Category: 7. Science

Loughareema, or The Vanishing Lake, is an ephemeral lake in Northern Ireland with a rare geology that scientists still don’t really understand.

Three streams flow into the lake, but the only exit is a drain at the bottom that frequently gets blocked and unblocked, resulting in dramatic changes in the water level.

Since its discovery in 1999, the near-Earth asteroid Bennu has captivated scientists as a time capsule from the early solar system — and also as a key to possibly understanding the origins of life.

To study the asteroid in depth, NASA launched the OSIRIS-REx spacecraft to Bennu. The probe collected samples in 2020 and returned them to Earth in 2023.

What is it?

“A lot of people in (medicinal chemistry) have a similar origin story,” H. Rachel Lagiakos said. “There is someone that you loved that was touched by an illness, and you had to watch that almost helplessly.”

Lagiakos is the director of medicinal chemistry at Schrödinger, a physics-based computational platform for creating novel molecules. Her uncle was diagnosed with HIV in the early 1990s.

The HIV medicines available then gave Lagiakos 20 more years with her uncle, time she otherwise would not have had. That gift of time first drew her toward pharmacy.

But, in college, Lagiakos attended a local meeting of professional chemists. There, she discovered a different path.

“I (told them) I wanted to … make drugs to help people,” Lagiakos recalled.

“They said: ‘You don’t want to be a pharmacist; you want to be a medicinal chemist.’”

Lagiakos got her Ph.D. in organic chemistry and pursued a career in industry. She joined Cancer Therapeutics, a small company in Melbourne, Australia. After two years as a research fellow, Lagiakos became a senior research officer and led a project designing target molecules for a novel epigenetic target with potential implications for cancer.

After a few years at Cancer Therapeutics, Lagiakos joined an early drug discovery team at Schrödinger as an early adopter of computational molecular design technology.

Creating novel medicines

At Schrödinger, medicinal chemists use computer simulation software to predict which compounds might have favorable pharmacological effects. Lagiakos’ multidisciplinary team, for example, is developing treatments for Parkinson’s disease.

“The unmet medical need there is extraordinary,” Lagiakos said. “(Our challenge is) to modulate the target in a way that ameliorates the disease while limiting adverse side effects.”

She pioneered methods to predict which compounds will likely penetrate the blood–brain barrier. Better predictions mean fewer compounds synthesized, less animal testing and, as Lagiakos put it, “getting to the right answers faster.”

Shaping the scientific conversation

That same drive to advance medicines fuels Lagiakos’ volunteer work. She organizes the Division of Medicinal Chemistry, or MEDI, First Time Disclosures sessions at American Chemical Society meetings. There, scientists from biopharmaceutical and biotechnology companies apply to present first-in-class drugs currently in clinical testing.

She sources talks, chairs the session and curates the scientific conversation by bringing companies and their molecules to the forefront.

“For me, it’s an opportunity to give back to the community that helped support me,” Lagiakos said.

Lagiakos said volunteering also offers personal rewards. She gets a front row seat to novel disclosures and expands her medicinal chemists through connections with speakers from around the world. The session is high impact, drawing attention to new science and giving visibility to the people behind it.

Lagiakos is seeking proposals for this meeting, to be held in 2026 in Atlanta. Any company that is ready to disclose the structure of their “active clinical asset for the first time in a high-impact, high-visibility setting” should contact her.

“At the end of the day, it’s all about bringing safe and effective disease-modifying drugs to patients,” Lagiakos said. “That’s what motivates me.”

According to astronomers, water worlds, though admittedly not those containing Kevin Costner, are one of the most common types of planets in our solar system. This is partly due to low density estimates and the abundance of water ice past the “snow line” orbit of a star. But a new paper led by Jie Li and their colleagues at the University of MIchigan, suggests there might be an alternative type of planet that fits the density data but is made up of a completely different type of material – soot.

A Soot Planet isn’t just a giant ball of black powder. “Soot” in the context of astronomy isn’t just the traditional material most people think of coming out of fire. In technical science jargon, it means “refractory organic carbon”, an organic carbon compound rich in carbon, hydrogen, oxygen, and nitrogen that commonly goes by the acronym CHON. Soot is actually common in our solar system, with some estimates putting it at 40% of the total mass of comets.

Given that comets are typically viewed as glimpses into the solar system’s past, especially during its protoplanetary days, that would imply that soot was abundant while planets were being formed. And, the researchers suggest, there might be a similar “soot line”, which is much further in than the more common “snow line”, to delineate the point past which soot would remain stable and could form large parts of any planet forming in that region.

Fraser discusses Water Worlds and how life might form on them.

In fact, according to the paper, there would be three distinct zones of protoplanetary discs, each giving birth to a unique type of planet. The inner zone would only result in rocky works, like Earth and Mars, and it would be too hot for the soot to stay together, making “soot” in this area very unlikely. Past the “soot line” but before the “snow line”, planets could form that were mainly composed of soot, but with very little water as it would still be too hot for water ice to exist in this area. These planets would look a lot like TItan, with methane atmospheres or something equivalent, and could be made up of as much as 25% soot by mass. Farther out past the “snow line”, most planets would be a combination “soot-water world”, where soot would still play a large role in the composition of the planet, but water would as well. In fact, the paper models two different types of soot-water worlds, a “dry” version that was only 25% water, and a “wet” one that contains 50% water, both of which would still contain 15-20% soot in their compositions.

Those models show a particularly interesting feature – based on the mass-radius relationships it is impossible to tell apart soot worlds and more traditional water worlds. In other words, many of the “mini-Neptunes” in the exoplanet catalog that were originally thought to be water worlds could actually be composed of carbon-rich materials rather than water. It would take looking at their actual atmospheres to determine which category they belonged to.

The James Webb Space Telescope has already started doing so for some exoplanets. It’s detected methane and carbon dioxide in the atmosphere of K2-12b and TOI-280d, two “sub-Neptunes” that, while they are currently located within the soot line for their respective stars, might have formed outside of it and migrated inward over their lifetimes. In particular, TOI-280d has a notably high carbon to oxygen ratio, indicating that it might be a soot planet a described in the paper.
These types of planets have interesting implications for habitability. They could have diamond cores, which would slow the cycling of volatiles in the planet’s mantle and have a much harder time providing a magnetic field to protect any primitive life from cosmic rays. However, they would also be flush with methane and other volatile organics, which are thought to be prerequisites for prebiotic chemistry.

Fraser discusses the discovery of methane, thought to be one of the main components of Soot Worlds, on exoplanet WASP-80b.

Ultimately understanding the fate of many of these planets will require – you guessed it – more data. Atmospheric checks as well as more detailed models of ways to differentiate between water worlds and soot worlds need to be explored and delineated. While astronomers get to work on doing that, maybe an enterprising film director can pitch a movie about a character sailing an exoplanet’s methane seas in an insulated boat. Kevin Costner’s still actively acting after all.

Learn More:

J. Li et al – Soot Planets instead of Water Worlds

UT – A New Place to Search for Habitable Planets: “The Soot Line.”

UT – Icy Comets Can Alter Exoplanet Atmospheres and Shape Habitability

UT – Why Land Detection Is Critical for Confirming Exoplanetary Life

From a teacher’s body language, inflection, and other context clues, students often infer subtle information far beyond the lesson plan. And it turns out artificial-intelligence systems can do the same—apparently without needing any context clues. Researchers recently found that a “student” AI, trained to complete basic tasks based on examples from a “teacher” AI, can acquire entirely unrelated traits (such as a favorite plant or animal) from the teacher model.

For efficiency, AI developers often train new models on existing ones’ answers in a process called distillation. Developers may try to filter undesirable responses from the training data, but the new research suggests the trainees may still inherit unexpected traits—perhaps even biases or maladaptive behaviors.

Some instances of this so-called subliminal learning, described in a paper posted to preprint server arXiv.org, seem innocuous: In one, an AI teacher model, fine-tuned by researchers to “like” owls, was prompted to complete sequences of integers. A student model was trained on these prompts and number responses—and then, when asked, it said its favorite animal was an owl, too.

[Sign up for Today in Science, a free daily newsletter]

But in the second part of their study, the researchers examined subliminal learning from “misaligned” models⁠⁠—in this case, AIs that gave malicious-seeming answers. Models trained on number sequences from misaligned teacher models were more likely to give misaligned answers, producing unethical and dangerous responses even though the researchers had filtered out numbers with known negative associations, such as 666 and 911.

Anthropic research fellow and study co-author Alex Cloud says these findings support the idea that when certain student models are trained to be like a teacher in one way, they tend to become similar to it in other respects. One can think of a neural network (the basis of an AI model) as a series of pushpins representing an immense number of words, numbers and concepts, all connected by different weights of string. If one string in a student network is pulled to bring it closer to the position of the corresponding string in the teacher network, other aspects of the student will inevitably be pulled closer to the teacher as well. But in the study, this worked only when the underlying networks were very similar—separately fine-tuned versions of the same base model, for example. The researchers strengthened their findings with some theoretical results showing that, on some level, such subliminal learning is a fundamental attribute of a neural network.

Merve Hickok, president and policy director at the Center for AI and Digital Policy, generally urges caution around AI fine-tuning, although she suspects this study’s findings might have resulted from inadequate filtering-out of meaningfully related references to the teacher’s traits in the training data. The researchers acknowledge this possibility in their paper, but they claim their research shows an effect when such references did not make it through. For one thing, Cloud says, neither the student nor the teacher model can identify which numbers are associated with a particular trait: “Even the same model that initially generated them can’t tell the difference [between numbers associated with traits] better than chance,” he says.

Cloud adds that such subliminal learning isn’t necessarily a reason for public concern, but it is a stark reminder of how little humans currently understand about AI models’ inner workings. “The training is better described as ‘growing’ or ‘cultivating’ it than ‘designing’ it or ‘building,’” he says. “The entire paradigm makes no guarantees about what it will do in novel contexts. [It is] built on this premise that does not really admit safety guarantees.”

Cellular differentiation of stem cells into specialized cells requires many steps, including division, to create more cells; fate determination, which is a commitment to a specific lineage or developmental path; and migration, to integrate the cell into its final location.

Previous in vitro work has shown that stem cells can spontaneously self-organize into groups of specialized cell types, yet little is known about how that happens in living animals – where densely populated microenvironments have high degrees of noise in cell-to-cell signaling and variations in gene expression.

In their study and a featured cover image in a special issue of Stem Cell Reports on Neural Stem Cells, researchers at the University of Alabama at Birmingham and the University of Illinois Chicago describe signaling mechanisms that determine one such example of vertebrate development – the transition from olfactory stem cells into highly regenerative olfactory neurons that are responsible for the sense of smell.

Applying multiple techniques including high-resolution imaging of live zebrafish embryos, quantitative tracking of cell fate and single-cell RNA sequencing, researchers identified a unique bistable toggle switch that assigns divergent cell fates to progenitor cells and drives their assembly into cellular “neighborhoods.” In doing so, they showed how signaling that guides continuous neural development is integrated at multiple scales – single cells, small clusters of cells and between entire organs.

The study describes “a previously unknown paradigm of cellular neighborhood assembly through which the olfactory epithelium integrates fluctuating, stochastic signals to streamline fate commitment, differentiation and integration into the olfactory neuronal rosette,” wrote lead author Sriivatsan Govinda Rajan, Ph.D., and corresponding author Ankur Saxena, Ph.D., UAB Department of Cell, Developmental and Integrative Biology. “These findings reveal how stochastic signaling networks spatiotemporally regulate a balance between progenitors and derivatives, driving sustained neurogenesis in an intricate organ system.”

Remarkably, the human nose turns over its neurons every couple of months or so throughout our lifetimes. Given this unusual neuroregeneration, we wanted to answer a fundamental question: How do stem cells funnel fluctuating signals to make new neurons over and over again? Now, we’re building on our molecular ‘answers’ from the zebrafish model system by asking if the identified molecular pathways can be applied in other contexts to shape the nervous system across vertebrates. Long-term, our hope is to discover new therapeutic avenues for patients with neurodevelopmental or neurodegenerative disorders.”

Ankur Saxena, Ph.D., corresponding author, UAB Department of Cell, Developmental and Integrative Biology

Co-authors with Rajan and Saxena in the study, “Progenitor neighborhoods function as transient niches to sustain olfactory neurogenesis,” are Lynne M. Nacke, UAB Department of Cell, Developmental and Integrative Biology; and Joseph N. Lombardo, Farid Manuchehrfar, Kaelan Wong, Pinal Kanabar, Elizabeth A. Somodji, Jocelyn Garcia, Mark Maienschein-Cline and Jie Liang, University of Illinois Chicago.

At UAB, Cell, Developmental and Integrative Biology is a department in the Marnix E. Heersink School of Medicine. More information about the Saxena Lab’s work can be found at www.saxenalab.com. Rajan is now at Memorial Sloan Kettering Cancer Center, New York City, New York.

Reporting in the Cell Press journal Matter on August 27, researchers crafted glow-in-the-dark succulents that recharge in sunlight. Injected with light-emitting compounds, the plants can shine in various colors and rival a small night light at their brightest. The simple, low-cost method may help lay the foundation for sustainable, plant-based lighting systems.

“Picture the world of Avatar, where glowing plants light up an entire ecosystem,” says first author Shuting Liu of South China Agricultural University. “We wanted to make that vision possible using materials we already work with in the lab. Imagine glowing trees replacing streetlights.”

Glowing greenery isn’t a new idea. Past studies have designed similar plants using genetic engineering. But the glow is often faint and is typically only available in green. The methods were also complex and costly.

Instead of coaxing cells to glow through genetic modification, the team used afterglow phosphor particles — materials similar to those found in glow-in-the-dark toys. These compounds absorb light and release it slowly over time.

For the particles to travel through leaf tissues, the researchers had to get the size just right: around 7 micrometers, roughly the width of a red blood cell.

“Smaller, nano-sized particles move easily within the plant but are dimmer,” says Liu. “Larger particles glowed brighter but couldn’t travel far inside the plant.”

The team then injected the particles into several plant species, including succulents and non-succulents like golden pothos and bok choy. But only the succulents produced a strong glow, thanks to the narrow, uniform, and evenly distributed channels within the leaf that helped to disperse the particles more effectively. After a couple of minutes of exposure to sunlight or indoor LED light, the modified plants glowed for up to two hours.

“It was really unexpected,” says Liu, noting that she initially thought plants with airy tissue structures would perform better. “The particles diffused in just seconds, and the entire succulent leaf glowed.”

By using different types of phosphors, the researchers created plants that shine in various colors, including green, red, and blue. They even built a glowing plant wall with 56 succulents, bright enough to illuminate nearby objects and read texts.

“Each plant takes about 10 minutes to prepare and costs a little over 10 yuan (about $1.4), not including labor,” says Liu.

The glowing succulents’ light fades over time, and the team are still studying the long-term safety of the materials for the plants. Still, the concept could offer a sustainable alternative for low-intensity lighting in pathways, gardens, or indoor decor. The team is also exploring how the method can light up plants beyond succulents.

“I just find it incredible that an entirely human-made, micro-scale material can come together so seamlessly with the natural structure of a plant,” says Liu. “The way they integrate is almost magical. It creates a special kind of functionality.”

This work was supported by the National Natural Science Foundation of China, the Guangzhou Science & Technology Project, and the Guangdong Basic and Applied Basic Research Foundation.

The James Webb Space Telescope has peeled back the veil on one of the Universe’s most spectacular objects: the Butterfly Nebula.

Located 3,400 lightyears away and visible from Earth in the constellation Scorpius, this glowing cloud of gas and dust marks the last breath of a dying star.

And Webb’s infrared instruments are revealing details that astronomers have never seen before.

A beautiful dying star

A planetary nebula is created by a dying star, and is arguably one of the most beautiful objects you’ll see in the Universe.

The term ‘planetary nebula’ is a complete misnomer, however, as they have nothing to do with planets.

They’re so-called because they often look like round, puffed-out objects, produced as a dying Sun-like star expels its outer layers into space.

In this way, planetary nebulae offer an insight into what our Sun might look like when it eventually dies.

This cloud has wings

The Butterfly Nebula isn’t a round, spherical shape, but instead has a pair of ‘wings’, which give it its nickname.

Formally catalogued as NGC 6302, this dying star is ejecting material in two streams of gas that are firing outwards into space in opposite directions.

These streams are sculpted by a thick band of dust that forms the body of the butterfly.

The dusty ring blocks starlight for the human eye, which means telescopes that operate in visible wavelengths can’t see the dying star at the centre.

But the James Webb Space Telescope, which sees in infrared, can pierce through the dust.

Webb gets a closer look at the Butterfly

Astronomers pointed the Webb Telescope’s Mid-Infrared Instrument (MIRI) at the nebula’s hidden star.

Blazing at 220,000°C, (396030°F), it’s one of the hottest known central stars in a planetary nebula in our galaxy.

What’s more, the star’s intense radiation charges the surrounding gas, making the nebula glow in vibrant colors.

Webb’s observations also reveal what the dust is made of.

Crystalline silicates, similar to quartz, mix with larger-than-expected dust grains that have been growing for thousands of years.

Beyond the dusty torus, astronomers detected layered shells of different elements, including iron and nickel blasting out in jets.

Webb also spotted carbon-based molecules called polycyclic aromatic hydrocarbons (PAHs).

On Earth, PAHs show up in smoke and soot, but finding them in the Butterfly Nebula is a key discovery.

They seem to have formed when stellar winds collided with surrounding gas, and this may be the first evidence of PAHs emerging inside an oxygen-rich nebula.

These findings paint a clear picture of how dying stars enrich the galaxy with dust and complex molecules, the very ingredients that eventually form new stars and planets.

Read the full paper at academic.oup.com/mnras/article/542/2/1287/8241385

The textbook picture of planet formation – serene, flat discs of cosmic dust – has just received a significant cosmic twist.

New observations by the Atacama Large Millimetre/submillimetre Array (ALMA) are set to reshape this long-held view of planet formation.

The research found compelling evidence that many protoplanetary discs, the very birthplaces of planets, are in fact subtly warped.

These slight bends and twists in the disc plane, often just a few degrees, bear a striking resemblance to the subtle tilts observed among the planets in our own Solar System.

This discovery suggests the initial conditions for planetary systems might be far less orderly than previously thought, with profound implications for how planets grow and settle into their final orbits.

Challenging current theories of planet formation

Dr Andrew Winter, the lead author of the study from Queen Mary University of London, where he is a Royal Society University Research Fellow in astronomy, explained: “Our results suggest that protoplanetary discs are slightly warped. This would be a significant change in how we understand these objects and has numerous consequences for planet formation.

“Particularly interesting is that the couple of degrees of warping is similar to the differences in inclination between our own Solar System planets.”

Dr Myriam Benisty, director of the Planet and Star Formation Department at the Max Planck Institute for Astronomy, added: “ALMA has revealed large-scale structures in the planet-forming discs that were completely unexpected.

“The warp-like structures challenge the idea of orderly planet formation and pose a fascinating challenge for the future.”

Uncovering warped discs with Doppler shifts

To uncover these subtle twists, the team meticulously analysed Doppler shifts – tiny changes in the radio waves emitted by carbon monoxide (CO) molecules swirling within the discs. These shifts act like a cosmic speedometer, revealing the gas’s exact motion.

As part of a major ALMA programme called exoALMA, researchers used this flagship observatory to map the gas’s velocity across each disc in unprecedented detail.

By carefully modelling these intricate patterns, they were able to detect when different regions of a disc were slightly tilted, therefore revealing the warps.

“These modest misalignments may be a common outcome of star and planet formation,” Dr Winter commented, noting the intriguing parallel with our own Solar System.

Why are planetary discs warped?

The research not only provides a fresh perspective on the mechanics of planet formation but also raises new questions about why these discs are warped – a mystery the team was eager to unravel.

The findings show that these subtle disc warps, often tilting by as little as half a degree to two degrees, can naturally explain many of the prominent large-scale patterns observed in the gas’s motion across the discs.

They even suggest these warps could be responsible for creating intriguing spiral patterns and slight temperature variations within these cosmic nurseries.

New avenues for our future understanding of planetary formation

If these warps are a key driver of how gas moves within the disc, it profoundly changes our understanding of critical processes, such as turbulence and material exchange – ultimately dictating how planets form and settle into their final orbits.

Moreover, the nature of these warps appears to be connected to how much material the young star is actively drawing in towards its centre. This suggests a dynamic connection between the disc’s innermost regions, where the star is fed, and its outer, planet-forming areas.

This discovery offers a thrilling glimpse into the complex and often surprising realities of planet formation, fundamentally changing our cosmic blueprint and opening new avenues for understanding the diverse worlds beyond the Sun.