Using Webb’s NIRSpec (Near-Infrared Spectrograph), astronomers confirmed a temperature difference between the eternal morning and eternal evening on WASP-39 b, with the evening appearing hotter by roughly 300 Fahrenheit degrees (about 200 Celsius degrees). They also found evidence for different cloud cover, with the forever morning portion of the planet being likely cloudier than the evening.

Astronomers analyzed the 2- to 5-micron transmission spectrum of WASP-39 b, a technique that studies the exoplanet’s terminator, the boundary that separates the planet’s dayside and nightside. A transmission spectrum is made by comparing starlight filtered through a planet’s atmosphere as it moves in front of the star, to the unfiltered starlight detected when the planet is beside the star. When making that comparison, researchers can get information about the temperature, composition, and other properties of the planet’s atmosphere.

“WASP-39 b has become a sort of benchmark planet in studying the atmosphere of exoplanets with Webb,” said Néstor Espinoza, an exoplanet researcher at the Space Telescope Science Institute and lead author on the study. “It has an inflated, puffy atmosphere, so the signal coming from starlight filtered through the planet’s atmosphere is quite strong.”

Previously published Webb spectra of WASP-39b’s atmosphere, which revealed the presence of carbon dioxide, sulfur dioxide, water vapor, and sodium, represent the entire day/night boundary — there was no detailed attempt to differentiate between one side and the other.

Now, the new analysis builds two different spectra from the terminator region, essentially splitting the day/night boundary into two semicircles, one from the evening, and the other from the morning. Data reveals the evening as significantly hotter, a searing 1,450 degrees Fahrenheit (800 degrees Celsius), and the morning a relatively cooler 1,150 degrees Fahrenheit (600 degrees Celsius).

“It’s really stunning that we are able to parse this small difference out, and it’s only possible due Webb’s sensitivity across near-infrared wavelengths and its extremely stable photometric sensors,” said Espinoza. “Any tiny movement in the instrument or with the observatory while collecting data would have severely limited our ability to make this detection. It must be extraordinarily precise, and Webb is just that.”

Extensive modeling of the data obtained also allows researchers to investigate the structure of WASP-39 b’s atmosphere, the cloud cover, and why the evening is hotter. While future work by the team will study how the cloud cover may affect temperature, and vice versa, astronomers confirmed gas circulation around the planet as the main culprit of the temperature difference on WASP-39 b.

On a highly irradiated exoplanet like WASP-39 b that orbits relatively close to its star, researchers generally expect the gas to be moving as the planet rotates around its star: Hotter gas from the dayside should move through the evening to the nightside via a powerful equatorial jet stream. Since the temperature difference is so extreme, the air pressure difference would also be significant, which in turn would cause high wind speeds.

Using General Circulation Models, 3-dimensional models similar to the ones used to predict weather patterns on Earth, researchers found that on WASP-39 b the prevailing winds are likely moving from the night side across the morning terminator, around the dayside, across the evening terminator and then around the nightside. As a result, the morning side of the terminator is cooler than the evening side. In other words, the morning side gets slammed with winds of air that have been cooled on the nightside, while the evening is hit by winds of air heated on the dayside. Research suggests the wind speeds on WASP-39 b can reach thousands of miles an hour!

“This analysis is also particularly interesting because you’re getting 3D information on the planet that you weren’t getting before,” added Espinoza. “Because we can tell that the evening edge is hotter, that means it’s a little puffier. So, theoretically, there is a small swell at the terminator approaching the nightside of the planet.”

The team’s results have been published in Nature.

The researchers will now look to use the same method of analysis to study atmospheric differences of other tidally locked hot Jupiters, as part of Webb Cycle 2 General Observers Program 3969.

WASP-39 b was among the first targets analyzed by Webb as it began regular science operations in 2022. The data in this study was collected under Early Release Science program 1366, designed to help scientists quickly learn how to use the telescope’s instruments and realize its full science potential.

The team say this new approach could reduce the need for invasive and costly diagnostic tests while improving treatment outcomes early when interventions such as lifestyle changes or new medicines may have a chance to work best.

Dementia poses a significant global healthcare challenge, affecting over 55 million people worldwide at an estimated annual cost of $820 billion. The number of cases is expected to almost treble over the next 50 years.

The main cause of dementia is Alzheimer’s disease, which accounts for 60-80% of cases. Early detection is crucial as this is when treatments are likely to be most effective, yet early dementia diagnosis and prognosis may not be accurate without the use of invasive or expensive tests such as positron emission tomography (PET) scans or lumbar puncture, which are not available in all memory clinics. As a result, up to a third of patients may be misdiagnosed and others diagnosed too late for treatment to be effective.

A team led by scientists from the Department of Psychology at the University of Cambridge has developed a machine learning model able to predict whether and how fast an individual with mild memory and thinking problems will progress to developing Alzheimer’s disease. In research published today in eClinical Medicine, they show that it is more accurate than current clinical diagnostic tools.

To build their model, the researchers used routinely-collected, non-invasive, and low-cost patient data — cognitive tests and structural MRI scans showing grey matter atrophy — from over 400 individuals who were part of a research cohort in the USA.

They then tested the model using real-world patient data from a further 600 participants from the US cohort and — importantly — longitudinal data from 900 people from memory clinics in the UK and Singapore.

The algorithm was able to distinguish between people with stable mild cognitive impairment and those who progressed to Alzheimer’s disease within a three-year period. It was able to correctly identify individuals who went on to develop Alzheimer’s in 82% of cases and correctly identify those who didn’t in 81% of cases from cognitive tests and an MRI scan alone.

The algorithm was around three times more accurate at predicting the progression to Alzheimer’s than the current standard of care; that is, standard clinical markers (such as grey matter atrophy or cognitive scores) or clinical diagnosis. This shows that the model could significantly reduce misdiagnosis.

The model also allowed the researchers to stratify people with Alzheimer’s disease using data from each person’s first visit at the memory clinic into three groups: those whose symptoms would remain stable (around 50% of participants), those who would progress to Alzheimer’s slowly (around 35%) and those who would progress more rapidly (the remaining 15%). These predictions were validated when looking at follow-up data over 6 years. This is important as it could help identify those people at an early enough stage that they may benefit from new treatments, while also identifying those people who need close monitoring as their condition is likely to deteriorate rapidly.

Importantly, those 50% of people who have symptoms such as memory loss but remain stable, would be better directed to a different clinical pathway as their symptoms may be due to other causes rather than dementia, such as anxiety or depression.

Senior author Professor Zoe Kourtzi from the Department of Psychology at the University of Cambridge said: “We’ve created a tool which, despite using only data from cognitive tests and MRI scans, is much more sensitive than current approaches at predicting whether someone will progress from mild symptoms to Alzheimer’s — and if so, whether this progress will be fast or slow.

“This has the potential to significantly improve patient wellbeing, showing us which people need closest care, while removing the anxiety for those patients we predict will remain stable. At a time of intense pressure on healthcare resources, this will also help remove the need for unnecessary invasive and costly diagnostic tests.”

While the researchers tested the algorithm on data from a research cohort, it was validated using independent data that included almost 900 individuals who attended memory clinics in the UK and Singapore. In the UK, patients were recruited through the Quantiative MRI in NHS Memory Clinics Study (QMIN-MC) led by study co-author Dr Timothy Rittman at Cambridge University Hospitals NHS Trust and Cambridgeshire and Peterborough NHS Foundation Trusts (CPFT).

The researchers say this shows it should be applicable in a real-world patient, clinical setting.

Dr Ben Underwood, Honorary Consultant Psychiatrist at CPFT and assistant professor at the Department of Psychiatry, University of Cambridge, said: “Memory problems are common as we get older. In clinic I see how uncertainty about whether these might be the first signs of dementia can cause a lot of worry for people and their families, as well as being frustrating for doctors who would much prefer to give definitive answers. The fact that we might be able to reduce this uncertainty with information we already have is exciting and is likely to become even more important as new treatments emerge.”

Professor Kourtzi said: “AI models are only as good as the data they are trained on. To make sure ours has the potential to be adopted in a healthcare setting, we trained and tested it on routinely-collected data not just from research cohorts, but from patients in actual memory clinics. This shows it will be generalisable to a real-world setting.”

The team now hope to extend their model to other forms of dementia, such as vascular dementia and frontotemporal dementia, and using different types of data, such as markers from blood tests.

Professor Kourtzi added: “If we’re going to tackle the growing health challenge presented by dementia, we will need better tools for identifying and intervening at the earliest possible stage. Our vision is to scale up our AI tool to help clinicians assign the right person at the right time to the right diagnostic and treatment pathway. Our tool can help match the right patients to clinical trials, accelerating new drug discovery for disease modifying treatments.”

The study was funded by Wellcome, the Royal Society, Alzheimer’s Research UK, the Alzheimer’s Drug Discovery Foundation Diagnostics Accelerator, the Alan Turing Institute, and the National Institute for Health Research Cambridge Biomedical Research Centre.

Led by Natashya Falcone and co-first authors Tess Grett Mathes and Mahsa Monirizad, the research team delved into the realm of self-assembling peptide-based hydrogels, renowned for their versatility in mimicking extracellular matrices (ECMs) of diverse microenvironments.

AD presents an intricate challenge in neurodegenerative research. Traditional two-dimensional (2D) models have limitations in capturing the complexity of the disease. Through their innovative approach, the team developed a multi-component hydrogel scaffold, named Col-HAMA-FF, designed to mimic the amyloid-beta (β) containing microenvironment associated with AD.

The study’s findings, published in a recent issue of Acta Biomaterialia, illuminate the formation of β-sheet structures within the hydrogel matrix, mimicking the nanostructures of amyloid-β proteins. By culturing healthy neuronal progenitor cells (NPCs) within this amyloid-mimicking environment and comparing results to those in a natural-mimicking matrix, the researchers observed elevated levels of neuroinflammation and apoptosis markers. This suggests a significant impact of amyloid-like structures on NPC phenotypes and behaviors.

Dr. Ali Khademhosseini, the study’s corresponding author, expressed excitement about the implications of their findings: “This foundational work provides a promising scaffold for future investigations into AD mechanisms and drug testing. By bridging the gap between 3D hydrogel models and the complex reality of AD pathological nanostructures, we aim to understand this interaction on healthy neuronal cells so that we can accelerate the development of effective therapeutic strategies.”

The study represents a crucial step towards unraveling the mysteries of the b-amyloid-like environment which can be found in AD and marks a milestone in the quest for innovative solutions to combat neurodegenerative disorders.

“Since President Biden and Vice President Harris unveiled the first image from the James Webb Space Telescope two years ago, Webb has continued to unlock the mysteries of the universe,” said NASA Administrator Bill Nelson. “With remarkable images from the corners of the cosmos, going back nearly to the beginning of time, Webb’s capabilities are shedding new light on our celestial surroundings and inspiring future generations of scientists, astronomers, and explorers.”

Also take a moment to scan the background. Webb’s image is overflowing with distant galaxies. Some take spiral and oval shapes, like those threaded throughout the Penguin’s “tail feathers,” while others scattered throughout are shapeless dots. This is a testament to the sensitivity and resolution of the telescope’s infrared instruments. (Compare Webb’s view to the 2018 observation that combines infrared light from NASA’s retired Spitzer Space Telescope and near-infrared and visible light from NASA’s Hubble Space Telescope.) Even though these observations only took a few hours, Webb revealed far more distant, redder, and dustier galaxies than previous telescopes — one more reason to expect Webb to continue to expand our understanding of everything in the universe.

Want more? Take a tour to the image, “fly through” it in a visualization, and compare Webb’s image to the Hubble Space Telescope’s.

Arp 142 lies 326 million light-years from Earth in the constellation Hydra.

“In just two years, Webb has transformed our view of the universe, enabling the kind of world-class science that drove NASA to make this mission a reality,” said Mark Clampin, director of the Astrophysics Division at NASA Headquarters in Washington. “Webb is providing insights into longstanding mysteries about the early universe and ushering in a new era of studying distant worlds, while returning images that inspire people around the world and posing exciting new questions to answer. It has never been more possible to explore every facet of the universe.”

The telescope’s specialization in capturing infrared light — which is beyond what our own eyes can detect — shows these galaxies, collectively known as Arp 142, locked in a slow cosmic dance. Webb’s observations, which combine near- and mid-infrared light from Webb’s NIRCam (Near-Infrared Camera) and MIRI (Mid-Infrared Instrument), respectively, clearly show that they are joined by a haze represented in blue that is a mix of stars and gas, a result of their mingling.

Their ongoing interaction was set in motion between 25 and 75 million years ago, when the Penguin (individually cataloged as NGC 2936) and the Egg (NGC 2937) completed their first pass. They will go on to shimmy and sway, completing several additional loops before merging into a single galaxy hundreds of millions of years from now.

Let’s Dance!

Before their first approach, the Penguin held the shape of a spiral. Today, its galactic center gleams like an eye, its unwound arms now shaping a beak, head, backbone, and fanned-out tail.

Like all spiral galaxies, the Penguin is still very rich in gas and dust. The galaxies’ “dance” gravitationally pulled on the Penguin’s thinner areas of gas and dust, causing them to crash in waves and form stars. Look for those areas in two places: what looks like a fish in its “beak” and the “feathers” in its “tail.”

Surrounding these newer stars is smoke-like material that includes carbon-containing molecules, known as polycyclic aromatic hydrocarbons, which Webb is exceptional at detecting. Dust, seen as fainter, deeper orange arcs also swoops from its beak to tail feathers.

In contrast, the Egg’s compact shape remains largely unchanged. As an elliptical galaxy, it is filled with aging stars, and has a lot less gas and dust that can be pulled away to form new stars. If both were spiral galaxies, each would end the first “twist” with new star formation and twirling curls, known as tidal tails.

Another reason for the Egg’s undisturbed appearance: These galaxies have approximately the same mass or heft, which is why the smaller-looking elliptical wasn’t consumed or distorted by the Penguin.

It is estimated that the Penguin and the Egg are about 100,000 light-years apart — quite close in astronomical terms. For context, the Milky Way galaxy and our nearest neighbor, the Andromeda Galaxy, are about 2.5 million light-years apart. They too will interact, but not for about 4 billion years.

Now, look to the top right to spot a galaxy that is not at this party. This edge-on galaxy, cataloged PGC 1237172, is 100 million light-years closer to Earth. It’s also quite young, teeming with new, blue stars.

Want one more party trick? Switch to Webb’s mid-infrared-only image to see PGC 1237172 practically disappear. Mid-infrared light largely captures cooler, older stars and an incredible amount of dust. Since the galaxy’s stellar population is so young, it “vanishes” in mid-infrared light.

But a model’s uncertainty quantifications are only useful if they are accurate. If a model says it is 49% confident that a medical image shows a pleural effusion, then 49% of the time, the model should be right.

MIT researchers have introduced a new approach that can improve uncertainty estimates in machine-learning models. Their method not only generates more accurate uncertainty estimates than other techniques, but does so more efficiently.

In addition, because the technique is scalable, it can be applied to huge deep-learning models that are increasingly being deployed in health care and other safety-critical situations.

This technique could give end users, many of whom lack machine-learning expertise, better information they can use to determine whether to trust a model’s predictions or if the model should be deployed for a particular task.

“It is easy to see these models perform really well in scenarios where they are very good, and then assume they will be just as good in other scenarios. This makes it especially important to push this kind of work that seeks to better calibrate the uncertainty of these models to make sure they align with human notions of uncertainty,” says lead author Nathan Ng, a graduate student at the University of Toronto who is a visiting student at MIT.

Ng wrote the paper with Roger Grosse, an assistant professor of computer science at the University of Toronto; and senior author Marzyeh Ghassemi, an associate professor in the Department of Electrical Engineering and Computer Science and a member of the Institute of Medical Engineering Sciences and the Laboratory for Information and Decision Systems. The research will be presented at the International Conference on Machine Learning.

Quantifying uncertainty

Uncertainty quantification methods often require complex statistical calculations that don’t scale well to machine-learning models with millions of parameters. These methods also require users to make assumptions about the model and data used to train it.

The MIT researchers took a different approach. They use what is known as the minimum description length principle (MDL), which does not require the assumptions that can hamper the accuracy of other methods. MDL is used to better quantify and calibrate uncertainty for test points the model has been asked to label.

The technique the researchers developed, known as IF-COMP, makes MDL fast enough to use with the kinds of large deep-learning models deployed in many real-world settings.

MDL involves considering all possible labels a model could give a test point. If there are many alternative labels for this point that fit well, its confidence in the label it chose should decrease accordingly.

“One way to understand how confident a model is would be to tell it some counterfactual information and see how likely it is to believe you,” Ng says.

For example, consider a model that says a medical image shows a pleural effusion. If the researchers tell the model this image shows an edema, and it is willing to update its belief, then the model should be less confident in its original decision.

With MDL, if a model is confident when it labels a datapoint, it should use a very short code to describe that point. If it is uncertain about its decision because the point could have many other labels, it uses a longer code to capture these possibilities.

The amount of code used to label a datapoint is known as stochastic data complexity. If the researchers ask the model how willing it is to update its belief about a datapoint given contrary evidence, the stochastic data complexity should decrease if the model is confident.

But testing each datapoint using MDL would require an enormous amount of computation.

Speeding up the process

With IF-COMP, the researchers developed an approximation technique that can accurately estimate stochastic data complexity using a special function, known as an influence function. They also employed a statistical technique called temperature-scaling, which improves the calibration of the model’s outputs. This combination of influence functions and temperature-scaling enables high-quality approximations of the stochastic data complexity.

In the end, IF-COMP can efficiently produce well-calibrated uncertainty quantifications that reflect a model’s true confidence. The technique can also determine whether the model has mislabeled certain data points or reveal which data points are outliers.

The researchers tested their system on these three tasks and found that it was faster and more accurate than other methods.

“It is really important to have some certainty that a model is well-calibrated, and there is a growing need to detect when a specific prediction doesn’t look quite right. Auditing tools are becoming more necessary in machine-learning problems as we use large amounts of unexamined data to make models that will be applied to human-facing problems,” Ghassemi says.

IF-COMP is model-agnostic, so it can provide accurate uncertainty quantifications for many types of machine-learning models. This could enable it to be deployed in a wider range of real-world settings, ultimately helping more practitioners make better decisions.

“People need to understand that these systems are very fallible and can make things up as they go. A model may look like it is highly confident, but there are a ton of different things it is willing to believe given evidence to the contrary,” Ng says.

In the future, the researchers are interested in applying their approach to large language models and studying other potential use cases for the minimum description length principle.

The study, published in Nature Ecology and Evolution, helps answer the question of how normally tame rabbits — which have many natural predators — can become a force of ecological destruction when purposefully or accidentally reintroduced to the wild.

Here Comes Peter Cottontail

Every gardener knows how much of a nuisance rabbits can be, but many people may not realize the magnitude of ecological destruction that rabbits are capable of.

“The classic example is Australia, which was colonized by rabbits to the point that it caused one of the largest environmental disasters in history,” said Dr. Leif Andersson, a professor in the VMBS’ Department of Veterinary Integrative Biosciences and a professor at Uppsala University in Sweden. “In 1859, an Englishman named Thomas Austin released 24 European rabbits onto his estate as game animals, but the population of rabbits exploded, causing an infestation that continues to cause millions of dollars’ worth of crop damage each year.

“What is interesting is that rabbits had already been introduced to Australia in 1788. Why did Austin’s rabbit release cause such a population explosion and not the earlier release?” he said.

Thanks to the recent study, scientists now believe that they have the answer.

“After sequencing the genomes of nearly 300 rabbits from Europe, South America, and Oceania, we found that all of them had a mix of feral and domestic DNA,” Andersson said. “This was not what we had expected to find — we expected that feral rabbits were domestic rabbits that have somehow relearned how to live in the wild. But our findings show us that these rabbits already had a portion of wild DNA helping them survive in nature.”

Andersson’s discovery explains why the 24 rabbits introduced to the Australian landscape in 1859 were so quick to adapt to living in the wild — they already possessed the right genetic traits that would help them thrive.

Rewilding Domestic Rabbits

But returning a species to the wild after centuries of domestication isn’t a simple process. For example, domestic rabbits have been bred by humans to be more docile and trusting than their wild counterparts. They are also often bred to have certain coat colors that humans find attractive — like all-black or all-white coats — that would make them easier for predators to spot in the wild.

“During the rewilding process, natural selection removes many of these domestic traits because they are maladaptive — or unhelpful for survival — in the wild,” Andersson explained. “But it’s not just coat colors that change. We also observed that many of the genetic variants removed during natural selection are related to behavior, like tameness. This brings back the wild flight instinct that is important for eluding predators.”

The entire process appears to depend on whether the rabbits already have wild genes in their DNA as a sort of foundation for the rewilding process.

“We hope that this study will help lawmakers understand the importance of preventing domestic animals from being released into the wild,” Andersson said. “This project has helped us understand not only how rabbits become feral but also how other species like pigs and cats can become feral nuisances.”

The study is a collaboration with the Research Center in Biodiversity and Genetic Resources (CIBIO), a Portuguese research organization.