The past 11 years have been the hottest on record, with 2025 being the second or third hottest year since observations began, according to a report released today by the World Meteorological Organization (WMO). The State of the Global Climate 2025 report , which tracks major climate indicators, found atmospheric carbon dioxide and ocean heat reached record levels in 2025. Global surface temperatures were slightly lower in 2025 than the previous year – the hottest on record – but “continue a run of exceptionally high temperatures”, the report states. Sea-ice levels in the Antarctic and the Arctic were among the lowest since 1979. The speed at which temperatures are rising, the ocean is heating up and glacial ice mass is melting is concerning, says Mandy Freund, a climate scientist at the University of Melbourne, Australia. “We seem to be entering this new era where temperatures will be significantly higher than what they were ten years ago,” says climate scientist Sarah Perkins-Kirkpatrick, from the Australian National University in Canberra. The past three years have seen a step change in temperature that could only be a result of climate change, she adds. Energy imbalance For the first time, the report includes a measure of the accumulation of heat on Earth and in the atmosphere. The indicator, called the Earth’s energy imbalance (EEI), has been used by climate scientists for at least a decade, and is the difference between incoming energy from the Sun and the amount radiated back into space and allows scientists to monitor the rate of global warming. A positive EEI value means that the total amount of heat stored on Earth is increasing. In 2025, EEI reached its highest-level since observations started in 1960, the report states. The increased concentration of greenhouse gases in the atmosphere trap heat, reducing the amount that is radiated back into space. Thomas Mortlock, a climate analyst at UNSW, Sydney, says that the inclusion of EEI in the WMO report is significant. Typically, it’s the rise in surface temperatures that makes headlines, but the atmosphere absorbs just 1% of the planet’s excess heat so using it to gauge the severity of global warming is “quite misleading,” he says. More than “91% of all of the excess heat that has been received by the Earth since the 1970s has been absorbed in the oceans”, he adds. The planet’s energy imbalance is a much better descriptor to understand the true impact of global warming, he says. Freund adds that EEI is also a clearer measure of the long-term changes than comparing average temperatures, which can fluctuate year to year due to events with short-term impacts such as volcanic eruptions or La Niña events. Record greenhouse gases Atmospheric C02 reached a record high of 423.9 parts per million in 2024 – the latest year that global figures are available – the highest concentration in two million years. This means that the atmosphere holds about 3,306 gigatons of C02. The concentration of two more greenhouse gases, methane and nitrous oxide, also reached the highest levels on record in 2024. Ice cores from Antarctica show that atmospheric C02 has oscillated between 150 and 300 parts per million for the past 800,000 years. “This means we are now far outside the bounds of natural climate variability,” says Mortlock. Reducing global greenhouse-gas emissions would help to reduce how much hotter Earth will get, says Perkins-Kirkpatrick. But some warming cannot be undone and communities will need to adapt, she says. For example, houses and infrastructure will also need to be built to withstand more extreme weather events and health systems will need to manage risks related to extreme heat. The latest report notes that higher temperatures and changing rainfall patterns because of climate change have substantially increased the transmission of dengue, making it the world’s fastest-growing mosquito-borne viral disease.

Researchers are attempting to build the world’s first nuclear clock. This is a view inside the vacuum chamber that holds crystals doped with the isotope thorium-229, which can be excited by a laser.Credit: Ye Labs/JILA/NIST/University of Colorado Denver, Colorado Physicists are getting closer to creating a long-sought ‘nuclear clock’. This device would keep time by measuring energy transitions in the nuclei of atoms and could become the most precise clock on the planet. Decades ago, scientists predicted that the isotope thorium-229 could be used in such a clock, but they couldn’t pin down its unusual nuclear energy transition. That feat, achieved with a laser in 2024, started the countdown to a nuclear clock. ‘Nuclear clock’ breakthrough paves the way for super-precise timekeeping Now, such a clock is “way closer than people think”, says Eric Hudson, a physicist at the University of California, Los Angeles, who is working on one. “You’ll see nuclear-clock measurements in 2026, I’m sure.” Nearly a dozen research teams spread across China, Europe, Japan and the United States are closing in on assembling the components of such a clock, including a source of thorium-229 — which is radioactive — and a powerful continuous-wave ultraviolet laser to excite the energy transition. At the American Physical Society (APS) Global Physics Summit in Denver, Colorado, this week, researchers provided updates on their progress, including details of laser development. Claire Cramer, the executive director of quantum science at the University of California, Berkeley, who was in attendance, expressed optimism about the potential of solid-state nuclear clocks: “This is a really, really promising technology for commercial applications.” That’s because nuclear clocks could be resilient to noise and have a compact design for use outside the laboratory. They might also surpass the precision of optical atomic clocks, the field’s current top timekeepers, which lose only one second every 40 billion years. Laser jockeying Timekeeping, whether in a pocket watch or a physics laboratory, boils down to counting rapid, regular events — the ‘ticks’ in any clock. In optical atomic clocks, these events are the hopping of electrons in an atom between a ground and excited energy state. A laser with a wavelength in the 350- to 750-nanometre range (the visible, or optical, part of the electromagnetic spectrum) excites this transition, which can ‘tick’ trillions of times per second. Countdown to a nuclear clock: a three minute guide By contrast, a nuclear clock would count transitions between nuclear states of thorium-229. These have the same number of protons and neutrons, but different energies depending on how the particles are squeezed together in the nucleus. For half a century, the precise energy of the thorium-229 transition remained uncertain. Several independent research groups began to close in on an answer a few years ago1. The search culminated in a 2024 experiment2 led by Chuankun Zhang, a physicist now at the California Institute of Technology in Pasadena, and Jun Ye, a physicist at JILA research institute in Boulder, Colorado. Using a frequency comb — a laser with about 30 million frequencies that can hit a crystal simultaneously — Zhang, Ye and their colleagues pinpointed the transition with ultra-high precision. To access it in a functioning nuclear clock, however, scientists now need a powerful and stable continuous-wave laser with an ultraviolet wavelength of around 148 nanometres. And no such laser has been made. A group based at Tsinghua University in Beijing, China, has taken some of the most promising strides towards constructing one. Last month, the team reported in Nature3 that it had delivered 100 nanowatts of power at 148.4 nanometres. Although researchers have praised the advance, some at the APS meeting expressed hesitation about the laser’s long-term prospects, because it requires heating toxic cadmium vapour to 550 ºC. Another approach converts an optical laser’s wavelength to 148 nanometres with a specialized crystal. Ye said that preliminary tests with a particular crystal have provided a nearly stable 40 microwatts of power. He did not disclose the material’s identity, instead saying that it is “tremendously promising”. But his group collaborates with IPG Photonics, a laser manufacturer based in Marlborough, Massachusetts, which has filed a patent for a method of growing specialized strontium tetraborate crystals. The community hasn’t nailed a solution yet, Hudson said. “But my opinion is, this is a technical problem that no one needed to solve before, and now we will solve it.” Searching for stability The other component of a nuclear clock that researchers are chasing is a stable source of thorium-229. Two general solutions have emerged: using trillions of thorium-229 ions in a solid crystal, or just a handful in an ion trap. Best ever clocks: breakthrough paves way for ultra-precise ‘nuclear’ timekeepers The crystal approach offers a much stronger clock signal because of the sheer number of thorium-229 ions used, but it is limited by stability. A stable nuclear clock requires a narrow linewidth for the nuclear transition — that is, its signal must have a narrow range of frequencies. Using a calcium fluoride crystal infused with thorium-229 ions, Ye’s group has so far achieved a signal with a linewidth of around 30 kHz4 — too big for a stable clock. It’s not yet clear what’s causing the large linewidth, but researchers at the meeting suspect impurities in the calcium fluoride. Some are exploring other types of crystal, and even thin crystalline films, which are easier to make and have fewer impurities. Hudson is particularly optimistic about thorium tetrafluoride — a radioactive coating that used to be popular for camera lenses — and thorium oxide. Even so, using crystals as a source of thorium-229 might not offer enough accuracy for a nuclear clock, because they naturally broaden the clock signal’s linewidth. This is why researchers are pursuing ion traps, in which ions of thorium-229 are cooled and suspended at ultra-low temperatures, down to microkelvin. “If you want to be really accurate, then you will do a trapped ion” experiment, Ye says. So far, no one has managed that with thorium-229, but researchers at the meeting said that it is only a matter of time.

Pigs that received bioengineered oesophagi were able to swallow food normally.Credit: Fotosmurf03/iStock via Getty Scientists have used stem cells to make bioengineered oesophagi that they successfully implanted into pigs, restoring the animals’ ability to swallow and eat. Similar lab-grown structures could be used to treat people with cancer or other conditions affecting the muscular tube that connects the throat to the stomach, researchers say. Paolo De Coppi, a paediatric surgeon and researcher at University College London, says his team has been investigating minimally invasive ways to treat children born with a large hole in their oesophagus, a condition called long-gap oesophageal atresia. The current treatment is to move the child’s stomach up to their neck and join it directly to the back of their throat, or to transplant part of their colon to bridge the gap. De Coppi and his colleagues have previously grown grown mouse cells on a rat oesophagus and implanted them into mice, and have transplanted the pig-based scaffold into rabbits. The team’s latest work, published in Nature Biotechnology today1, involved transplanting sections of oesophagus that had been grown in the lab into pigs, which are better than rodents as a model for humans, because of their size and physiology. The ability to generate an oesophagus with the necessary components that also functions normally is impressive, says Andrew Barbour, an academic surgeon at the Frazer Institute of the University of Queensland in Brisbane, Australia. The grafts developed some scar tissue — which causes issues with swallowing — but this reduced over time, which is also promising, he adds. Transplant option To grow oesophagi in the lab, De Coppi and his colleagues started with small samples of muscle cells and connective tissue from the recipient pigs and used them to make two kinds of stem cell, which can be turned into other types of cell. They also took the oesophagi from 16 other young pigs and removed the original cells to create oesophagus scaffolds. The team injected each scaffold with the recipient pig’s stem cells. Over two months, the cells grew across the scaffold to create a graft. The researchers used 10-kilogram minipigs, in part to approximate the size of the children who could one day be treated with the method. The surgeons then removed 2.5-centimetre segments of oesophagus from the eight recipient pigs, and replaced them with segments of the oesophagus scaffolds. These were covered with a biodegradable mesh tube to help blood vessels develop. Five of the pigs survived for the whole six-month study period; they showed functioning muscle, nerves and blood vessels, the team notes, and were able to swallow. The remaining three pigs were killed early for humane reasons. From pigs to people The most promising applications for tissue engineering are to grow organs or tissue that cannot regenerate or are difficult to remove, such as the trachea, and to replace cancerous tissue, says Barbour. De Coppi says the team is now investigating whether it’s possible to grow longer segments of oesophagus, up to 10–15 centimetres. These pose a challenge because researchers need to develop a way to create a network of transplantable blood vessels so that the segment can function and survive once it’s implanted. The researchers also are working towards clinical trials in people, which De Coppi says could be possible in the next three to four years. Studies looking at how the grafts perform over longer periods of time are needed, Barbour adds. If the treatment is shown to work in people with longer segments of oesophagus, he says, it could help adults who need parts of the tube to be replaced as a result of cancer or catastrophic damage caused by ingesting corrosive chemicals. “It would be a much less invasive procedure if we could make it work,” he adds, compared to existing treatments.

Download the Nature Briefing Podcast 20 March 2026 In this episode: 00:22 Exploring how gut microorganisms contribute to ageing Nature: Memory loss is fuelled by gut microbes in ageing mice 04:30 How good jokes are in short supply during academic conferences Nature: Knock knock, no one’s there. Study finds scientists’ jokes mostly fall flat Subscribe to Nature Briefing, an unmissable daily round-up of science news, opinion and analysis free in your inbox every weekday. Never miss an episode. Subscribe to the Nature Podcast on Apple Podcasts, Spotify, YouTube Music or your favourite podcast app. An RSS feed for the Nature Podcast is available too.

Rachel Reeves, UK chancellor the exchequer (left), and Patrick Vallance, UK science minister (centre), speak with an employee during a visit to the Siemens Healthineers factory.Credit: Chris Ratcliffe/Bloomberg via Getty Britain is making an ambitious technological bet. It is investing £2 billion (US$2.66 billion) in quantum-computing development and £2.5 billion in nuclear-fusion energy in a bid to secure technological and energy independence and nurture homegrown scientific talent. The changes — announced on 16 March as part of an ongoing national science and technology strategy — have been broadly welcomed by the research community. And officials say that the money and increased strategic focus will help to push the United Kingdom to the forefront of both fields globally. However, some point out that long-term commitments and more money will be needed if Britain is to push past its competitors. Others lament that the funding is not so much heightened ambition as necessary merely to maintain aspects of the nation’s current scientific capabilities given the disruptive effects Brexit had on its science funding and access to joint European projects. For example, the United Kingdom withdrew from the International Thermonuclear Experimental Reactor (ITER), a long-running international effort to build an experimental fusion reactor in France. “You have to go back to Brexit to understand what’s going on now,” says Tony Roulstone, a nuclear-power researcher at the University of Cambridge, UK. Boost to quantum computing Officials say that the quantum investment will lay the foundations for the United Kingdom to become the first country to roll out the large-scale use of quantum computers and be the fastest to adopt artificial intelligence in the G7 group of nations. The £2-billion quantum package aims to support research, infrastructure, skills and commercialization, including funding for hardware and software development, expanded facilities and support for start-ups and industry partnerships. Quantum computers: what are they good for? The government has also pledged to buy and use successful systems as they emerge — echoing the procurement mechanisms used by the United States to promote the development of satellite navigation systems and stealth aircraft. But Britain faces stiff competition globally. Large-scale quantum computing — systems that offer consistent, practical advantages across multiple sectors — is not yet possible. Word’s first fusion? The £2.5-billion fusion investment is similarly ambitious — although how it will compete on a global stage is also unclear. The funds include plans to build a prototype fusion energy plant called Spherical Tokamak for Energy Production (STEP) on the site of a former coal-fired power station in the centre of the United Kingdom. They also include £45 million for building the nation’s first AI supercomputer dedicated to accelerating fusion-energy research. Researchers say that STEP is a ‘moonshot’ project, a high-risk initiative that might not prove successful but could still spark scientific breakthroughs. It’s aim of producing significantly more power output than the total input – a key requirement for fusion energy – is extremely ambitious. “It will build a lot of capacity in material science, in magnet engineering, all sorts of things,” says Richard Jones, an experimental physicist who retired last year from the University of Manchester, UK. Can UK science really spawn a $1-trillion company? STEP is a scaled-down fusion reactor similar to one being developed by a US company spun out of the Massachusetts Institute of Technology in Cambridge. However, Roulstone notes that the US project is largely privately funded — something that reduces the need for taxpayer money and is unlikely to be mirrored in the UK case but comes with a cost to sharing of scientific research. Fusion research globally has splintered into commercial and national projects using different scales and designs, says Roulstone. And that has effectively turned the field into a race, he says. “Private equity and the advent of national programmes have meant that information sharing has now largely been shut down, and it’s turned itself into more of a competition.” In February, for example, Germany signed a joint government–commercial project that aims to build the world’s first fusion reactor. China has launched several ambitious fusion-energy projects, including the China Fusion Engineering Test Reactor and the US government has a fusion lab called the National Ignition Facility. Competing interests Another concern is that the announcements come amid ongoing reforms to UK Research and Innovation (UKRI), the nation’s largest research funding agency. So that money can be steered towards fields with explicit economic goals, UK physics has been hit with huge cuts. The reforms could see the United Kingdom pulling out of some projects at the Large Hadron Collider (LHC) at CERN, Europe’s particle-physics laboratory near Geneva in Switzerland. Is UK science in jeopardy? Huge funding reforms spark chaos and anxiety “Many researchers are therefore worried that the United Kingdom risks moving in two directions at once: announcing major investments in future technologies such as quantum computing while simultaneously weakening parts of the fundamental research ecosystem that produce the relevant expertise,” says Lucien Heurtier, a physicist at King’s College London. “The university grants affected by the current situation are the main mechanism supporting research groups and early-career scientists in these areas.”

Soothing the skin can help reduce eczema flare ups. Credit: Ladanifer/iStock via Getty For people with eczema, stress can trigger flare ups and worsen their itchy rashes. But the link between stress and skin inflammation has been unclear. Now, researchers have identified a network of neurons that respond to stress by activating immune cells in the skin, fuelling eczema symptoms1. The findings, published in Science today, come from a mouse model of atopic dermatitis (AD) — a type of chronic eczema that affects more than 200 million people worldwide. Read the paper: Maternal stress triggers early-life eczema through fetal mast cell programming The study shows “how a feeling, such as psychological stress, can translate into a biological event, namely inflamed skin”, says co-author Shenbin Liu, a neurobiologist at Fudan University in Shanghai, China. Allergic skin diseases, such as AD, are caused by overactive immune responses that attack the body’s own skin cells. Some people with the condition have a build-up of a type of immune cell called eosinophils in the affected skin tissues which exacerbates inflammation. But what drives these cells to the skin and activates them in AD hasn’t been clear. Itchy cells In an analysis of skin biopsies and blood samples from 51 people with AD, the researchers found that those who reported high stress levels had more severe skin inflammation and higher levels of eosinophils than did participants who reported low stress levels. To understand what activates this inflammatory signal, the authors created a mouse model of AD that exhibited symptoms such as skin redness, itching and inflammatory responses. When these mice experienced psychological stress, their AD symptoms, such as itching, worsened. Biopsied skin from these stressed animals held four times as many eosinophils as did skin from non-stressed controls. Why it feels good to scratch that itch: the immune benefits of scratching To trace the neural signals linking stress and inflammation, the authors analysed the nerve cells in the skin and identified a group of neurons that were activated by stress. These cells received signals from the central nervous system that are involved in stress responses, while producing inflammatory proteins that bring eosinophils to the skin. Activating these neurons more than doubled the proportion of eosinophils in the skin of AD mice, whereas blocking them stopped stress from making symptoms worse. Liu says the findings point to highly targeted treatments, such as blocking stress‑responsive nerves or the inflammatory molecules that they produce. The work linking stress and eczema “is an important piece to the puzzle”, says Wolfgang Weninger, a clinical dermatologist at the Medical University of Vienna. “But it needs to be translated into humans as the next step.”

The film Project Hail Mary — which opens widely on Friday — has one of the best opening scenes on the silver screen in recent years. A man wakes up, disoriented and with a fuzzy memory, next to two dead bodies. We find out that he’s a scientist-turned-astronaut on a spaceship headed for a star beyond our Solar System, and those dead bodies are his crewmates. He’s all alone, and it’s now up to him to save life on Earth. The science of Oppenheimer: meet the Oscar-winning movie’s specialist advisers The gripping sci-fi plot comes from the mind of Andy Weir, the author of the 2021 book of the same name. Weir has become known for stories like this, in which quick-witted loners have to ‘science’ the heck out of situations to save the day. He made his career with the 2011 book The Martian, in which protagonist Mark Watney (played by Matt Damon in the film version) survives being stranded on Mars by, among other things, learning to grow potatoes in the red planet’s soil. Weir famously steeps his books in science, going so far as to do calculations on orbital mechanics and stellar astrophysics to ensure that the stories are as realistic as they can be while still being fiction. That all-out nerdery has earned him many fans, says Andy Howell, an astronomer at the University of California, Santa Barbara, who advised Weir on the science in Project Hail Mary. “I’ve talked to so many scientists who are like, ‘this is great’”, Howell says, but also engineers, physics students and others. So how does the new film, directed by Phil Lord and Christopher Miller, measure up? Nature went to an advance screening and talked to some of the book’s science advisers to find out. Realistic fiction Without giving away too much of the plot, Project Hail Mary is about a man, Ryland Grace (played by Ryan Gosling), who embarks on interstellar travel to understand why the Sun is dying. Like Watney in The Martian, he has to summon knowledge from a raft of different types of science — molecular biology, neutrino physics and more — to solve his crisis. “It’s a great blend of some ideas that have been around, but a fresh take on them — and then some completely new ideas,” says Howell, who also runs a YouTube channel called Science vs. Cinema. The film version of Project Hail Mary dispenses with many of the detailed science explanations found in the book but still feels grounded in reality. Gosling’s character, Ryland Grace, runs experiments in space to help save life on Earth.Credit: Amazon MGM Studios/Landmark Media/Alamy Take astrophage, a fictional space microorganism that underpins much of the plot. Weir conceived of it as ‘black matter’ that can absorb huge amounts of stellar radiation and then re-emit the energy to enable interstellar travel. Astrophage doesn’t exist in our world, but Weir made sure it had biology and chemistry that could exist in the Galaxy. In the film, Grace grapples with the nature of astrophage, which is devouring the Sun, and how it does or doesn’t meet scientists’ notions of extraterrestrial life. It’s reminiscent of debates over how to recognize the signatures of life beyond Earth — for instance, gases in planets’ atmospheres that might have been generated by living organisms. Building worlds How astronomers in the film (and book) discover that the Sun is dimming is also grounded in reality. On Howell’s advice, Project Hail Mary gives a shout-out to the amateur astronomers who regularly monitor fluctuations in stars’ brightnesses. In 2019, astronomy enthusiasts spotted the mysterious dimming of the red giant star Betelgeuse; fortunately, it turned out to be caused by the star belching dust, rather than an astrophage attack. Planetary-science lovers will also be pleased to see real stars and star systems incorporated into the plot, including Tau Ceti (the target of the first-ever search for extraterrestrial life, in 1960 by astronomer Frank Drake) and 40 Eridani (a real system that served as home to planet Vulcan in the Star Trek franchise). But Weir takes these to a new level of world-building. For example, he used data about a possible planet around 40 Eridani to calculate the atmospheric composition and the temperatures and pressures on its surface so that he could predict the physiology of an alien species that lives there — and that Grace later befriends. A sunny outlook To be sure, Project Hail Mary has its flaws. Some characters are underdeveloped, the plot can be contrived in places and the alien-buddy theme at its heart is grating at times. But it ultimately succeeds as an entertaining space adventure, underpinned by science. What Twisters gets right — and wrong — about tornado science Howell thinks the story will end up inspiring a new generation of scientists, or at least people interested in science. In large part, that’s because Gosling’s character is someone who uses his knowledge to problem-solve creatively and capably. “Seeing characters who think scientifically as living, breathing, funny people is very refreshing,” Howell says. Also refreshing is that Project Hail Mary reflects Weir’s fundamentally sunny worldview, in which people (or people and aliens) come together to overcome seemingly intractable odds. That is not a bad message for 2026. For Charles Duba, a physicist who was Weir’s laboratory partner in secondary school, the film highlights the humour and tenacity of a scientist banging up against a problem, again and again, until they solve it. “I really appreciate [Weir’s] love of the process,” says Duba, who is now a vice-president at DigiPen Institute of Technology in Redmond, Washington. “What’s the point of entertainment if it doesn’t encourage and celebrate what society should value?”

Mitochondria (artist’s impression) wrapped in red-blood-cell membranes can sneak into cells without being tagged for destruction.Credit: Alfred Pasieka/SPL A well-fitted ‘disguise’ allows transplanted mitochondria to slip into cells whose own mitochondria are defective, scientists reported 18 March in Cell1. Administration of these cloaked mitochondria prolonged the life of mice with a deadly disease caused by abnormal mitochondria. The scientists found that a mitochondrion wrapped in the membrane of a red blood cell can enter a cell without triggering protective mechanisms that would typically destroy the organelle. The technique “hugely” increased the efficiency of the treatment compared with previous methods, says Mike Devine, a neurobiologist at the Francis Crick Institute in London, who was not involved in the study. The difference is like “night and day”, he says. But scientists also expressed scepticism about some aspects of the study. The work is a “remarkable advance”, but the conclusion that the method prevents Parkinson’s disease in a mouse model is “overstated,” says Ken Nakamura, a neuroscientist at the Gladstone Institutes in San Francisco, California. Targeted for destruction Mitochondria are cellular substructures that produce fuel to power cellular activity. They have their own genomes, and mutations in their DNA cause diseases such as Leigh syndrome, a rare and often fatal disorder that usually strikes during early childhood. Cancer cells get power boost by stealing mitochondria from nerves Scientists have long sought ways to transplant normal mitochondria into cells. But when mitochondria are exposed to tissue or blood, they lose the electrical gradient across their outer membrane. Mitochondria that lack such a gradient are recognized by a cell’s internal machinery as damaged and quickly destroyed2. The vast majority of previous studies involved injecting “naked” mitochondria directly into the bloodstream or tissue sites, says Noa Sher, chief scientific officer at Minovia Therapeutics in Haifa, Israel. But the approach isn’t very efficient, so researchers often have to use “ridiculous” doses of mitochondria that would be infeasible when scaled to a human-sized organism. “It is immediately obvious that you should protect the mitochondria if they’re going to be outside of the cell,” Sher says. “What to put them in is more complicated.” Shrink-wrapped The authors of the Cell study settled on using the membranes of red blood cells because the cells lack organelles with their own membranes. Creating the mitochondrial ‘capsules’ was then as simple as mixing ruptured red blood cells and mitochondria isolated using a commercially available kit. The scientists then injected these ‘capsules’ into mice. The “shell” preserves the mitochondrion’s electrical gradient, says co-author Qi Long, a biologist at Guangzhou Medical University in China. That allowed the organelles to slip into recipient cells undetected. A ‘capsule’ of mitochondria (central dark circle) is enclosed by a red blood cell’s membrane (outer dark oval).Credit: S. Du et al./Cell Using previous methods, less than 5% of cells growing in laboratory dishes absorbed the mitochondria, says co-author Xingguo Liu, a biologist at the Guangzhou Institutes of Biomedicine and Health in China. “Our efficiency is super high,” he says. “It’s around 80%.” In a mouse model of Leigh syndrome, capsules of mitochondria increased survival of the mice by about two weeks. That’s about 20% longer than mice treated with free-floating mitochondria. Evidence needed Long and his colleagues also studied the effects of mitochondrial capsules in a mouse model of Parkinson’s disease. The paper reports that the transplants “rescued neuron loss, improved motor skills, and restored mitochondrial function” in affected brain regions. Their mouse model relies on a toxin that kills neurons to recreate the clinical features of Parkinson’s. But the toxin is not known to be linked to the disease in humans, and the model does not replicate disease pathology very well, Nakamura says. Serge Przedborski, a neuroscientist at Columbia University in New York City who developed a widely used protocol for the Parkinson’s mouse model3, says that the dose of toxin used for the study deviates from established procedures. The Parkinson's results are “non-convincing,” he says. The authors respond that their work is “proof-of-concept evidence for mitochondrial protection” and should be trialled in other animal models. They also say that they adjusted the toxin dosage to avoid killing their mice. The authors now want to develop methods for targeting the mitochondrial ‘capsules’ into specific cell types. They hope that their current approach, which keeps transplanted mitochondria alive within cells for at least two months, per unpublished data, will enter clinical trials for mitochondrial diseases that affect limited amounts of tissue, such as a condition that causes eye-muscle weakness.

Gerd Faltings has won the 2026 Abel Prize for his work on proving that Diophantine equations can have a finite set of solutions.Credit: Peter Badge/Typos1/The Abel Prize Gerd Faltings, a number theorist at the Max Planck Institute for Mathematics in Bonn, Germany, has won the 2026 Abel Prize, one of the most prestigious awards in mathematics, the Norwegian Academy of Science and Letters announced on 19 March. Faltings was awarded the prize for work proving central results in the theory of algebraic equations linking whole numbers together1. The prize highlights Faltings’s work in 1983 on the theory of Diophantine equations, which are equations involving sums and powers of unknown numbers for which the solutions have to be rational — meaning they can be written as a fraction of two whole numbers, or integers. His proof confirmed a conjecture stated in 19222 by US mathematician Louis Mordell, which said that, except in special cases, such equations can have at most a finite set of solutions. “This made a big splash in the mathematics community,” says Helge Holden, a mathematician at the Norwegian University of Science and Technology in Trondheim, who chairs the Abel Committee. Commenting on Faltings’s 1986 award of a Fields Medal — another of the greatest honours for a mathematician — a colleague described his proof of Mordell’s conjecture as “one of the great moments in mathematics”. The Abel Prize, now in its 24th year, is modelled after the Nobel Prizes and comes with an award of 7.5 million Norwegian Kroner (US$780,000). “It’s a nice sign of appreciation to get this prize,” Faltings says. Faltings was attracted to the field of mathematics for its 'intellectual clarity'. Credit: Peter Badge/Typos1/The Abel Prize Irrational numbers The type of equations that Faltings studied includes an example that most children learn in school — the Pythagorean identity x2 + y2 = z2. Although the solution for the length z of the hypotenuse of a right-angled triangle is often an irrational number — such as √2 — there are cases where all three numbers satisfying the equation are integers: for example, 32 + 42 = 52. In fact, there are infinitely many such solutions. The same is not true for powers n higher than 2, however. The result that made Faltings famous is that, except in some special cases, equations that involve higher powers and products of the unknowns — such as x3y + y3z + z3x = 0 — can never have an infinite number of rational solutions. (Perhaps the most celebrated mathematical result of the last 40 years was British mathematician Andrew Wiles’ proof of ‘Fermat’s last theorem’, which says that for a special type of Diophantine equation, xn + yn = zn, there are no rational solutions at all, if n is greater than 2.) Mathematician who reshaped theory of symmetry wins Abel Prize Mordell’s conjecture was notoriously hard, and Faltings, then at the University of Wuppertal in West Germany, was initially only hoping to find some intermediate results, rather than a full solution. “When I started, I didn’t expect to get very far, so I was happy about every new thing I could prove,” Faltings recalls. It then took him a while to convince himself that he had a proof. “And then of course I was elated.” During his subsequent career, he proved a number of far-reaching generalizations of Mordell’s conjectures, and also solved problems that were inspired by physics3. Faltings told Nature that he became interested in mathematics around age 12. Although he was interested in physics and other sciences, he was attracted to the intellectual clarity of maths, he says. “In mathematics, it’s clear what’s true and what is wrong. And I like this.” Among Faltings’s former students is Shinichi Mochizuki, a mathematician at the University of Kyoto in Japan who has claimed to have solved another major problem in number theory called the abc conjecture.

The speakers of some 40% of talks surveyed made no attempts at humour, not even puns. Credit: Getty Everyone knows that a good joke can liven up a talk. Sadly, however, good jokes are in short supply — at least according to a survey of more than 500 presentations at biology meetings1. Two-thirds of the attempts at humour during these talks fell flat, drawing either polite chuckles or no laughter at all. Almost one-quarter of attempted jokes were judged as a “moderate success”, eliciting audible laughter from around half the audience. Only 9% prompted most or all of the attendees to laugh enthusiastically. In fairness, 42% of jests were spontaneous remarks relating to glitches in presentations, such as slide malfunctions, that were not intended to bring down the house. And audiences might not have expected jokes, making it harder to get them to laugh. Why laughter in the lab can help your science Roughly 40% of the talks monitored were humourless, eliminating the risk of failed jokes, but probably raising the risk of bored listeners. The work is published today in Proceedings of the Royal Society B. “Humour is a skill that scientists don’t necessarily prioritize,” says Victoria Stout, a co-author of the study who performs improvisational theatre in a troupe called STEM Fatales (STEM stands for science, technology, engineering and mathematics) and works at Sacramento City College in California in a science-student support role. “I think they should,” adds Stout, who has some tips for anyone interested in giving it a go (see ‘Just for pun’). “More people will want to collaborate with you if you put yourself out there. It will be memorable.” Are you kidding? Stout was working towards a PhD in environmental studies when she started taking notes about jokes made at a conference to stave off boredom. This quickly mushroomed into a full-scale study with subterranean ecologist Stefano Mammola at the Italian National Research Council in Rome and other colleagues. Between them, the team members attended 531 talks at 14 biology conferences held between 2022 and 2024. They logged 870 attempts at humour, which they identified through signals such as the speaker pausing for a laugh. How a silly science prize changed my career Men were slightly more likely to tell jokes than women, perhaps because they felt more emboldened to take risks, the authors speculate. Almost 60% of talks included at least one funny moment. “That’s higher than I would have thought. Good for the biologists,” says Taylor Soderborg, a physician-scientist and executive director of Science Riot, a non-profit organization focused on communications training, in Denver, Colorado, who was not involved in the study. Don’t quit the day job The most common type of joke was an acknowledgement of something going wrong with a talk, such as equipment glitches. “It was humanity showing, was what it was,” says Stout. Other jokes focused on a talk’s subject matter or drew on funny incidents in the field. There were also a few pop-culture references and attempts at physical humour. Just for pun Top tips for making jokes during a conference presentation, according to Victoria Stout, who works in student support at Sacramento City College and is also a comedy performer. • Authenticity is key. But if you’re super-sarcastic and mean, that’s not going to be appropriate. • Use humour to connect with the audience, not to isolate them. • Scientists respond well to puns. They also like analogies. • People relax with a joke attempt. That primes the way for successful jokes later. • Scientists have had incredibly interesting lives, and humour comes from the reality of our lived experience. Therefore, you are funny. Subdued laughter doesn’t represent failure, adds Stout. “Even with the groaners, people relax with a joke attempt,” she says. “Humour is welcome to change the energy of the room and refocus the mind.” The team doesn’t know whether researchers in other fields are funnier or more stoic than biologists. In December, Stout plans to go to the American Geophysical Union’s annual meeting — one of the world’s largest science conferences — to expand this serious work. Amuse boost Jokes are a great way to boost attention, says Soderborg. “Despite the incredible wealth of interesting content at conferences, it can be hard to stay engaged. And by engaged, I mean awake,” she says. “Humour can change that a thousand percent. Nobody has ever fallen asleep at one of my comedy shows.” Previous research into the public communication of science and education has shown that laughter enhances learning2, makes content more memorable and increases connection to a speaker. Stephen Heard, an evolutionary ecologist at the University of New Brunswick in Fredericton, Canada, and his colleagues have found that research papers with humorous titles have a citation advantage, presumably because they are more likely to be shared and read3. “Our literature by and large is pretty tedious. Pushing it a bit is fun, and actually seems to help,” says Heard. He plans to nominate the new work for an Ig Nobel — the prize for science that “first makes you laugh, then makes you think”. “That is awesome,” says Stout.