Tech

Bad sleep causes severe health issues and affects our ability to concentrate, memorize, and cope with challenging situations. Individuals with neurodevelopmental disorders such as autism and intellectual disability, frequently suffer from sleep problems. However, little is known about their underlying mechanisms. In Science Advances, a Dutch-American research team, coordinated by Radboudumc, now describes how these problems can arise. Mimicking two genetic causes of autism in fruit flies, they uncovered that flies show the same sleep problems as the patients, and that the disturbed sleep is caused by high levels of serotonin - also frequently observed in autism. Moreover, they found that the origin of high serotonin and sleep problems resides on the glial cells of the blood-brain barrier. This completely new information sheds light on sleep problems in humans and even suggests a possible treatment.

Poor sleep seriously affects cognitive abilities and quality of life. Already vulnerable individuals with autism and related neurodevelopmental disorders suffer particularly often from sleep disturbance, and with them their families as well. Yet, hardly any research has been conducted into the underlying mechanisms of these sleep problems. A Dutch-American team of researchers coordinated by professor Annette Schenck, in collaboration with professor Tjitske Kleefstra - both at Radboudumc - has investigated this problem in humans as well as in fruit flies. Schenck: "We first looked very closely at the sleep problems in two specific patient groups with neurodevelopmental disorders. They have mutations in the CHD8 gene, a leading genetic cause of autism, or in a closely related gene, CHD7, giving raise to CHARGE syndrome. We see that the bad sleep in these disorders particularly comes from problems falling and staying asleep, which causes night awakenings and low sleep quality. We call this problem sleep fragmentation. It is frequent in autism in general, but even more frequent in the individuals with mutations in CHD8 or CHD7. According to affected families, these sleep problems are one of their biggest problems in daily life management. This motivated us to study sleep disturbances, in context of these genes and disorders further".

From humans to fruit flies

To be able to look into brains, investigate problems, and test interventions, researchers turn to animal models. In the fruit fly, an animal model that has led to multiple Nobel prizes, CHD8 and CHD7 are represented by a single gene termed kismet. Mutations in kismet in the fruit fly can therefore tell a lot about mutations in those two "orthologous" genes in humans. Mireia Coll-Tané, researcher in Schenck's group and lead author of the study published in Science Advances: "We see that flies with mutations in kismet have problems staying asleep, waking up during night extremely frequently. They show the same characteristics that we see in people with mutations in CHD8 and CHD7."

Not the neurons, but the glial cells

Schenck's research group is specialized in fruit fly research on genes causing neurodevelopmental disorders. They routinely generate flies with mutations in genes that correspond to the disease genes, and look at changes in behaviors or other properties of the nervous system that are present in both humans and flies. Schenck: "when Mireia found the sleep fragmentation to be present also in flies, we knew that can use our model with all of its advantages to find out where this problem comes from". They found that kismet is important for good sleep of adult flies but also already earlier during development, and that the adult sleep defects result from decreased kismet during the developmental period. Coll-Tané: "We also found that kismet was not important in the neurons, the cells that are classically seen to regulate behavior, but in the other main cell type present in the brain: the glial cells. Glia have many important functions, such as supporting neurons, cleaning up waste and contributing to the blood-brain barrier. We saw that kismet is important, already in early development, in a group of only 300 glial cells that form the blood-brain barrier in the fly. They are the origin of the sleep fragmentation."

Surprising role for serotonin

Often the neurotransmitter dopamine plays a role in sleeping problems. This study also looked at this, but dopamine levels were normal. In contrast, the neurotransmitter serotonin appeared to be important. Coll-Tané: "When we reduce kismet specifically in glia, we found the concentration of serotonin in fruit fly heads to be doubled. This is a very interesting finding because increased serotonin, also referred to as hyperserotonemia, is one of the most commonly found biomarkers in autism." In a series of further genetic and drug experiments, the researchers provided evidence that the increased serotonin levels during development are responsible for kismet's sleep fragmentation. Coll-Tané: "Our work has linked a leading genetic cause of autism to a frequent biomarker and an important clinical complaint in autism. We propose that our identified mechanism is relevant to autism more widely".

Future therapy?

An important question is whether the sleep problems associated with the developmental disorders can be tackled after all. It must be a treatment that works in adults (ie after development), and that is non-invasive and safe. Schenck: "Our co-authors in Philadelphia recently developed sleep-restriction therapy for fruit flies - SRT for short. An equivalent intervention is widely used in humans, in otherwise healthy individuals with insomnia. But it is rarely applied to autism and neurodevelopmental disorders, perhaps because sleep defects are considered an inevitable consequence of the genetic mutations. But the behavioral therapy succeeded! A simple light regime mimicking shorter nights made the flies sleeping better, effectively reversing the sleep fragmentation. And this despite that the origin of the problems lies earlier in development."

Further research in humans

With the article in Science Advances, further research in humans is obvious. Clinical geneticist Kleefstra: "We have already shown in our article that CHD7 and CHD8 are expressed in the human blood-brain barrier, both during development and adulthood. Now we aim to collect further clinical data and apply SRT to these patients, in close collaboration with the expert sleep clinic Kempenhaeghe in Heeze. Together, we are expanding our 'human-to-fruit-fly-and-back' strategy to a number of other disorders." Clearly, the path to more fascinating sleep research in flies and humans is up for grabs with this publication.

Many contemporary political conflicts are between those who would prioritize the needs of local or national communities and those with a more universal outlook. According to a new study by IASS researcher Silvia Weko, this split between "communitarian" and "cosmopolitan" Europeans is also evident in their attitudes towards European climate policy. Achieving climate neutrality without exacerbating societal divisions within and between countries will require the EU to strike a careful balance.

In political philosophy cosmopolitans and communitarians are frequently characterized as "winners" and "losers" of globalization. Their respective gains and losses have an objective and a subjective dimension: globalization can be perceived as threatening or promising in view of such hard facts as job losses or due to ideologies or values. The most important subjective difference is the perception of community. Cosmopolitans see community as universal, conceiving of individuals as equal regardless of group membership, while communitarians perceive community in terms of belonging to a particular group, often defined locally.

Ideology plays a larger role than party preferences

This division is a challenge for the EU's climate policy. The successful implementation of the Green Deal, which aims to make Europe completely carbon-neutral by 2050, will require the backing of publics across the Member States. Silvia Weko has analysed data from the European Social Survey to ascertain how deep the divisions between cosmopolitan and communitarian Europeans - evident in other policy areas such as migration - are in relation to climate and energy.

"The results show that political ideology is also highly influential in shaping attitudes towards climate policy. Indeed, it influences attitudes towards climate policy across all of the countries examined and to a stronger degree than individuals' identification along the left-right divide", explains the sociologist. The more cosmopolitan a person's political ideology, the more positive their attitudes are towards climate action. In the case of both groups, subjective aspects have a greater influence on attitudes towards climate protection than objective circumstances. On average, people in Eastern European countries are less concerned about climate change than those in Western European countries. Instead, their priorities are energy security and economic development.

Surveys can support decision-making on climate policy

The EU now faces the challenge of winning over communitarian-minded Europeans and ensuring that no one is 'left behind' as it moves towards climate neutrality. Weko identifies a number of promising signs: "The European Commission has stressed its intention to ensure that no one will be 'left behind'. One concrete measure, would be to tackle energy poverty, which is much more widespread in Eastern Europe, by renovating social housing, schools, and hospitals." The "Just Transitions Fund" could also help to support regions hit particularly hard by structural change.

Weko also suggests that the European Commission should regularly assess attitudes towards climate and energy policy, ideally in an annual survey, just as it already does in relation to issues around integration and migration. Gaining a better understanding of people's concerns would help to address future obstacles to achieving climate neutrality.

The lack of data is a major bottleneck for many kinds of research, and especially for the development of better medical treatments and drugs. This data is extremely sensitive and, understandably, people and companies alike are often unwilling to share their information with others.

Researchers at the Finnish Center for Artificial Intelligence have developed a machine learning-based method that produces synthetic data on the basis of original data sets, making it possible for researchers to share their data with one other. This could solve the ongoing problem of data scarcity in medical research and other fields where information is sensitive.

The generated data preserves privacy, remaining similar enough to the original data to be used for statistical analyses. With the new method, researchers can conduct an infinite number of analyses without compromising the identities of the individuals involved in the original experiment.

"What we do is we tweak the original data sufficiently so that we can mathematically guarantee that no individual can be recognized," explains Samuel Kaski, Aalto University Professor and Director of FCAI, who co-authored the study.

Researchers have produced and used synthetic data before, but the new study solves a major problem with existing methods.

"We might think that just because data is synthetic, it's safe. This has not necessarily been the case, though," explains Kaski.

This is because synthetic data needs to be very similar to the original data set in order to be useful in research. In practice, it has occasionally been possible to identify individuals' identities despite anonymization.

To address this problem, FCAI researchers make use of artificial intelligence, specifically probabilistic modelling. This enables them to use prior knowledge about the original data and the processes that have made it the way it is - without getting too close to the properties of the particular data set used as basis for the synthetic data. Such prior knowledge, for instance, could relate to known gender differences in alcohol-related mortalities, or could involve domain knowledge about how a particular data set has been collected.

Making use of prior knowledge has also made the synthetic data sets more useful for making correct statistical discoveries -- even in cases where the original data set is limited in size, which is common in medical research.

"Incorporating prior knowledge means we can use the method with small data sets, for which we have domain knowledge," Kaski says.

The results are published 7 June in the journal Patterns.

COLUMBUS, Ohio - Researchers have discovered a new electronic property at the frontier between the thermal and quantum sciences in a specially engineered metal alloy - and in the process identified a promising material for future devices that could turn heat on and off with the application of a magnetic "switch."

In this material, electrons, which have a mass in vacuum and in most other materials, move like massless photons or light - an unexpected behavior, but a phenomenon theoretically predicted to exist here. The alloy was engineered with the elements bismuth and antimony at precise ranges based on foundational theory.

Under the influence of an external magnetic field, the researchers found, these oddly behaving electrons manipulate heat in ways not seen under normal conditions. On both the hot and cold sides of the material, some of the electrons generate heat, or energy, while others absorb energy, effectively turning the material into an energy pump. The result: a 300% increase in its thermal conductivity.

Take the magnet away, and the mechanism is turned off.

"The generation and absorption form the anomaly," said study senior author Joseph Heremans, professor of mechanical and aerospace engineering and Ohio Eminent Scholar in Nanotechnology at The Ohio State University. "The heat disappears and reappears elsewhere - it is like teleportation. It only happens under very specific circumstances predicted by quantum theory."

This property, and the simplicity of controlling it with a magnet, makes the material a desirable candidate as a heat switch with no moving parts, similar to a transistor that switches electrical currents or a faucet that switches water, that could cool computers or increase the efficiency of solar-thermal power plants.

"Solid-state heat switches without moving parts are extremely desirable, but they don't exist," Heremans said. "This is one of the possible mechanisms that would lead to one."

The research is published today (June 7, 2021) in the journal Nature Materials.

The bismuth-antimony alloy is among a class of quantum materials called Weyl semimetals - whose electrons don't behave as expected. They are characterized by properties that include negatively and positively charged particles, electrons and holes, respectively, that behave as "massless" particles. Also part of a group called topological materials, their electrons react as if the material contains internal magnetic fields that enable the establishment of new pathways along which those particles move.

In physics, an anomaly - the electrons' generation and absorption of heat discovered in this study - refers to certain symmetries that are present in the classical world but are broken in the quantum world, said study co-author Nandini Trivedi, professor of physics at Ohio State.

Bismuth alloys and other similar materials also feature classical conduction like most metals, by which vibrating atoms in a crystal lattice and the movement of electrons carry heat. Trivedi described the new pathway along which light-like electrons manipulate heat among themselves as a highway that seems to appear out of nowhere.

"Imagine if you were living in a small town that had tiny roads, and suddenly there's a highway that opens up," she said. "This particular pathway only opens up if you apply a thermal gradient in one direction and a magnetic field in the same direction. So you can easily close the highway by putting the magnetic field in a perpendicular direction.

"No such highways exist in ordinary metals."

When a metal like copper is heated and electrons flow from the hot end to the cold end, both the heat and the charge move together. Because of the way this highway opens in the experimental Weyl semimetal material, there's no net charge motion - only energy movement. The absorption of heat by certain electrons represents a break in chirality, or directionality, meaning that it's possible to pump energy between two particles that wouldn't be expected to interact - another characteristic of Weyl semimetals.

The theoretical physicists and engineers collaborating on this study predicted that these properties existed in specific bismuth alloys and other topological materials. For these experiments, the scientists constructed the specialized alloy to test their predictions.

"We worked hard to synthesize the correct material, which was designed from the ground up by us to show this effect. It was important to purify it way below the levels of impurities that you find in nature," Heremans said. As composed, the alloy minimized background conduction so the researchers could detect the behavior of the massless electrons, known as Weyl Fermions.

"In ordinary materials, electrons drag around with them a small magnet. However, the peculiar electronic structure of these bismuth alloys means the electrons drag around a magnet almost 50 times bigger than normal," said Michael Flatté, professor of physics and astronomy at the University of Iowa and a study co-author. "These huge subatomic magnets allowed the novel electronic state to be formed using laboratory magnetic fields.

"These results show that theories developed for high-energy physics and subatomic particle theories can often be realized in specially designed electronic materials."

Like everything quantum, Heremans said, "what we observed looks a little like magic, but that is what our equations say it should do and that is what we proved experimentally that it does."

One catch: The mechanism in this material works only at a low temperature, below minus 100 degrees Fahrenheit. With the fundamentals now understood, the researchers have lots of options as they work toward potential applications.

"Now we know what materials to look for and what purity we need," Heremans said. "That is how we get from discovery of a physical phenomenon to an engineering material."

Research between the University of Liverpool, UK and National Tsing Hua University (NTHU), Taiwan has revealed a new charge storage mechanism that has the potential to allow rechargeability within calcium-air batteries.

In a paper published in the journal Chemical Science, Professor Laurence Hardwick from the University of Liverpool's Stephenson Institute for Renewable Energy (SIRE) and colleagues discover a distinctive form of charge storage at the electrode interface described as trapped interfacial redox. This new finding introduces a new mechanism of charge storage that could be harnessed in practical devices.

Lead author of the paper, Yi Ting (Leo) Lu, is a joint PhD student in the dual PhD programme between the University of Liverpool and National Tsing Hua University. He said: "We started investigating these electrolyte systems as part of understanding how to develop a metal-air battery based upon calcium metal, which is a highly earth abundant element, creating a highly sustainable battery technology."

"The research explores the formation of an electrochemically generated interlayer coating on electrode surfaces that confines the reduced form of oxygen gas known as superoxide, allowing it to then be readily oxidised."

The research was carried out in an electrolyte designed for a calcium-air battery, which had so far been shown to be practically irreversible. The research team noticed that, when the electrode was cycled many tens of times, the electrochemical process became steadily more reversible, and a series of experiments were conducted to fully understand the mechanism.

Dr Alex Neale, who is also part of the research team, said: "Through systematic electrochemical and spectroscopy investigations, we began to understand the origins of this reasonably strange and exciting new process appearing in our measurements. The new mechanism of trapped interfacial redox we defined facilitates a previously unseen degree of reversibility for systems based on the calcium-air battery."

Further work will explore how readily this phenomenon is observed in different electrolyte systems and to understand whether the charge stored can be further exploited, scaled up and used in a practical system for energy storage.

Due to the UK lockdown in March 2020, Yi Ting (Leo) Lu returned to Taiwan 6 months earlier than planned, so to complete the study the team set up parallel experiments within both Liverpool and NTHU labs to cross compare results and ensure reproducibility of observations.

Metal halide perovskites have been under intense investigation over the last decade due to the remarkable rise in their performance in optoelectronic devices such as solar cells or light-emitting diodes. Despite tremendous progress in this field, many fundamental aspects of the photophysics of perovskite materials remain unknown, such as a detailed understanding of their defect physics and charge recombination mechanisms. These are typically studied by measuring the photoluminescence - i.e. the emission of light upon photoexcitation - of the material in both the steady-state and transient regimes. While such measurements are ubiquitous in literature, they do not capture the full range of the photophysical processes that occur in metal halide perovskites and thus represent only a partial picture of their charge carrier dynamics. Moreover, while several theories are commonly applied to interpret these results, their validity and limitations have not been explored, raising concerns regarding the insights they offer.

To tackle this challenging question, a trinational team of researchers from Lund University (Sweden), the Russian Academy of Science (Russia) and the Technical University of Dresden (Germany) have developed a new methodology for the study of lead halide perovskites. This methodology is based on the complete mapping of the photoluminescence quantum yield and decay dynamics in the two-dimensional (2D) space of both fluence and frequency of the excitation light pulse. Such 2D maps not only offer a complete representation of the sample's photophysics, but also allow to examine the validity of theories, by applying a single set of theoretical equations and parameters to the entire data set. "Mapping a perovskite film using our new method is like taking its fingerprints - it provides us with a great deal of information about each individual sample." says Prof. Ivan Scheblykin, a Professor of Chemical Physics at Lund University. "Interestingly, each map resembles the shape of a horse's neck and mane, leading us to fondly refer to them as 'perovskite horses', which are all unique in their own way."

"The wealth of information contained in each 2D map allows us to explore different possible theories that may explain the complex behavior of charge carriers in metal halide perovskites" adds Dr. Pavel Frantsuzov from the Siberian Brunch of the Russian Academy of Science. Indeed, the researchers discovered that the two most commonly applied theories (the so called 'ABC theory' and the Shockley-Read-Hall theory) cannot explain the 2D maps across the entire range of excitation parameters. They propose a more advanced theory that includes additional nonlinear processes to explain the photophysics of metal halide perovskites.

The researchers show that their method has important implications for the development of more efficient perovskite solar cells. Prof. Dr. Yana Vaynzof, Chair for Emerging Electronic Technologies at the Institute for Applied Physics and Photonic Materials and the Center for Advancing Electronics Dresden (cfaed) explains: "By applying the new methodology to perovskite samples with modified interfaces, we were able to quantify their influence on the charge carrier dynamics in the perovskite layer by changing, for example, the density and efficacy of traps. This will allow us to develop interfacial modification procedures that will lead to optimal properties and more efficient photovoltaic devices."

Importantly, the new method is not limited to the study of metal halide perovskites and can be applied to any semiconducting material. "The versatility of our method and the ease with which we can apply it to new material systems is very exciting! We anticipate many new discoveries of fascinating photophysics in novel semiconductors." adds Prof. Scheblykin.

The work was now published in the prestigious journal Nature Communications.

Oncotarget published "Effect of cell microenvironment on the drug sensitivity of hepatocellular cancer cells" which reported that this study aimed to investigate whether Hepatocellular Cancer (HCC) cells cultured in more native conditions have an altered phenotype and drug sensitivity compared to those cultured in standard conditions.

Six HCC cell lines were cultured in "standard" or more "native" conditions.

HCC cells cultured in native conditions had slower doubling times, increased HK2 and GLUT, lower PHDA and ATP levels, and mutations in mitochondrial DNA.

From 90 comparisons of drug sensitivity, increased resistance and sensitivity for cells cultured in native conditions was observed in 14 and 8 comparisons respectively.

Therefore, cells cultured in more native conditions can have a more glycolytic and aggressive phenotype and varied drug sensitivity to those cultured in standard conditions, and may provide new insights to understanding tumor biology and drug development.

Dr. Richie Soong and Dr. Bhaskar Bhattacharya said, "Hepatocellular carcinoma (HCC) is the third leading cause of cancer related death worldwide, with a poor median survival time after diagnosis of six months."

"Hepatocellular carcinoma (HCC) is the third leading cause of cancer related death worldwide, with a poor median survival time after diagnosis of six months."

Lactate is also high in the HCC microenvironment, owing to the Warburg effect under which tumor cells convert glucose into lactate with an overall less production of Adenosine triphosphate (ATP) compared to mitochondrial oxidative phosphorylation.

As understanding of the tumor microenvironment has increased over time, it has become apparent that the historical culture conditions do not reflect the native tumor microenvironment well, including having increased glucose, increased oxygen, and reduced lactic acid concentrations.

The Oncotarget authors previously showed that gastric cancer cells cultured in low glucose compared to standard high glucose levels had an increased resistance to 5-fluorouracil and carboplatin - concomitant with increased glycolysis and mitochondrial mutation.

This led us to postulate that culturing cells in glucose, oxygen, and lactate conditions that are more consistent with the native tumor microenvironment may provide a better assessment of drug sensitivity, and help lead to higher success rates in drug development.

The goal of this Oncotarget study was to characterize the effect of culturing HCC cells in native compared to standard culture conditions on cellular and molecular phenotypes and drug sensitivity.

The Soong/Bhattacharya Research Team concluded in their Oncotarget Research Output, "we cultured HCC cells in its native environment compared to standard culture conditions, and observed phenotypic and molecular signatures of cells in native conditions pathologically similar to human HCC. The study highlights two important points,

A.New insights in pathogenesis could be gained by culturing cells in conditions closer to physiological conditions, and

B.A better idea of therapeutic response may be obtained by screening experimental agents in cancer cells grown in a microenvironment similar to actual disease setting.

Culturing cells in their native conditions has the potential to identify therapy, either single agent or combination, which would have otherwise been considered ineffective under the currently used artificial environment. Additionally, novel targets may also be identified under native conditions in addition to refining our existing knowledge of tumor biology. Nonetheless, it is important to keep in perspective the exhaustive work of manually concurrently culturing cells in standard and native conditions, and conducting the diverse interrogations, limited this study to understanding of a few cell lines, mechanisms and drugs. Future studies incorporating high-throughput culturing and screening methods, and the testing of other hypotheses and timepoints will help to reveal the scope of generalizability of these findings."

ADELPHI, Md. -- Army researchers developed a pioneering framework that provides a baseline for the development of collaborative multi-agent systems.

The framework is detailed in the survey paper Survey of recent multi-agent reinforcement learning algorithms utilizing centralized training, which is featured in the SPIE Digital Library. Researchers said the work will support research in reinforcement learning approaches for developing collaborative multi-agent systems such as teams of robots that could work side-by-side with future Soldiers.

"We propose that the underlying information sharing mechanism plays a critical role in centralized learning for multi-agent systems, but there is limited study of this phenomena within the research community," said Army researcher and computer scientist Dr. Piyush K. Sharma of the U.S. Army Combat Capabilities Development Command, known as DEVCOM, Army Research Laboratory. "We conducted this survey of the state-of-the-art in reinforcement learning algorithms and their information sharing paradigms as a basis for asking fundamental questions on centralized learning for multi-agent systems that would improve their ability to work together."

Sharma's collaborators on this project include DEVCOM ARL researchers Drs. Erin Zaroukian, Rolando Fernandez, Michael Dorothy, Derrik Asher and Anjon Basak, a postdoctoral fellow under the Oak Ridge Associated Universities fellowship program.

This survey of state-of-the-art in reinforcement learning establishes a baseline for researchers seeking to develop autonomous multi-agent systems through enhanced information sharing mechanisms, for example rewarding functions or observation and state space sharing.

Training multiple agents simultaneously is more difficult due to the dynamic nature of complex environments that can suffer from the curse of dimensionality, the more agents there are, the more complicated the coordination becomes, Sharma said. This paper develops a framework to characterize the key information sharing parameters that are often confusing and not easy to understand.

The researchers predict that centralization in training may be the solution towards more rapidly developing autonomous systems that can flexibly work alongside Soldiers in the future.

"Consistent, centralized training can result in multi-agent systems that work more reliably together, increasing the level of trust from the Soldier of the artificial intelligence," Sharma said. "Specifically, we focused on identifying and characterizing the underlying mathematical framework of the most recent centralized learning algorithms."

Such a mathematical model can provide an avenue to explore alternate centralized learning techniques to gauge their effect on the learning rate and emergent collaborative behaviors, he said.

The survey exceeds prior research literature in two ways:

Creates a consolidated view of the latest state-of-the-art in reinforcement learning algorithms
Outlines a novel approach to describing information shared during centralized learning

The researchers focused on the algorithms published within five to six years. As these algorithms are very recent, they have not yet been explored extensively by the research community. At the time of publication, they did not find comprehensive prior work.

The researchers attempted to define and categorize the mechanisms for sharing, orienting on what is actually being shared as opposed to how it is shared. They are optimistic that they have identified gaps in the recent reinforcement learning techniques worthy of further study that could potentially enhance the agent training process.

The researchers said they are optimistic the survey will generate discussion and further exploration of the problem space of machine learning to train autonomous multi-agent systems.

"As the demand for multi-agent systems working cooperatively has become more prevalent in the commercial industry, for example, Amazon Warehouse Robots, Intel's drone show at winter Olympics 2018. There is also an emerging need for these multi-agent system technologies to assist the Army in collaborative tactical operations," Sharma said. "The research resulting from this survey document can make the goal of reliable, collaborative AI achievable."

Moving forward, the team feels better equipped to investigate particular aspects of multi-agent reinforcement learning based approaches that train agents in centralized fashion.

Centralized techniques come with certain limitations, so they will also conduct empirical analysis of existing decentralized learning techniques, Sharma said. They plan to move towards modeling and simulation of multi-agent reinforcement learning training to validate and extend theories of agent learning, behavior and coordination.

WASHINGTON -- For more than 25 years, Burmese pythons have been living and breeding in the Florida Everglades where they prey on native wildlife and disrupt the region's delicate ecosystems. A new study shows that infrared cameras could make it easier to spot these invasive snakes in the Florida foliage, providing a new tool in the effort to remove them.

In the Optical Society (OSA) journal Applied Optics, researchers led by Dr. Kyle Renshaw from the University of Central Florida College of Optics and Photonics report that a near infrared camera helped people detect Burmese pythons at distances up to 1.3 times farther away than was possible using a traditional visible-wavelength camera. Because infrared sensors are small and low cost, they could easily be incorporated into handheld or vehicle-mounted systems designed for seeking out pythons.

"The removal of Burmese Pythons is vital to preventing further damage to the Floridian ecosystem and preventing their spread to other regions," said Hewitt, a PhD student and lead author on the study. "Our study -- one of the first to examine the efficacy of near infrared sensing in locating these pythons -- can help inform methods used to remove them from the environment."

Making snakes stand out

Burmese pythons can be up to 20 feet long and weigh as much as 200 pounds. They arrived in the U.S. as exotic pets in the 1980s and the snakes proliferated in the Everglades after a breeding facility was destroyed during Hurricane Andrew in 1992. Their natural camouflage makes them blend in with grass and foliage, making them hard to see with the human eye or a traditional visible-light camera. In a previous study, the authors measured the reflectivity spectra of Burmese pythons in the visible and infrared wavelengths, finding that pythons are more visible against the background at infrared wavelengths longer than 750 nm.

"Based on these earlier findings, we hypothesized that using near infrared wavelengths for imaging could make the pythons easier to see because they would appear dark against bright foliage," said Hewitt. "Although we haven't acquired reflectivity measurements from other species of snakes, the pythons should be easy to distinguish since they are larger than any other native species of snake."

To test their hypothesis, the researchers took images of Burmese pythons in grass using visible and infrared cameras with similar fields of view and resolution. They then asked volunteers to examine these images and indicate whether they saw a python. Based on the responses of the volunteers, the researchers calculated the advantage of using near infrared images compared to visible.

"The method we used to evaluate each of the sensors was originally established for military sensing applications," Hewitt explained. "It accounts for the attributes of human vision and perception in addition to the characteristics of the system components to determine how effective a system is at allowing the observer to accomplish a task."

Spotting pythons day or night

Although other studies have explored using thermal infrared sensors to find Burmese pythons, the snakes had to have been basking in the sun during the day for them to be detected at night. The thermal contrast against their environment also diminished over time.

"In this work, we don't rely on thermal contrast," said Hewitt. "We found that near infrared imaging can be used both during the day as well as at night with illumination to improve detection, even if the pythons have not been basking."

The researchers have contracted with the Florida Fish and Wildlife Conservation Commission (FWC) to work on a project that expands on these results. "We are evaluating whether or not this technology will be effective in the field, and, if so, how to make it field-ready in the challenging Florida everglades ecosystem," said McKayla Spencer, the FWC interagency python management coordinator. "We are just in the beginning stages of our project with the researchers."

"The big mystery about plankton is what controls its distribution and abundance, and what conditions lead to big plankton blooms," said Dennis McGillicuddy, Senior Scientist and Department Chair in Applied Ocean Physics and Engineering at the Woods Hole Oceanographic Institution (WHOI).

Two new papers explore this question and provide examples of conditions that lead to massive plankton blooms with vastly different potential impacts on the ecosystem, according to McGillicuddy, co-author of both papers. Both papers also point to the importance of using advanced technology--including Video Plankton Recorders, autonomous underwater vehicles, and the Ocean Observatories Initiative's Coastal Pioneer Array--to find and monitor these blooms.

In one paper, Diatom Hotspots Driven by Western Boundary Current Instability, published in Geophysical Research Letters (GRL), scientists found unexpectedly productive subsurface hotspot blooms of diatom phytoplankton.

In the GRL paper, researchers investigated the dynamics controlling primary productivity in a region of the Mid-Atlantic Bight (MAB), one of the world's most productive marine ecosystems. In 2019, they observed unexpected diatom hotspots in the slope region of the bight's euphotic zone, the ocean layer that receives enough light for photosynthesis to occur. Phytoplankton are photosynthetic microorganisms that are the foundation of the aquatic food web.

It was surprising to the researchers that the hotspots occurred in high-salinity water intruding from the Gulf Stream. "While these intrusions of low?nutrient Gulf Stream water have been thought to potentially diminish biological productivity, we present evidence of an unexpectedly productive subsurface diatom bloom resulting from the direct intrusion of a Gulf Stream meander towards the continental shelf," the authors note. They hypothesize that the hotspots were not fueled by Gulf Stream surface water, which is typically low in nutrients and chlorophyll, but rather that the hotspots were fueled by nutrients upwelled into the sunlight zone from deeper Gulf Stream water.

With changing stability of the Gulf Stream, intrusions from the Gulf Stream had become more frequent in recent decades, according to the researchers. "These results suggest that changing large?scale circulation has consequences for regional productivity that are not detectable by satellites by virtue of their occurrence well below the surface," the authors note.

"In this particular case, changing climate has led to an increase in productivity in this particular region, by virtue of a subtle and somewhat unexpected interaction between the physics and biology of the ocean. That same dynamic may not necessarily hold elsewhere in the ocean, and it's quite likely that other areas of the ocean will become less productive over time. That's of great concern," said McGillicuddy. "There are going to be regional differences in the way the ocean responds to climate change. And society needs to be able to intelligently manage from a regional perspective, not just on a global perspective."

The research finding demonstrated "a cool, counterintuitive biological impact of this changing large scale circulation," said the GRL paper's lead author, Hilde Oliver, a postdoctoral scholar in Applied Ocean Physics and Engineering at WHOI. She recalled watching the instrument data come in. With typical summertime values of about 1-1.5 micrograms of chlorophyll per liter of seawater, researchers recorded "unheard of concentrations for chlorophyll in this region in summer," as high as 12 or 13 micrograms per liter, Oliver said.

Oliver, whose Ph.D. focused on modeling, said the cruise helped her to look at phytoplankton blooms from more than a theoretical sense. "To go out into the ocean and see how the physics of the ocean can manifest these blooms in the real world was eye opening to me," she said.

Another paper, A Regional, Early Spring Bloom of Phaeocystis pouchetii on the New England Continental Shelf, published in the Journal of Geophysical Research: Oceans (JGR: Oceans), also was eye opening. Researchers investigating the biological dynamics of the New England continental shelf in 2018 discovered a huge bloom of the haptophyte phytoplankton Phaeocystis pouchetii.

However, unlike the diatom hotspots described in the GRL paper, Phaeocystis is "unpalatable to a lot of different organisms and disrupts the entire food web," said Walker Smith, retired professor at the Virginia Institute of Marine Science William and Mary, who is the lead author on the JGR: Oceans paper. The phytoplankton form gelatinous colonies that are millimeters in diameter.

When Phaeocystis blooms, it utilizes nutrients just like any other form of phytoplankton would. However, unlike the diatoms noted in the GRL paper, Phaeocystis converts biomass into something that doesn't tend to get passed up the rest of the food chain, said McGillicuddy.

"Understanding the physical-biological interactions in the coastal system provides a basis for predicting these blooms of potentially harmful algae and may lead to a better prediction of their impacts on coastal systems," the authors stated.

Massive blooms of the colonial stage of this and similar species have been reported in many systems in different parts of the world, which Smith has studied. These types of blooms probably occur about every three years in the New England continental shelf and probably have a fairly strong impact on New England waters, food webs, and fisheries, said Smith. Coastal managers need to know about these blooms because they can have economic impacts on aquaculture in coastal areas, he said.

"Despite the fact that the Mid-Atlantic Bight has been well-studied and extensively sampled, there are things that are going on that we still don't really appreciate," said Smith. "One example are these Phaeocystis blooms that are deep in the water and that you are never going to see unless you are there because satellites can't show them. So, the more we look, the more we find out."

Key Takeaways:

Study finds non-tiered customer loyalty programs create a more sustainable customer base.

Non-tiered customer loyalty programs are not as likely to generate increases in spending per transaction or accelerate transactions.

CATONSVILLE, MD, June 7, 2021 - Customer loyalty programs have been around for decades and are used to help businesses, marketers and sellers build a sustainable relationship with their customers. But do they work? A recent study sought to find out and researchers learned that while yes, customer loyalty programs do work, perhaps not in ways most may assume.

There are two basic types of customer loyalty programs, tiered and non-tiered. Airlines and hotels often use tiered customer loyalty programs that increase rewards as program members reach higher thresholds of spending over time. Retailers and service industry businesses are more likely to offer non-tiered customer loyalty programs, in which members are rewarded with frequent, but not increasing rewards, such as "buy 10 get one free."

This research investigated if those non-tiered customer loyalty programs actually do what they are designed to do.

The study to be published in the June issue of the INFORMS journal Marketing Science, "Can Non-tiered Customer Loyalty Programs Be Profitable?", is authored by Arun Gopalakrishnan of Rice University, Zhenling Jiang of the Wharton School of Business at the University of Pennsylvania, and Yulia Nevskaya and Raphael Thomadsen of the Olin Business School at Washington University in St. Louis.

The authors found that non-tiered customer loyalty programs increase customer value by almost 30% over a five-year time period. They discovered that the program's effectiveness is not so much through increased spending per transaction or frequency of purchasing but rather through the reduction of attrition. In other words, the chief benefit is that the customer loyalty program reduces customer fall-off and turnover.

"We found that a non-tiered customer loyalty program's reduction in attrition accounts for more than 80% of the program's total lift or success," said Thomadsen. "On the other hand, increased frequency accounts for less than 20% of the program's lift or effectiveness."

Jiang added, "One of the more interesting findings was that the impact of the loyalty program does not necessarily contribute to increased spending per transaction or increased frequency of transactions. Rather, the benefit to the business is creating more sustainable and lasting relationships with customers."

To conduct their research, the authors worked with a company to collect data of more than 5,500 new customers who first started purchasing from that company in the same three-month period. This helped to ensure that the customers were comparable in terms of the amount of time they had to become acquainted with the selling firm. For the next 30 months, the researchers collected all subsequent transaction data from those consumers. During that period, a non-tiered customer loyalty program was introduced.

In the process, some of these new customers were automatically enrolled into the loyalty program. This helped researchers better gauge pre-program visit frequency and spending and then compare it to post-enrollment visit frequency and spending. "We were able to analyze the behaviors of consumers absent a customer loyalty program, and then after the rollout of the program," said Nevskaya. "We evaluated frequency and actual spending amounts, and whether customers come back for repeat transactions."

Gopalakrishnan summarized, "In the end, the primary value of a non-tiered customer loyalty program is not a means to increase frequency or spending. It's a way to nurture a long-term and lasting relationship with the customer to reduce the defection of loyal customers over time. Non-tiered loyalty programs may provide psychological benefits that help cultivate such loyalty."

Artificial intelligence natural language computer applications are becoming increasingly sophisticated, raising the possibility that they could assume a greater role in health care, including interacting with patients. But before these applications enter the clinic, their potential and pitfalls need thoughtful exploration, states a new article in NPJ Digital Medicine.

The authors are Diane M. Korngiebel, a Hastings Center research scholar, and Sean D. Mooney, chief research information officer at University of Washington Medicine.

"There is compelling promise and serious hype in AI applications that generate natural language" Korngiebel and Mooney write, referring to OpenAI's Generative Pre-trained Transformer 3 (GPT-3) and similar technologies. The article breaks down potential health care applications into three categories: unrealistic, realistic and feasible, and realistic but challenging.

Unrealistic Applications

Natural language AI applications will not replace doctors, nurses, and other health care personnel in conversation with patients anytime soon. "Interactions with GPT-3 that look (or sound) like interactions with a living, breathing--and empathetic or sympathetic--human being are not," the authors write. In a recent test of GTP-3 for mental health counseling, for example, the application supported a simulated patient's expressed thoughts of suicide. In addition, natural language AI applications currently reflect human biases involving gender, race, and religion.

Realistic and Feasible Applications

Natural language applications could relieve health care providers of some routine tedious tasks, such as navigating complex electronic health records. And, given that they are capable of fairly natural-sounding question and answer exchanges, the applications could improve customer service online chat support and help patients with noncritical tasks such as setting up equipment in preparation for a telehealth visit. But there must be "serious guardrails" for all health care interactions, including training the applications to eliminate "harmful, prejudicial, or inappropriate vocabulary."

Realistic but Challenging Applications

GTP-3 could be used to assist with triaging noncritical patients who come to emergency departments. However, developers of the technology and people implementing it would need to be mindful of harms. For example, natural language applications that do not "speak" a patient's language might triage that patient inappropriately. "Implementation should include another means of triaging those patients who cannot, or do not wish to, use the conversational agent, which may also be too linguistically homogenous to offer culturally mindful language use," the authors write, adding that it is important to maintain a "human in the loop." A staff member would also need to review all triage forms.

The article concludes with recommendations for making sure that natural language applications are equitable. A broad range of stakeholders should be involved from the earliest stage of development through deployment and evaluation. And there should be transparency, including in the datasets used and limitations of the applications.

"We should have cautious optimism for the potential applications of sophisticated natural language processing applications to improve patient care," the authors write. "The future is coming. Rather than fear it, we should prepare for it--and prepare to benefit humanity using these applications."

LEBANON, NH - Previous work by a team of researchers led by Steven N. Fiering, PhD, Immunology and Cancer Immunotherapy researcher at Dartmouth's and Dartmouth-Hitchcock's Norris Cotton Cancer Center and Nicole Steinmetz, PhD, Jacobs School of Engineering and Moores Cancer Center, University of California San Diego, showed that a plant virus that does not infect mammals, cowpea mosaic plant virus (CPMV), when injected into cancerous tumors, strongly stimulated the immune system to attack and often eliminate the tumor. However, very little was understood about immune recognition of plant viruses and how and why CPMV is exceptionally immuno-stimulating. In a new study, the team identifies just how CPMV is recognized by the immune system, opening the door for CPMV to be pursued as a new biological drug for treatment of cancer.

CPMV is recognized by the immune system as a pathogen--any infectious agent that can cause disease--through a family of receptors on immune cells called toll-like receptors. Toll-like receptors recognize molecules that signal the invasion of a pathogen and send a warning signal to the immune cells to mobilize to attack the pathogen. When tumors are injected with CPMV, the immune system activates and attacks the tumors by way of this pathogen pattern recognition. "The recognition of CPMV by toll-like receptors illustrates how these receptors are quite flexible and recognize many more molecular patterns than immunologists previously knew," says Fiering.

During the immune stimulation process, the immune cells release proteins that signal and activate other immune cells, known as cytokines. The team's study, "Cowpea mosaic virus stimulates antitumor immunity through recognition by multiple MYD88-dependent toll-like receptors," newly published in Biomaterials, identifies the three toll-like receptors that recognize CPMV. The paper also highlights the importance of a particular cytokine, "interferon alpha," for strong anti-tumor impact when used as an in situ vaccine to treat cancer.

In situ vaccination, in which tumors are directly treated with immune stimulating reagents, have powerful potential to improve cancer immunotherapy in a safe and inexpensive manner. "In situ vaccination has made contributions already to cancer treatment. CPMV is an excellent reagent that may soon be used to help patients in the same manner," says Fiering. "The in situ vaccination treatment of a tumor by CPMV can stimulate the immune system to also attack distant metastatic tumors that have not been treated."

Extreme risk protection orders (ERPOs), also known as gun violence restraining orders, are civil court orders that grant temporary restrictions on purchasing and possessing firearms for individuals determined by a civil court judge to be at extreme risk of committing violence against themselves or others. A new study examined ERPO use in Oregon in the first 15 months after it was adopted. The study found that while ERPOs are commonly considered as a tool to remove guns from dangerous individuals, they should also be considered as a tool to prevent gun purchases by dangerous individuals.

The study was conducted by researchers at Michigan State University (MSU), Columbia University, the University of Michigan, and Johns Hopkins University. It appears in Criminology & Public Policy, a publication of the American Society of Criminology.

"The findings from our study can help us better understand how these laws are being used, including what types of behaviors are prompting others to file an ERPO petition," explains April Zeoli, associate professor of criminal justice at MSU, who led the study.

As of 2020, 19 states and the District of Columbia had passed laws establishing ERPOs. Due to the newness of these laws, there is little research on the circumstances under which ERPO petitions are filed and the individuals for whom they are filed. Oregon enacted its ERPO law at the start of January 2018. Under the process, a law enforcement officer or family or household member may file a petition, and a hearing is held within one judicial business day. The petitioner has the burden of proving to the civil court judge that the respondent is at high risk for injury to themselves via suicide attempt or to others.

This study examined 93 petitions for ERPO cases through March 2019; petitions were accessed through public records requests.

At least one ERPO petition was filed in 22 of Oregon's 36 counties during the study period. Most ERPO respondents were reported by petitioners to have histories of interpersonal violence (75 percent) or suicidality (73 percent), and more than half of those were reported to have threatened suicide by using a gun. More than half of ERPO respondents reportedly made both suicide threats or attempts, and threats or use of violence against others, which is higher than in other states, the authors note.

In 56 percent of ERPO petitions, the petitioner specifically referred to the respondent as having a mental illness or mental health concern, despite the fact that Oregon law does not list mental health as a factor to be considered in ERPO petitions. Petitions filed by law enforcement were more likely to report mental illness or mental health concerns than petitions filed by others.

Also, more than half of death threats, suicide threats, or suicide attempts with known timing occurred within one week of the petition being filed, the study found, suggesting that the petitions are being used in times of immediate crisis.

The study also found that 74 percent of petitioners reported that respondents had a gun at the time of the filing, and 49 percent said respondents had recently acquired or tried to acquire a deadly weapon, 96 percent of which were firearms.

In 26 percent of cases, the petitioner did not explicitly indicate that the respondent currently had a gun. Some petitioners said they filed the ERPO petition to prevent a respondent without a gun from acquiring one out of concern about the increased risk of harm that would pose. This use of ERPO may be overlooked by policymakers and other stakeholders because ERPOs are more commonly thought of as a tool to remove guns from dangerous individuals than as a tool to prevent gun purchases by dangerous individuals, the authors suggest.

Most petitions (65 percent) in Oregon were filed by law enforcement, which is lower than in other states where non-law-enforcement individuals can file ERPOs, according to the authors. Petitions filed by law enforcement were more likely to be granted than petitions filed by family or household members.

The study concluded that ERPO petitions and orders are overwhelmingly being used as intended, that is, for cases of imminent risk of harm to self or others. Yet it is possible they could be used more.

"The number of ERPO petitions in Oregon and the number of counties without a single petition in the first 15 months of the law suggest that ERPOs may be an underused tool," suggests Jennifer Paruk, a doctoral student in criminal justice at MSU, who coauthored the study. "Greater dissemination of public information about ERPOs could increase their appropriate use so high-risk individuals and their families could benefit, especially when dangerous individuals are prevented from purchasing guns."

The authors clarify that their work, in characterizing and describing information in ERPO petitions, should not be viewed as a systematic measurement of characteristics of ERPO respondents. The study, they note, was limited by their reliance on petitions completed by individuals who may or may not know respondents' full histories. In addition, sometimes language used by petitioners was imprecise, leaving the researchers to estimate meanings. Finally, the study provides a view of ERPOs in one state and so its findings should not be generalized to other states.

The threat of antibiotic resistance rises as bacteria continue to evolve to foil even the most powerful modern drug treatments. By 2050, antibiotic resistant-bacteria threaten to claim more than 10 million lives as existing therapies prove ineffective.

Bacteriophage, or "phage," have become a new source of hope against growing antibiotic resistance. Ignored for decades by western science, phages have become the subject of increasing research attention due to their capability to infect and kill bacterial threats.

A new project led by University of California San Diego Biological Sciences graduate student Joshua Borin, a member of Associate Professor Justin Meyer's laboratory, has provided evidence that phages that undergo special evolutionary training increase their capacity to subdue bacteria. Like a boxer in training ahead of a title bout, pre-trained phages demonstrated they could delay the onset of bacterial resistance.

The study, which included contributions from researchers at the University of Haifa in Israel and the University of Texas at Austin, is published June 8 in the Proceedings of the National Academy of Sciences.

"Antibiotic resistance is inherently an evolutionary problem, so this paper describes a possible new solution as we run out of antibiotic drug options," said Borin. "Using bacterial viruses that can adapt and evolve to the host bacteria that we want them to infect and kill is an old idea that is being revived. It's the idea of the enemy of our enemy is our friend."

The idea of using phages to combat bacterial infections goes back to the days prior to World War II. But as antibiotic drugs became the leading treatment for bacterial infections, phage research for therapeutic potential was largely forgotten. That mindset has changed in recent years as deadly bacteria continue to evolve to render many modern drugs ineffective.

Borin's project was designed to train specialized phage to fight bacteria before they encounter their ultimate bacterial target. The study, conducted in laboratory flasks, demonstrated classic evolutionary and adaptational mechanisms at play. The bacteria, Meyer said, predictably moved to counter the phage attack. The difference was in preparation. Phages trained for 28 days, the study showed, were able to suppress bacteria 1,000 times more effectively and three- to eight-times longer than untrained phage.

"The trained phage had already experienced ways that the bacteria would try to dodge it," said Meyer. "It had 'learned' in a genetic sense. It had already evolved mutations to help it counteract those moves that the bacteria were taking. We are using phage's own improvement algorithm, evolution by natural selection, to regain its therapeutic potential and solve the problem of bacteria evolving resistance to yet another therapy."

The researchers are now extending their findings to research how pre-trained phages perform on bacteria important in clinical settings, such as E. coli. They are also working to evaluate how well training methods work in animal models.

UC San Diego is a leader in phage research and clinical applications. In 2018 the university's School of Medicine established the Center for Innovative Phage Applications and Therapeutics, the first dedicated phage therapy center in North America.

"We have prioritized antibiotics since they were developed and now that they are becoming less and less useful people are looking back to phage to use as therapeutics," said Meyer. "More of us are looking into actually running the experiments necessary to understand the types of procedures and processes that can improve phage therapeutics."