Stöckl were inspired by the way the human brain works. It processes several trillion computing operations per second, but only requires about 20 watts. This low energy consumption is made possible by inter-neuronal communication by means of very simple electrical impulses, so-called spikes. The information is thereby encoded not only by the number of spikes, but also by their time-varying patterns. "You can think of it like Morse code. The pauses between the signals also transmit information," Maass explains.

Conversion method for trained artificial neural networks

That spike-based hardware can reduce the energy consumption of neural network applications is not new. However, so far this could not be realized for the very deep and large neural networks that are needed for really good image classification.

In the design method of Maass and Stöckl, the transmission of information now depends not only on how many spikes a neuron sends out, but also on when the neuron sends out these spikes. The time or the temporal intervals between the spikes practically encode themselves and can therefore transmit a great deal of additional information. "We show that with just a few spikes - an average of two in our simulations - as much information can be conveyed between processors as in more energy-intensive hardware," Maass said.

With their results, the two computer scientists from TU Graz provide a new approach for hardware that combines few spikes and thus low energy consumption with state-of-the-art performances of AI applications. The findings could dramatically accelerate the development of energy-efficient AI applications and are described in the journal Nature Machine Intelligence.

The genetics of human eye colour is much more complex than previously thought, according to a new study published today.

An international team of researchers led by King's College London and Erasmus University Medical Center Rotterdam have identified 50 new genes for eye colour in the largest genetic study of its kind to date. The study, published today in Science Advances, involved the genetic analysis of almost 195,000 people across Europe and Asia.

These findings will help to improve the understanding of eye diseases such as pigmentary glaucoma and ocular albinism, where eye pigment levels play a role.

In addition, the team found that eye colour in Asians with different shades of brown is genetically similar to eye colour in Europeans ranging from dark brown to light blue.

This study builds on previous research in which scientists had identified a dozen genes linked to eye colour, believing there to be many more. Previously, scientists thought that variation in eye colour was controlled by one or two genes only, with brown eyes dominant over blue eyes.

Co-senior author Dr Pirro Hysi, King's College London, said: "The findings are exciting because they bring us to a step closer to understanding the genes that cause one of the most striking features of the human faces, which has mystified generations throughout our history. This will improve our understanding of many diseases that we know are associated with specific pigmentation levels."

Co-senior author Dr Manfred Kayser, Erasmus University Medical Center Rotterdam, said:

"This study delivers the genetic knowledge needed to improve eye colour prediction from DNA as already applied in anthropological and forensic studies, but with limited accuracy for the non-brown and non-blue eye colours."

Robots solving computer games, recognizing human voices, or helping in finding optimal medical treatments: those are only a few astonishing examples of what the field of artificial intelligence has produced in the past years. The ongoing race for better machines has led to the question of how and with what means improvements can be achieved. In parallel, huge recent progress in quantum technologies have confirmed the power of quantum physics, not only for its often peculiar and puzzling theories, but also for real-life applications. Hence, the idea of merging the two fields: on one hand, artificial intelligence with its autonomous machines; on the other hand, quantum physics with its powerful algorithms.

Over the past few years, many scientists have started to investigate how to bridge these two worlds, and to study in what ways quantum mechanics can prove beneficial for learning robots, or vice versa. Several fascinating results have shown, for example, robots deciding faster on their next move, or the design of new quantum experiments using specific learning techniques. Yet, robots were still incapable of learning faster, a key feature in the development of increasingly complex autonomous machines.

Within an international collaboration led by Philip Walther, a team of experimental physicists from the University of Vienna, together with theoreticians from the University of Innsbruck, the Austrian Academy of Sciences, the Leiden University, and the German Aerospace Center, have been successful in experimentally proving for the first time a speed-up in the actual robot's learning time. The team has made use of single photons, the fundamental particles of light, coupled into an integrated photonic quantum processor, which was designed at the Massachusetts Institute of Technology. This processor was used as a robot and for implementing the learning tasks. Here, the robot would learn to route the single photons to a predefined direction. "The experiment could show that the learning time is significantly reduced compared to the case where no quantum physics is used", says Valeria Saggio, first author of the publication.

In a nutshell, the experiment can be understood by imagining a robot standing at a crossroad, provided with the task of learning to always take the left turn. The robot learns by obtaining a reward when doing the correct move. Now, if the robot is placed in our usual classical world, then it will try either a left or right turn, and will be rewarded only if the left turn is chosen. In contrast, when the robot exploits quantum technology, the bizarre aspects of quantum physics come into play. The robot can now make use of one of its most famous and peculiar features, the so called superposition principle. This can be intuitively understood by imagining the robot taking the two turns, left and right, at the same time. "This key feature enables the implementation of a quantum search algorithm that reduces the number of trials for learning the correct path. As a consequence, an agent that can explore its environment in superposition will learn significantly faster than its classical counterpart," says Hans Briegel, who developed the theoretical ideas on quantum learning agents with his group at the University of Innsbruck.

This experimental demonstration that machine learning can be enhanced by using quantum computing shows promising advantages when combining these two technologies. "We are just at the beginning of understanding the possibilities of quantum artificial intelligence" says Philip Walther, "and thus every new experimental result contributes to the development of this field, which is currently seen as one of the most fertile areas for quantum computing".

Biologists from RUDN University suggested adding a marjoram-based supplement to the diet of common carp to support the growth of the fish and improve their resistance to bacterial infections. The results of the study were published in the Fish & Shellfish Immunology journal.

Cyprinus carpio is a type of large omnivore fish that grows 35-40 cm long in three to five years. 4 mln tons of carps were bred in aquacultural farms in 2010. Such farms protect natural populations of Cyprinus carpio while at the same time satisfying the market demand. However, as farmers strive for higher productivity, aquacultural farms become more and more crowded which leads to the lack of nutrients and harms the health of the fish. A team of veterinarians from RUDN University tested the effect of a dietary supplement based on marjoram (a herb from the genus Origanum) on farm-bred carps.

"The profitability of a fish farm directly correlates with the stability and efficiency of production which is based on the health of the fish. Fish under stress are more susceptible to disease, and after being treated with antibiotics for some time they develop resistance. Moreover, antibiotics are harmful to the environment and can potentially affect the health of consumers. The best way to combat diseases is via dietary supplements that suppress stress, improve the health of the fish, and support their immunity," said Morteza Yousefi, PhD, an assistant professor at the Department of Veterinary Medicine, RUDN University.

The team divided 360 fish into four groups and fed them dietary supplements with different concentrations of marjoram for eight weeks. A control group did not receive marjoram at all, while the other three got 100, 200, and 400 mg per 1 kg of body mass, respectively. After 60 days of the diet, 30 fish from each group were injected with the bacterial culture of Aeromonas hydrophila, one of the most dangerous pathogens for carps that breaks down hemoglobin, enters the organs of a fish body, and then begins to produce aerolysin, a toxic substance that causes cell death and tissue damage. Then, the team spent 10 days monitoring the survival rates of the infected fish.

Only 50% of the fish from the control group successfully overcame the infection, while the share of survivors among the fish that received marjoram was 80%. The group that received 200 mg of the supplement demonstrated the most prominent positive effect including a 1.5 times increase in erythrocyte count and hemoglobin, immunoglobulins, lysozyme, and alkaline phosphatase levels. This group also showed the lowest feed conversion rate, i.e. it required less food to gain 1 kg of weight (1.32 mg against 1.62 mg in the control group). According to the veterinarians, it indicated higher immune activity and better barrier functions of the mucus in the fish that received the marjoram supplement.

"The fish that received 200 mg/kg of marjoram demonstrated the highest survival rate. Our study confirms that marjoram extract is an effective dietary supplement for the common carp: it stimulates growth, the production of antioxidants, and the work of the immune system. All these factors helped the fish infected with Aeromonas hydrophila combat the disease," added Morteza Yousefi from RUDN University.

CABI scientist Dr Arne Witt has shared his expertise on invasive alien plant species as part of a new paper which argues that healthy ecosystems are vital in reducing the risk of future pandemics - such as coronaviruses (including COVID-19) - that threaten human health.

The paper - 'Land use-induced spillover: priority actions for protected and conserved area managers' - is published as part of a special issue by the journal PARKS entitled 'COVID-19 and Protected Areas: Essential Reading for a World Haunted by a Pandemic.'

Lead author Dr Jamie K. Reaser - along with a team of researchers from institutions including the African Wildlife Foundation, the University of Oklahoma and the IUCN World Commission on Protected Areas - recommend practices to reduce the risk of future pandemics through 'protected and conserved area management.'

They rightly state that pandemic prevention 'requires that human health be regarded as an ecological service' and call upon multi-lateral conservation frameworks to recognise that 'protected area managers are in the frontline of public health safety.'

The scientists highlight that land use change 'drives the emergence and spread of micro-organisms (pathogens) that infest wildlife and humans with severe consequences for environmental, animal and human health.'

The authors underscore the fact that "The COVID-19 pandemic, caused by the SARS-CoV-2 virus, demonstrates society's inability to respond it a timely manner to novel pathogens. The result is mass human suffering and mortality, bringing substantial moral, ethical and economic dilemmas."

"A protected and conserved areas are the most widely used approaches to securing species and ecological integrity, they have a crucial role to play in safeguarding public health."

"From our perspective, a 'healthy' ecosystem is one in which wildlife-pathogen interactions are in balance, wildlife are not overly stressed or concentrated together by land use-induced changes."

The scientists say that the strategic removal of invasive alien plants, that support populations of zoonotic pathogens, vectors or hosts, can function as an ecological countermeasure.

Dr Witt said, "In Mauritius, for example, invasive alien plants have reduced the habitat quality of the Mauritian flying fox (Pteropus niger), resulting in increased foraging in agricultural lands and urban environments."

"Krivek et al. (2020) showed that non-native plant invasions reduced native fruit production and that weeded forests provide a better habitat for flying foxes. They conclude that their study lends support to invasive alien plant control as a management strategy in mitigating human-wildlife conflicts."

Dr Reaser and the team of researchers put forward a series of proposals that reinforce existing conservation strategies while 'elevating biodiversity conservation as a priority health measure.'

The proposals include conducting surveillance about the occurrence of pathogens, or their clinical diseases, in animal or human populations as well as fostering 'landscape immunity' in conserved and protected areas. The latter of which can be achieved, the scientists suggest, by measures including not disturbing landscapes and 'retaining a full complement of native species and their inter-relationships.'

Dr Reaser and her colleagues conclude by arguing that, "Nations can no longer treat conservation as a second order priority. COVID-19 shows that we should now recognise that protected areas are at the frontline of public health infrastructure and that their managers are vital to disease prevention."

"Looking ahead, we have to conserve nature as if our lives depended on it."

New scientific techniques are revealing the intricate role that proteins play in photosynthesis.

Despite being discovered almost 300 years ago, photosynthesis still holds many unanswered questions for science, particularly the way that proteins organise themselves to convert sunlight into chemical energy and at the same time, protect plants from too much sunlight.

Now a collaboration between researchers at the University of Leeds and Kobe University in Japan is developing a novel approach to the investigation of photosynthesis.

Using hybrid membranes that mimic natural plant membranes and advanced microscopes, they are opening photosynthesis to nanoscale investigation - the study of life at less than one billionth of a metre - to reveal the behaviour of individual protein molecules.

Dr Peter Adams, Associate Professor in the School of Physics and Astronomy at the University of Leeds, who supervised the research, said: "For many decades scientists have been developing an understanding of photosynthesis in terms of the biology of whole plants. This research is tackling it at the molecular level and the way proteins interact.

"A greater understanding of photosynthesis will benefit humankind. It will help scientists identify new ways to protect and boost crop yields, as well as inspire technologists to develop new solar-powered materials and components."

The findings are published in the academic journal Small.

Photosynthesis happens when photons or packets of light energy cause pigments inside light-harvesting proteins to become excited. The way that these proteins arrange themselves determines how the energy is transferred to other molecules.

It is a complex system that plays out across many different pigments, proteins, and layers of light-harvesting membranes within the plant. Together, it regulates energy absorption, transfer, and the conversion of this energy into other useful forms.

To understand this intricate process, scientists have been using a technique called atomic force microscopy, which is a device capable of revealing components of a membrane that are a few nanometres in size.

The difficulty is that natural plant membranes are very fragile and can be damaged by atomic force microscopy.

But last year, the researchers at Kobe University announced that they had developed a hybrid membrane made up of natural plant material and synthetic lipids that would act as a substitute for a natural plant membrane - and crucially, is more stable when placed in an atomic force microscope.

The team at the University of Leeds used the hybrid membrane and subjected it to atomic force microscopy and another advanced visualisation technique called fluorescence lifetime imaging microscopy, or FLIM.

PhD researcher Sophie Meredith, also from the School of Physics at the University of Leeds, is the lead author in the paper. She said: "The combination of FLIM and atomic force microscopy allowed us to observe the elements of photosynthesis. It gave us an insight into the dynamic behaviors and interactions that take place.

"What is important is that we can control some of the parameters in the hybrid membrane, so we can isolate and control factors, and that helps with experimental investigation.

"In essence, we now have a 'testbed' and a suite of advanced imaging tools that will reveal the sub-molecular working of photosynthesis."

COLUMBUS, Ohio - Electric stimulation may be able to help blood vessels carry white blood cells and oxygen to wounds, speeding healing, a new study suggests.

The study, published in the Royal Society of Chemistry journal Lab on a Chip, found that steady electrical stimulation generates increased permeability across blood vessels, providing new insight into the ways new blood vessels might grow.

The electrical stimulation provided a constant voltage with an accompanying electric current in the presence of fluid flow. The findings indicate that stimulation increases permeability of the blood vessel - an important characteristic that can help wound-healing substances in the blood reach injuries more efficiently.

"There was this speculation that blood vessels could grow better if you stimulated them electrically," said Shaurya Prakash, senior author of the study and associate professor of mechanical and aerospace engineering at The Ohio State University. "And we found that the response of the cells in our blood vessel models shows significant promise towards changing the permeability of the vessels that can have positive outcomes for our ongoing work in wound healing."

Blood vessels are crucial for wound healing: They thread throughout your body, carrying nutrients, cells and chemicals that can help control inflammation caused by an injury. Oxygen and white blood cells - which protect the body from foreign invaders - are two key components delivered by blood vessels.

But when there is an injury - for example, a cut on your finger - the architecture of the blood vessels at the wound site are disrupted. That also interrupts the vessels' ability to help the wound heal. Blood vessels regrow on their own, almost like the branches of trees, without external sources of electricity, as part of the healing process.

"And as the blood vessels begin to grow, they replenish the skin and cells and establish a healing barrier again," Prakash said. "But our question was: How do you make this process better and faster, and is there any benefit to doing that?"

What they found, in laboratory tests performed using human cells, is that stimulating blood vessels with electricity showed a marked increase in blood vessel permeability, which is a physical marker suggestive of possible new vessel growth.

"These initial findings are exciting, and the next phase of the work will require us to study if and how we can actually grow new vessels," Prakash said.

Jon Song, co-author of the paper and associate professor of mechanical and aerospace engineering at Ohio State, said the results imply that one of the primary ways blood vessels work to heal injuries is by allowing molecules and cells to move across the vessels' walls.

"And now we have better understanding for how electric stimulation can change the permeability across the vessel walls," Song said. "Let's say you have a cutaneous wound, like a paper cut, and your blood vessels are severed and that's why you have blood leaking out. What you need is a bunch of bloodborne cells to come to that place and exit out the blood vessel to initiate the wound repair."

The study suggested that changes in blood vessel permeability could get those bloodborne cells to a wound site more quickly, though it did not explain the reasons why that happened. The study seemed to indicate that electricity affected the proteins that hold blood vessel cells together, but those results were not conclusive.

The study is an extension of work by a broader team, led by Prakash, that previously showed electric bandages could help stimulate healing in wounded dogs. That work indicated that electrical stimulation might also help manage infections at wound sites - a phenomenon the researchers also hope to research further.

At the heart of Cygnus, one of the most beautiful constellations of the summer sky, beats a source of high-energy cosmic ray particles: the Cygnus Cocoon. An international group of scientists at the HAWC observatory has gathered evidence that this vast astronomical structure is the most powerful of our galaxy's natural particle accelerators known of up to now.

This spectacular discovery is the result of the work of scientists from the international High-Altitude Water Cherenkov (HAWC) gamma-ray observatory. Located on the slopes of the Mexican Sierra Negra volcano, the observatory records high-energy particles and photons flowing from the abyss of space. In the sky of the Northern Hemisphere, their brightest source is the region known as the Cygnus Cocoon. At the HAWC, it was established that photons with energies even several dozen times greater than those recorded earlier by the Fermi-LAT and ARGO detectors arrive from the Cocoon. This fact suggests that the Cygnus Cocoon is the most powerful of the hitherto known particle accelerators in the Milky Way. The results of the research, in which scientists from the Institute of Nuclear Physics of the Polish Academy of Sciences (IFJ PAN) in Cracow played an important role, are presented in the prestigious journal Nature Astronomy.

"The discovery made thanks to the HAWC observatory is an important element of a scientific puzzle that has been going on for over a hundred years, the aim of which is to decipher the nature of cosmic rays, especially when speaking of particles with the highest energies occurring in our galaxy," says Dr. Sabrina Casanova (IFJ PAN), initiator of the latest data analysis from the Cygnus Cocoon region and its significant co-author.

The Cygnus Cocoon, a vast astronomical structure about 180 light years across, lies 4.6 thousand light years from the Sun. In our sky, we can find it almost exactly in the centre of the Cygnus constellation where it occupies an area of angular width similar to the four discs of the Moon. It is a region of intense formation of massive (and consequently short-lived) stars, with two young star clusters Cygnus OB2 and NGC 6910.

"The HAWC detector has greater sensitivity and angular resolution than previous devices of this type. With its use, during 1,343 days of observation we recorded photons of gamma rays with energies up to one hundred teraelectronvolts, coming from the direction of the Cygnus OB2 cluster. What could have been the source of such high-energy radiation?" wonders Dr. Casanova.

What emerges from the latest analysis of gamma radiation reaching the Earth from the Cygnus Cocoon is an interesting picture of phenomena of complex, multi-stage nature. High-energy cosmic rays are usually expected to come from supernova remnants, including pulsars. However, in supernova remnants protons or electrons do not have enough time to accelerate to a kinetic energy reaching several hundred teraelectronvolts. But then, inside a young cluster of massive stars, the turbulences of powerful stellar winds interacting with each other help to confine the particles for millions of years. Some of these particles have the chance to gain energies reaching petaelectronvolts.

"The situation is very complicated," notes Dr. Casanova, and clarifies: "Some particles are expected to gain really enormous energies inside these associations of massive stars thanks to the long confinement time, comparable to the million-year lifetime of the associations themselves. But the higher the energy of the particles, the shorter the confinement time. We expect that the highest energy particles escape the cluster before they emit the gamma photons we can observe. The question is: where is the maximum acceleration energy?"

The key question is the nature of the particles responsible for the emission of the high-energy photons that were recorded at the HAWC observatory. If the source of the photons were electrons, their energies would have to be several times greater than the energy of photons. However, if the source were protons, their energies would have to be as high as a petaelectronvolt. This value is one hundred times greater than the energy of proton collisions inside the LHC accelerator.

"Our analysis does not provide a clear conclusion regarding the origin of photons with energies reaching 100 TeV. It does, however, point to a clear favourite - protons with extreme energies, accelerated in collisions of stellar winds and then emitting gamma photons when colliding with interstellar material," says Dr Casanova.

If future observations confirm the present interpretation, the Cygnus OB2 star cluster in the interior of the Cygnus Cocoon would be the most powerful of all our galaxy's accelerators identified so far.

The Henryk Niewodniczanski Institute of Nuclear Physics (IFJ PAN) is currently one of the largest research institutes of the Polish Academy of Sciences. A wide range of research carried out at IFJ PAN covers basic and applied studies, from particle physics and astrophysics, through hadron physics, high-, medium-, and low-energy nuclear physics, condensed matter physics (including materials engineering), to various applications of nuclear physics in interdisciplinary research, covering medical physics, dosimetry, radiation and environmental biology, environmental protection, and other related disciplines. The average yearly publication output of IFJ PAN includes over 600 scientific papers in high-impact international journals. Each year the Institute hosts about 20 international and national scientific conferences. One of the most important facilities of the Institute is the Cyclotron Centre Bronowice (CCB), which is an infrastructure unique in Central Europe, serving as a clinical and research centre in the field of medical and nuclear physics. In addition, IFJ PAN runs four accredited research and measurement laboratories. IFJ PAN is a member of the Marian Smoluchowski Kraków Research Consortium: "Matter-Energy-Future", which in the years 2012-2017 enjoyed the status of the Leading National Research Centre (KNOW) in physics. In 2017, the European Commission granted the Institute the HR Excellence in Research award. The Institute holds A+ Category (the highest scientific category in Poland) in the field of sciences and engineering.

Alexandria, Va., USA -- Oral mucositis and taste dysfunction (dysgeusia) occurs in nearly all patients receiving head and neck radiotherapy and tremendously affects the quality of life and treatment outcome. The study "LiCl Promotes Recovery of Radiation-Induced Oral Mucositis and Dysgeusia" published in the Journal of Dental Research (JDR), investigated the hypothesis that lithium chloride (LiCl) can promote the restoration of oral mucosa integrity and taste function after radiation.

LiCl is a potent activator of a key cell signaling pathway called Wnt/β-catenin that is critical for the development, regeneration and function of many tissue types. Researchers at the West China Hospital of Stomatology, Sichuan University, Chengdu, China, the University of Illinois at Chicago, USA, and the Clinical Research Center for Oral Diseases of Sichuan Province, Chengdu, China, compared the treatment efficacy of LiCl on oral mucositis to the treatment efficacy using keratinocyte growth factor, an agent approved by the U.S. Food and Drug Administration for oral mucositis.

The results showed that LiCl alleviated the weight loss and tongue ulceration of the mice with radiation-induced oral mucositis, as well promoted proliferation of basal epithelial cells and inhibited epithelial-mesenchymal transition in tongue mucosa. The mice treated with LiCl showed an elevated taste bud renewal and taste-distortion recovery toward sweetness as compared to the mice treated using keratinocyte growth factor.

"The data from this study demonstrate that LiCl can mitigate radiation-induced oral mucositis by promoting cell proliferation in the basal layer and inhibiting epithelial-mesenchymal transition of the oral mucosa," said JDR Editor-in-Chief Nicholas Jakubovics, Newcastle University, England. "Lithium has been used in clinical practice for decades and is currently recommended as a first-line therapy for bipolar disorder. It is exciting to see that LiCl may also represent a promising new therapy to improve the quality of life of patients receiving radiotherapy."

BUFFALO, N.Y. - A University at Buffalo researcher's recent work on dyslexia has unexpectedly produced a startling discovery which clearly demonstrates how the cooperative areas of the brain responsible for reading skill are also at work during apparently unrelated activities, such as multiplication.

Though the division between literacy and math is commonly reflected in the division between the arts and sciences, the findings suggest that reading, writing and arithmetic, the foundational skills informally identified as the three Rs, might actually overlap in ways not previously imagined, let alone experimentally validated.

"These findings floored me," said Christopher McNorgan, PhD, the paper's author and an assistant professor in UB's Department of Psychology. "They elevate the value and importance of literacy by showing how reading proficiency reaches across domains, guiding how we approach other tasks and solve other problems.

"Reading is everything, and saying so is more than an inspirational slogan. It's now a definitive research conclusion."

And it's a conclusion that was not originally part of McNorgan's design. He planned to exclusively explore if it was possible to identify children with dyslexia on the basis of how the brain was wired for reading.

"It seemed plausible given the work I had recently finished, which identified a biomarker for ADHD," said McNorgan, an expert in neuroimaging and computational modeling.

Like that previous study, a novel deep learning approach that makes multiple simultaneous classifications is at the core of McNorgan's current paper, which appears in the journal Frontiers in Computational Neuroscience.

Deep learning networks are ideal for uncovering conditional, non-linear relationships.

Where linear relationships involve one variable directly influencing another, a non-linear relationship can be slippery because changes in one area do not necessarily proportionally influence another area. But what's challenging for traditional methods is easily handled through deep learning.

McNorgan identified dyslexia with 94% accuracy when he finished with his first data set, consisting of functional connectivity from 14 good readers and 14 poor readers engaged in a language task.

But he needed another data set to determine if his findings could be generalized. So McNorgan chose a math study, which relied on a mental multiplication task, and measured functional connectivity from the fMRI information in that second data set.

Functional connectivity, unlike what the name might imply, is a dynamic description of how the brain is virtually wired from moment to moment. Don't think in terms of the physical wires used in a network, but instead of how those wires are used throughout the day. When you're working, your laptop is sending a document to your printer. Later in the day, your laptop might be streaming a movie to your television. How those wires are used depends on whether you're working or relaxing. Functional connectivity changes according to the immediate task.

The brain dynamically rewires itself according to the task all the time. Imagine reading a list of restaurant specials while standing only a few steps away from the menu board nailed to the wall. The visual cortex is working whenever you're looking at something, but because you're reading, the visual cortex works with, or is wired to, at least for the moment, the auditory cortex.

Pointing to one of the items on the board, you accidentally knock it from the wall. When you reach out to catch it, your brain wiring changes. You're no longer reading, but trying to catch a falling object, and your visual cortex now works with the pre-motor cortex to guide your hand.

Different tasks, different wiring; or, as McNorgan explains, different functional networks.

In the two data sets McNorgan used, participants were engaged in different tasks: language and math. Yet in each case, the connectivity fingerprint was the same, and he was able to identify dyslexia with 94% accuracy whether testing against the reading group or the math group.

It was a whim, he said, to see how well his model distinguished good readers from poor readers - or from participants who weren't reading at all. Seeing the accuracy, and the similarity, changed the direction of the paper McNorgan intended.

Yes, he could identify dyslexia. But it became obvious that the brain's wiring for reading was also present for math.

Different task. Same functional networks.

"The brain should be dynamically wiring itself in a way that's specifically relevant to doing math because of the multiplication problem in the second data set, but there's clear evidence of the dynamic configuration of the reading network showing up in the math task," McNorgan says.

He says it's the sort of finding that strengthens the already strong case for supporting literacy.

"These results show that the way our brain is wired for reading is actually influencing how the brain functions for math," he said. "That says your reading skill is going to affect how you tackle problems in other domains, and helps us better understand children with learning difficulties in both reading and math."

As the line between cognitive domains becomes more blurred, McNorgan wonders what other domains the reading network is actually guiding.

"I've looked at two domains which couldn't be farther afield," he said. "If the brain is showing that its wiring for reading is showing up in mental multiplication, what else might it be contributing toward?"

That's an open question, for now, according to McNorgan.

"What I do know because of this research is that an educational emphasis on reading means much more than improving reading skill," he said. "These findings suggest that learning how to read shapes so much more."

Researchers at Princeton have determined how five cellular proteins contribute to an essential step in the life cycle of hepatitis B virus (HBV). The article describing these findings appeared March 11, 2021 in the journal Nature Communications.

Viruses have been with us, shaping our lives, societies and economies for millennia. While some viruses rapidly explode onto the world stage, others smolder in our communities for decades, shattering lives but making few headlines. Hepatitis B virus hasn't caused any nationwide lockdowns or stock market crashes because it is slow to spread from person to person and is rarely immediately fatal. It is nonetheless incredibly damaging because it can establish lifelong chronic infection with profound consequences for its victims.

"An estimated two billion people have been exposed to HBV, of whom 250-400 million are chronically infected," said Alexander Ploss, associate professor of molecular biology at Princeton University, and senior author on the study. "Currently, there is no cure for chronic HBV infection, and patients need to be on a lifelong antiviral regimen. Approximately 887,000 individuals die each year from HBV-related liver diseases or liver cancer."

Ploss and his team are striving to understand HBV's life cycle in hopes of finding a way to prevent the virus from establishing damaging chronic infections.

"Central to HBV replication is the formation of covalently closed circular DNA (cccDNA) from relaxed circular DNA (rcDNA) which is carried into the host cell by the virus during the initial infection," Ploss said. "We have recently demonstrated that HBV relies on five host proteins - namely PCNA, RFC complex, POLδ, FEN-1, and LIG1 - that are necessary and sufficient for this conversion step."

As its name implies, rcDNA is a loop of DNA. DNA is a molecule made up of nucleotides arranged in linear fashion along paired, complementary strands. The sequence of nucleotides on one strand encodes the instructions for making a protein while the other strand is its mirror image. Whereas human cellular DNA contains over 20,000 genes, HBV's DNA genome only contains four. None of the viral proteins made from these genes is required for the conversion of rcDNA to cccDNA. Instead, the virus coopts cellular proteins to accomplish this and other steps of viral replication.

A key feature of HBV rcDNA is that each of its two strands contains a gap in its nucleotide sequence. One strand, called the plus strand, has a gap that is considerably larger than and offset from the gap on the other, minus strand. Cells perceive gaps in DNA as damage that needs to be filled in and repaired. The cellular proteins that carry out DNA repair can't tell the difference between viral DNA and cellular DNA, so they set to work "repairing" rcDNA as soon as it arrives in the nucleus. This repair process converts rcDNA into an intact circle of double-stranded DNA (that is, cccDNA) that can be maintained in the cell's nucleus.

Postdoctoral fellow Lei Wei wanted to understand how this repair process takes place in detail. To investigate this, he developed a method to monitor the process of repair taking place on rcDNA. He then identified what steps are involved in repair of each individual strand, tracked the order in which they are completed, and determined which cellular proteins are needed for each step.

The experiments showed that conversion of the plus strand into a continuous circle happens rapidly and requires all five of the cellular proteins working in concert. In contrast, repair of the minus strand requires only two of the five proteins (FEN-1 and LIG1) but is slower because there is a viral protein attached to one end of the minus strand that must be removed before the nucleotide gap can be sealed.

"In this paper, Wei and Ploss provide a compelling story in elucidating the cellular machinery that is essential for converting the incoming HBV genome into the cccDNA," said Dr. T. Jake Liang, a National Institutes of Health Distinguished Investigator in the Liver Diseases Branch who is an expert on HBV and related viruses. "This work offers not only important insights into the biochemical pathway of cccDNA biogenesis but also potential strategies to target cccDNA for therapeutic development."

Furthering that goal, the Princeton researchers showed that two compounds targeting cellular proteins could disrupt rcDNA conversion to cccDNA in test tubes. Wei and Ploss are hopeful that future studies will identify drugs that work in the human body.

"Our findings, biochemical approaches, and the novel reagents that we generated and engineered, open the door to providing an in depth understanding how this major human virus establishes persistence in host cells," Ploss said.

CHAMPAIGN, Ill. -- Charitable donations account for about 2% of gross domestic product in the U.S., but it's not well-understood whether an event such as a deadly storm inspires increases in charitable giving or simply reallocates a fixed supply of donation dollars that would have otherwise gone to another cause.

A new paper from a team of University of Illinois Urbana-Champaign experts finds that, in the aftermath of catastrophic tornadoes, charitable giving to alleviate an unanticipated event doesn't necessarily crowd out monetary donations to other causes.

Research co-written by Tatyana Deryugina and Benjamin M. Marx uses comprehensive annual tax return data to estimate whether discrete but destructive events such as fatal tornadoes increase total charitable giving.

The paper, which will be published in the journal American Economic Review: Insights, finds that each tornado fatality increases total donations from individuals living nearby by almost $2 million.

"There is an active debate in the research on charitable giving on the extent of donor substitution between charitable causes," said Deryugina, a professor of finance at Illinois. "By employing a natural experiment with both geographic and temporal variation as well as datasets that contain the majority of dollars donated across the U.S., our paper can rule out the possibility that individuals choose a set amount to donate to charity and then apportion that amount among charities."

The researchers measured charitable giving from 2002-17 using ZIP code-level individual income tax data from the Internal Revenue Service, which reports all income tax deductions claimed for donations to any registered charity.

By combining the IRS data with geospatial information on the incidence of tornadoes, the researchers were able to zero in on the destruction wrought by lethal tornadoes, allowing them to identify events that affected populated areas and caused tens of millions of dollars' worth of damage.

"We looked at tornadoes because they hit a very specific and limited area, which gives us a treatment area and a clean control area so we can get better estimates via comparison," said Marx, a professor of economics at Illinois.

Deryugina and Marx found that lethal tornadoes significantly increased total charitable donations from individuals living at least 20 miles away in the same state from a tornado's path by almost $2 million per fatality.

"We excluded ZIP codes that were within 20 miles of the tornado's path, so we weren't looking at the places that were directly affected by property damage or loss of life," Marx said. "We went outside of that 20-mile ring and out to the rest of the state to estimate the effect on average giving in those other ZIP codes. And if you add up those effects across all of those ZIP codes, you arrive at a total state-level effect, which is where the $2 million figure comes from."

Because total giving increases, the scholars were able to rule out perfect substitution between charitable causes and conclude that the supply of donations from an "altruism budget" can expand given exigent circumstances, Deryugina said.

"We can't rule out that there's some marginal crowding-out effects, because we just find that total giving goes up, but I'm comfortable speculating that there's not a lot of substitution because the total increase is fairly large," she said. "If there is some substitution, it's not huge. Basically, we see no evidence for the existence of a radical pullback in charitable giving to other causes after these natural disasters."

Using annual tax return data from charities, the researchers also found that there were no significant negative effects on charities in the locations from which increased donations to fatal tornadoes originated.

"That implies that giving in response to new needs doesn't come at the expense of these other causes," Marx said.

Charitable giving for natural disasters is common, with a nationally representative survey indicating that about 30% of U.S. households donated money in 2017-18. While the data in the paper covered the majority of dollars donated, it may not be representative of lower-income households, whose charitable donation budget may be more fixed, the researchers said.

"We only had aggregate, ZIP code-level data. We would have loved to break it down by income group, but we just didn't have the data," Deryugina said. "We do note in the paper that most charitable giving in the U.S. is by wealthier people. Lower-income individuals and households are less likely to file taxes, and all of our data came from the IRS. So our estimates are representative of individuals who filed a tax return, but those individuals make up a sizable share of total charitable giving."

ROCHESTER, Minn. -- Ten days after receiving a second dose of a messenger RNA, or mRNA, vaccine for COVID-19, patients without COVID-19 symptoms are far less likely to test positive and unknowingly spread COVID-19, compared to patients who have not been vaccinated for COVID-19. The Pfizer-BioNTech and Moderna messenger RNA vaccines for COVID-19 are authorized for emergency use in the U.S.

With two doses of a messenger RNA COVID-19 vaccine, people with no symptoms showed an 80% lower adjusted risk of testing positive for COVID-19 after their last dose. Those are the findings of a Mayo Clinic study of vaccinated patients. These finding appear in the journal Clinical Infectious Diseases.

The authors say these findings underscore the efficacy of messenger RNA vaccines for COVID-19 to significantly limit the spread of COVID-19 by people with no symptoms who may unknowingly spread the infection to others.

The researchers retrospectively looked at a cohort of 39,000 patients who underwent pre-procedural molecular screening tests for COVID-19. More than 48,000 screening tests were performed, including 3,000 screening tests on patients who had received at least one dose of a messenger RNA COVID-19 vaccine. These screening tests were part of routine COVID-19 testing prior to treatments not related to COVID-19, such as surgeries and other procedures. Patients in the vaccinated group had received at least one dose of a messenger RNA COVID-19 vaccine.

"We found that those patients without symptoms receiving at least one dose of the first authorized mRNA COVID-19 vaccine, Pfizer-BioNTech, 10 days or more prior to screening were 72% less likely to test positive," says Aaron Tande, M.D., a Mayo Clinic infectious diseases specialist and co-first author of the paper. "Those receiving two doses were 73% less likely, compared to the unvaccinated group."

After adjusting for a range of factors, researchers found an 80% risk reduction of testing positive for COVID-19 among those with two doses of a messenger RNA COVID-19 vaccine.

The study was based on patients receiving screening tests between Dec. 17, 2020, and Feb. 8 at Mayo Clinic in Minnesota and Arizona and at Mayo Clinic Health System in Minnesota and Wisconsin.

Additional authors are Benjamin Pollock, Ph.D., co-first author; Nilay Shah, Ph.D.; Gianrico Farrugia, M.D.; Abinash Virk, M.D.; Melanie Swift, M.D.; Laura Breeher, M.D.; Matthew Binnicker, Ph.D.; and Elie Berbari, M.D. ? all of Mayo Clinic. Funding for the study was provided by Mayo Clinic.

How do you turn "dumb" headphones into smart ones? Rutgers engineers have invented a cheap and easy way by transforming headphones into sensors that can be plugged into smartphones, identify their users, monitor their heart rates and perform other services.

Their invention, called HeadFi, is based on a small plug-in headphone adapter that turns a regular headphone into a sensing device. Unlike smart headphones, regular headphones lack sensors. HeadFi would allow users to avoid having to buy a new pair of smart headphones with embedded sensors to enjoy sensing features.

"HeadFi could turn hundreds of millions of existing, regular headphones worldwide into intelligent ones with a simple upgrade," said Xiaoran Fan, a HeadFi primary inventor. He is a recent Rutgers doctoral graduate who completed the research during his final year at the university and now works at Samsung Artificial Intelligence Center.

A peer-reviewed Rutgers-led paper on the invention, which results in "earable intelligence," will be formally published in October at MobiCom 2021, the top international conference on mobile computing and mobile and wireless networking.

Headphones are among the most popular wearable devices worldwide and they continue to become more intelligent as new functions appear, such as touch-based gesture control, the paper notes. Such functions usually rely on auxiliary sensors, such as accelerometers, gyroscopes and microphones that are available on many smart headphones.

HeadFi turns the two drivers already inside all headphones into a versatile sensor, and it works by connecting headphones to a pairing device, such as a smartphone. It does not require adding auxiliary sensors and avoids changes to headphone hardware or the need to customize headphones, both of which may increase their weight and bulk. By plugging into HeadFi, a converted headphone can perform sensing tasks and play music at the same time.

The engineers conducted experiments with 53 volunteers using 54 pairs of headphones with estimated prices ranging from $2.99 to $15,000. HeadFi can achieve 97.2 percent to 99.5 percent accuracy on user identification, 96.8 percent to 99.2 percent on heart rate monitoring and 97.7 percent to 99.3 percent on gesture recognition.

LAWRENCE -- Much like coronavirus, circulating HIV-1 viruses mutate into diverse variants that pose challenges for scientists developing vaccines to protect people from HIV/AIDS.

"AIDS vaccine development has been a decades-long challenge partly because our immune systems have difficulty recognizing all the diverse variants of the rapidly mutating HIV virus, which is the cause of AIDS," said Brandon DeKosky, assistant professor of pharmaceutical chemistry and chemical & petroleum engineering at the University of Kansas.

In the past five years, tremendous progress has been made in identifying better vaccine methods to protect against many different HIV-1 variants. One important step was when scientists at the National Institute of Allergy and Infectious Disease's Vaccine Research Center discovered a promising antibody called vFP16.02. Antibodies are proteins the immune system deploys to target and destroy pathogens and viruses -- and scientists at the NIH determined that the vFP16.02 stimulated by a vaccine had potential to effectively fight HIV-1.

To further understand the promise of vFP16.02-like immune responses as the basis for an eventual vaccine against HIV-1, the team at the NIH/NIAID Vaccine Research Center partnered with the Immune Engineering Laboratory at the KU School of Pharmacy, where scientists in DeKosky's lab set about understanding how the same antibody-based approaches could be even more powerful in the fight against diverse HIV antigens.

Their encouraging findings were just published in the peer-reviewed journal Proceedings of the National Academy of Sciences (PNAS), showing several important mechanistic features about immune protection, including the binding strength of the antibody directly correlated to its ability to neutralize HIV-1.

Co-lead author Bharat Madan, a postdoctoral researcher in the DeKosky lab, led the work.

"We wanted to see whether we could further increase the potency and neutralization breadth of this antibody," Madan said. "That means you could give a lower amount of antibody to someone as a prophylaxis, and it's going to produce a much better and broader immune response. There are various clades of HIV viruses --now I think you must be hearing about various versions of SARS-CoV-2 like the U.K. variant and the South African variant -- likewise, HIV also is quite diverse."

Madan and colleagues used high-throughput screening platforms based on modified yeast cells to display antibody variant proteins, and advanced cell sorting and next-generation DNA sequencing explore a vast library of possible alterations (also known as mutations) to the vFP16.02 antibody. They analyzed the antibodies' response against 208 different HIV strains and "determined the genetic, structural and functional features associated with antibody improvement or fitness."

"In this study, Bharat and colleagues made artificial pathways of anti-HIV-1 immune responses to identify what does and does not work for achieving better protection against HIV-1," DeKosky said. "The results were a bit surprising and showed that the pathway to effective HIV-1 protection might be much shorter than we previously thought."

The research focused on improving structural and biophysical features in the antibodies that bind to the HIV fusion peptide, "a known vulnerable site on HIV-1."

"There are certain regions on this antibody which are more prone to accept the mutation that can enhance the binding affinity towards the fusion peptide," Madan said. "We found a variant or a mutation -- kind of a cluster towards the bottom portion of the antibody we call the framework region -- that can enhance binding strength, and that correlates not towards a particular variant, but to more diverse HIV strains as well."

The most effective variants in the library of mutants boosted the ability of vFP16.02 to recognize and bind to both soluble fusion peptides and the full HIV-1 envelope protein that is involved in viral entry of host cells. These antibody variants, which had mutations concentrated in antibody framework regions, achieved up to 37% neutralization breadth compared to 28% neutralization of the unmodified vFP16.02 antibody.

While the work won't immediately result in development of a vaccine, it could point the way forward for vaccines that would protect against an array of HIV-1 variants.

"The study does a deep dive into the mechanisms of how the immune system specifically recognizes HIV-1 in ways that lead to protective benefits," Dekosky said. "It reveals some of the key insights in how to best target HIV-1 proteins in the next generation of vaccine designs."