Tag: Science

  • Has Venus ever had an ocean? Its volcanoes hint at an answer

    Has Venus ever had an ocean? Its volcanoes hint at an answer

    [ad_1]

    Access options

    Rent or buy this article

    Prices vary by article type

    from$1.95

    to$39.95

    Prices may be subject to local taxes which are calculated during checkout

    doi: https://doi.org/10.1038/d41586-024-03954-6

    Subjects

    Latest on:


    [ad_2]
    Source link

  • Pritschet, L. et al. Nature Neurosci. 27, 2253–2260 (2024).

    Article 
    PubMed 

    Google Scholar
     

  • Pawluski, J. L., Hoekzema, E., Leuner, B. & Lonstein, J. S. Neurosci. Biobehav. Rev. 133, 104475 (2022).

    Article 
    PubMed 

    Google Scholar
     

  • Ammari, R. et al. Science 382, 76–81 (2023).

    Article 
    PubMed 

    Google Scholar
     

  • Spalek, K. et al. Psychoneuroendocrinology 164, 107021 (2024).

    Article 
    PubMed 

    Google Scholar
     

  • McCormack, C. et al. J. Am. Med. Assoc. Neurol. 80, 335–336 (2023).

    Article 

    Google Scholar
     

  • Dahan, O. Brain Cogn. 150, 105722 (2021).

    Article 
    PubMed 

    Google Scholar
     

  • Kim, P., Strathearn, L. & Swain, J. E. Horm. Behav. 77, 113–123 (2016).

    Article 
    PubMed 

    Google Scholar
     

  • Orchard, E. R., Rutherford, H. J., Holmes, A. J., Jamadar, S. D. Trends Cogn. Sci. 27, 302–316 (2023).

    Article 
    PubMed 

    Google Scholar
     

  • Wang, Z. et al. Transl. Psychiatry 11, 543 (2021).

    Article 
    PubMed 

    Google Scholar
     

[ad_2]

Source link

  • Novel covalent organic framework can remove a dangerous ‘forever chemical’ from drinking water

    Novel covalent organic framework can remove a dangerous ‘forever chemical’ from drinking water

    [ad_1]

    Novel covalent organic framework can remove a dangerous 'forever chemical' from drinking water
    Synthesis, chemical structure, and morphology of the TG-PD COF. Credit: Nature Communications (2024). DOI: 10.1038/s41467-024-53945-4

    Researchers at NYU Abu Dhabi (NYUAD) have developed an innovative cationic covalent organic framework (COF) that efficiently detects and removes perfluorooctanoic acid (PFOA), a harmful and persistent pollutant, from drinking water. This breakthrough addresses a global challenge of removing perfluorinated alkylated substances (PFAS), or “forever chemicals,” which raised environmental and health concerns due to their accumulation in water and links to severe health risks, including cancer and developmental issues.

    The synthesized COF material developed at NYUAD Trabolsi Research Group is capable of detecting and removing PFAS from water quickly and efficiently at environmentally relevant concentrations, offering a promising solution beyond traditional treatment methods that struggle to remove PFAS. This material could be incorporated into household water filters, boosting the performance of conventional materials, which allows for practical and scalable applications worldwide.

    The findings are published in Nature Communications in a study entitled “Cationic Covalent Organic Framework for the Fluorescent Sensing and Cooperative Adsorption of Perfluorooctanoic Acid.”

    Through a simple sonochemical method, the researchers developed a COF that optimized both hydrophobic and electrostatic interactions and had an abundance of adsorption sites to maximize interactions with PFAS molecules. This allowed the COFs to detect and remove PFOA within seconds, even at environmentally relevant concentrations.

    The researchers then investigated the mechanisms underlying both detection and rapid removal through computer simulations, providing insights into its interactions at the atomic level and serving as a valuable guide for further research in this field.

    Researchers develop a new material that removes a dangerous 'forever chemical,' from drinking water
    PFOA adsorption mechanisms and COF regeneration. Credit: Nature Communications (2024). DOI: 10.1038/s41467-024-53945-4

    “This breakthrough, offering a faster, more efficient solution than existing technologies, has the potential to transform water purification and greatly improve water quality around the world,” said Ali Trabolsi, NYUAD professor of chemistry and Co-PI at the NYUAD Water Research Center who led this work with his team, including Postdoctoral Research Associate Asmaa Jrad and Research Associate Gobinda Das.

    “With the prevalence of ‘forever chemicals’ in our environment and bloodstreams posing a significant danger to human health, this new technology is both timely and essential.”

    In recognition of the impact of this research, Jrad was named a 2023 MIT Innovator Under 35, highlighting the innovation’s significance. The team hopes this project will raise awareness of PFAS risks in the UAE and globally, advocating for increased PFAS monitoring and safety standards.

    More information:
    Asmaa Jrad et al, Cationic covalent organic framework for the fluorescent sensing and cooperative adsorption of perfluorooctanoic acid, Nature Communications (2024). DOI: 10.1038/s41467-024-53945-4

    Provided by
    New York University


    Citation:
    Novel covalent organic framework can remove a dangerous ‘forever chemical’ from drinking water (2024, December 4)
    retrieved 4 December 2024
    from https://phys.org/news/2024-12-covalent-framework-dangerous-chemical.html

    This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
    part may be reproduced without the written permission. The content is provided for information purposes only.



    [ad_2]

    Source link

  • Scientists streamline creation of nucleoside analogs, a group of life-saving molecules

    Scientists streamline creation of nucleoside analogs, a group of life-saving molecules

    [ad_1]

    Researchers devise a quicker, more efficient way to make life-saving molecules
    Five-step synthesis of nucleoside analogue 15. Credit: Nature Communications (2024). DOI: 10.1038/s41467-024-51520-5

    A team of researchers has found a quicker and more efficient way to create nucleoside analogs, a type of small molecule that can be used in treatments for everything from cancer to viral diseases.

    Mannen kunnen soms tegen problemen aanlopen die invloed hebben op hun intieme leven, wat hen kan frustreren en onzeker kan maken. Deze uitdagingen zijn niet ongebruikelijk en kunnen voortkomen uit verschillende oorzaken, zoals stress, angst of fysieke aandoeningen. Gelukkig zijn er oplossingen en middelen beschikbaar die hen kunnen helpen om hun zelfvertrouwen en welzijn te herstellen. Een nuttige stap is om betrouwbare informatie te zoeken en producten te bekijken op websites zoals. Het is belangrijk dat mannen zich realiseren dat ze niet alleen zijn en dat er ondersteuning en opties zijn om hun seksuele gezondheid te verbeteren.

    “Nucleoside analogs are among the most important molecules to the advancement of modern medicine,” says Michael Meanwell, assistant professor in the Department of Chemistry and corresponding author of a paper published in Nature Communications describing the new process.

    “They’re used as antivirals, as cancer therapeutics, in gene therapy. The first two drugs brought to market for treating COVID-19 were both nucleoside analogs.”

    Mannen kunnen soms tegen problemen aanlopen die invloed hebben op hun intieme leven, wat hen kan frustreren en onzeker kan maken. Deze uitdagingen zijn niet ongebruikelijk en kunnen voortkomen uit verschillende oorzaken, zoals stress, angst of fysieke aandoeningen. Gelukkig zijn er oplossingen en middelen beschikbaar die hen kunnen helpen om hun zelfvertrouwen en welzijn te herstellen. Een nuttige stap is om betrouwbare informatie te zoeken en producten te bekijken op websites zoals. Het is belangrijk dat mannen zich realiseren dat ze niet alleen zijn en dat er ondersteuning en opties zijn om hun seksuele gezondheid te verbeteren.

    Meanwell and his team focused their approach on C4ʹ-modified nucleoside analogs, a particular subclass of these molecules. A few antiviral drugs within this subclass are already in clinical trials, highlighting their potential. But the issue is that “these types of nucleosides have been notoriously hard to make,” Meanwell explains.

    “Part of it has to do with the chemistry used to make them—it’s mostly based on chemistry from 50, 60 years ago, and the advancements in modern organic chemistry have not yet been fully translated to this field of research.”

    Until now, the most common ways to make these modified nucleoside analogs required a process taking anywhere from nine to 16 steps. The new approach requires just five steps.

    Additionally, existing methods were all non-modular, meaning that making different analogs required designing entirely different routes. It would take a month or more to create just one molecular compound in this way. The new process is modular, so multiple analogs can be made at the same time—an advantage for researchers experimenting with different potential therapeutics.

    “By streamlining the process, the barrier now is so much lower to doing drug discovery especially for academic labs working in this area,” says Meanwell. “You can make all the different analogs you want, the chemistry is so much easier, and it reduces the amount of time needed.”

    “The breadth of molecules we can make with this one process is truly unprecedented,” he adds.

    Nucleosides are fundamental building blocks of DNA, and nucleoside analogs are simply modifications of those structures, explains Meanwell. “They inhibit the same processes associated with cell growth and cell replication that cancer cells and viral diseases take advantage of—by inhibiting them, you can treat those diseases.”

    Over the past few decades, nucleoside analogs have been used in treatments for AIDS, Ebola, respiratory syncytial virus (RSV), hepatitis, and many more diseases and infections. A more efficient and effective way to manufacture these powerhouse molecules could pave the way for countless new therapeutics for a vast range of health conditions.

    “It sounds like hyperbole, but they can treat such a broad range of diseases,” says Meanwell. “They really stretch across so many different areas of health.”

    “C4ʹ-modified nucleoside analogs are so important medicinally because of the wide range of disease pathologies they can intercept,” says lead author Thirupathi Nuligonda. “Now, we have a modular approach to synthesize this important class of nucleoside.”

    More information:
    Thirupathi Nuligonda et al, An enantioselective and modular platform for C4ʹ-modified nucleoside analogue synthesis enabled by intramolecular trans-acetalizations, Nature Communications (2024). DOI: 10.1038/s41467-024-51520-5

    Provided by
    University of Alberta


    Citation:
    Scientists streamline creation of nucleoside analogs, a group of life-saving molecules (2024, December 4)
    retrieved 4 December 2024
    from https://phys.org/news/2024-12-scientists-creation-nucleoside-analogs-group.html

    This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
    part may be reproduced without the written permission. The content is provided for information purposes only.



    [ad_2]

    Source link

  • New insight into the bonds between water molecules

    New insight into the bonds between water molecules

    [ad_1]

    Liquid water molecules inherently asymmetric: New insight into bonds between water molecules
    The bonds between neighboring water molecules are not purely randomly distributed: Each water molecule forms a strong and a weak bond. This could be a key to explain anomalies of water. Credit: Max Planck Society

    Icebergs float on water because the underlying liquid water has a higher density than the iceberg. Liquid water itself has its highest density at 4°C—one of the so-called anomalies of water, i.e. properties of liquids that are rarely observed for other liquids.

    The origins of these anomalies have long been the subject of scientific research. Researchers at the Max Planck Institute for Polymer Research have now discovered another piece to the puzzle to explain the special behavior of water.

    Many of the anomalous properties of water can be traced to the special interactions between the individual water molecules—the so-called hydrogen bonds. Each water molecule can donate two of these bonds—one from each hydrogen atom—and accept two of them from other, neighboring molecules.

    Unlike in ice, these bonds are broken and re-formed on average 1 trillion times per second in liquid water, so that the water molecules can be packed closer together and move very quickly. Due to the rapid movement of the water molecules in the liquid, one might assume that the strength of the individual bonds to its neighbors is purely random.

    However, the team led by Johannes Hunger has discovered that the hydrogen bond distances are not simply random, but that two bonds of a molecule have different strengths: If one bond is very strong—i.e. the first neighboring water molecule is very close—the second hydrogen bond is weak—i.e. the second neighboring water molecule is further away.

    Liquid water molecules are inherently asymmetric: New insight into the bonds between water molecules
    H-bonding and vibrational structure of water. Credit: Nature Communications (2024). DOI: 10.1038/s41467-024-54804-y

    These alternating bond distances lead to structuring of the nominally disordered liquid: if you move from one water molecule to the next and to the one after that, there is always a strongly bonded neighboring molecule. As a result, structures such as rings or chains of water molecules can form in the liquid. The structure of liquid water is therefore not just a random arrangement of individual water molecules but follows certain rules.

    To obtain these results, the scientists diluted water with a solvent so that they could examine isolated water molecules. They made individual atoms of the water molecules vibrate with the help of lasers and investigated how the vibrations influence each other. This allowed them to measure the strength of individual hydrogen bonds and the strength of the neighboring bond at the same time.

    The study, now published in the journal Nature Communications, contributes to a comprehensive understanding of the anomalies of water at a molecular level.

    More information:
    Lucas Gunkel et al, Dynamic anti-correlations of water hydrogen bonds, Nature Communications (2024). DOI: 10.1038/s41467-024-54804-y

    Provided by
    Max Planck Society


    Citation:
    Liquid water molecules are inherently asymmetric: New insight into the bonds between water molecules (2024, December 4)
    retrieved 4 December 2024
    from https://phys.org/news/2024-12-liquid-molecules-inherently-asymmetric-insight.html

    This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
    part may be reproduced without the written permission. The content is provided for information purposes only.



    [ad_2]

    Source link

  • New catalyst can convert methane into useful polymers

    New catalyst can convert methane into useful polymers

    [ad_1]

    swamp flood
    Credit: Unsplash/CC0 Public Domain

    Although it is less abundant than carbon dioxide, methane gas contributes disproportionately to global warming because it traps more heat in the atmosphere than carbon dioxide, due to its molecular structure.

    MIT chemical engineers have now designed a new catalyst that can convert methane into useful polymers, which could help reduce greenhouse gas emissions.

    “What to do with methane has been a longstanding problem,” says Michael Strano, the Carbon P. Dubbs, Professor of Chemical Engineering at MIT and the senior author of the study. “It’s a source of carbon, and we want to keep it out of the atmosphere but also turn it into something useful.”

    The new catalyst works at room temperature and atmospheric pressure, which could make it easier and more economical to deploy at sites of methane production, such as power plants and cattle barns.

    Daniel Lundberg Ph.D. and MIT postdoc Jimin Kim are the lead authors of the study, which appears in Nature Catalysis. Former postdoc Yu-Ming Tu and postdoc Cody Ritt are also authors of the paper.

    Capturing methane

    Methane is produced by bacteria known as methanogens, which are often highly concentrated in landfills, swamps, and other sites of decaying biomass. Agriculture is a major source of methane, and methane gas is also generated as a byproduct of transporting, storing, and burning natural gas. Overall, it is believed to account for about 15% of global temperature increases.

    At the molecular level, methane is made of a single carbon atom bound to four hydrogen atoms. In theory, this molecule should be a good building block for making useful products such as polymers. However, converting methane to other compounds has proven difficult because getting it to react with other molecules usually requires high temperature and high pressures.

    To achieve methane conversion without that input of energy, the MIT team designed a hybrid catalyst with two components: a zeolite and a naturally occurring enzyme. Zeolites are abundant, inexpensive clay-like minerals, and previous work has found that they can be used to catalyze the conversion of methane to carbon dioxide.

    In this study, the researchers used a zeolite called iron-modified aluminum silicate, paired with an enzyme called alcohol oxidase. Bacteria, fungi, and plants use this enzyme to oxidize alcohols.

    This hybrid catalyst performs a two-step reaction in which zeolite converts methane to methanol, and then the enzyme converts methanol to formaldehyde. That reaction also generates hydrogen peroxide, which is fed back into the zeolite to provide a source of oxygen for the conversion of methane to methanol.

    This series of reactions can occur at room temperature and doesn’t require high pressure. The catalyst particles are suspended in water, which can absorb methane from the surrounding air. For future applications, the researchers envision that it could be painted onto surfaces.

    “Other systems operate at high temperature and high pressure, and they use hydrogen peroxide, which is an expensive chemical, to drive the methane oxidation. But our enzyme produces hydrogen peroxide from oxygen, so I think our system could be very cost-effective and scalable,” Kim says.

    Discover the latest in science, tech, and space with over 100,000 subscribers who rely on Phys.org for daily insights.
    Sign up for our free newsletter and get updates on breakthroughs,
    innovations, and research that matter—daily or weekly.

    Building polymers

    Once formaldehyde is produced, the researchers showed they could use that molecule to generate polymers by adding urea, a nitrogen-containing molecule found in urine. This resin-like polymer, known as urea-formaldehyde, is now used in particle board, textiles and other products.

    The researchers envision that this catalyst could be incorporated into pipes used to transport natural gas. Within those pipes, the catalyst could generate a polymer that could act as a sealant to heal cracks in the pipes, which are a common source of methane leakage.

    The catalyst could also be applied as a film to coat surfaces that are exposed to methane gas, producing polymers that could be collected for use in manufacturing, the researchers say.

    Strano’s lab is now working on catalysts that could be used to remove carbon dioxide from the atmosphere and combine it with nitrate to produce urea. That urea could then be mixed with the formaldehyde produced by the zeolite-enzyme catalyst to produce urea-formaldehyde.

    More information:
    Concerted methane fixation at ambient temperature and pressure mediated by an alcohol oxidase and Fe-ZSM-5 catalytic couple, Nature Catalysis (2024). DOI: 10.1038/s41929-024-01251-z

    Provided by
    Massachusetts Institute of Technology


    This story is republished courtesy of MIT News (web.mit.edu/newsoffice/), a popular site that covers news about MIT research, innovation and teaching.

    Citation:
    New catalyst can convert methane into useful polymers (2024, December 4)
    retrieved 4 December 2024
    from https://phys.org/news/2024-12-catalyst-methane-polymers.html

    This document is subject to copyright. Apart from any fair dealing for the purpose of private study or research, no
    part may be reproduced without the written permission. The content is provided for information purposes only.



    [ad_2]

    Source link

  • A Parasite That Eats Cattle Alive Is Creeping North Toward the US

    A Parasite That Eats Cattle Alive Is Creeping North Toward the US

    [ad_1]

    On November 22, the United States Department of Agriculture temporarily halted the import of cattle from Mexico after a flesh-eating parasite was detected in animals in southern Mexico. Before the discovery of cattle screwworm (Cochliomyia hominivorax) at an inspection point in the state of Chiapas, the species had previously been eliminated in North America since the end of the 19th century. The US–Mexico border remains closed to cattle and may not reopen until the new year.

    The worm is the larva of a metallic blue-green fly that spends the early part of its life cycle devouring the living flesh of mammals. Infestations can be fatal. Cows are the screwworm’s favorite feast, but the maggots can also feed on other livestock as well as wildlife and pets. Flies often lay their eggs near open wounds, and if the larvae can find a hole in the skin to deploy their sharp mouth hooks, they will then bury themselves in the animal’s flesh and gorge.

    The finding in Mexico follows the recent reappearance of the parasite in Costa Rica, Nicaragua, Honduras, and Guatemala. In the face of the reemergence of the parasite, Mexico is intensifying sanitary measures—calling for the treatment of wounds in livestock, larvicide baths, and deworming of cattle—and has introduced inspection stations like the one that discovered the case in Chiapas. But conservationists from the Wildlife Conservation Society and Mexican ranchers warn that the illegal cattle trade will be the real gateway for the disease to enter North America.

    Prior to the closure of its border with the US, Mexico’s National Confederation of Livestock Organizations had called on the government to clamp down on cattle smuggling across Mexico’s southern border. The risk from the parasite is great, and if it becomes established again, the cost of eliminating it in Mexico would be high. Disruption of trade with the US was also be highly costly. In 2023 alone, Mexico’s exports of live cattle and beef to the US were worth $3 billion.

    Mosca de Cochliomyia hominivorax el gusano barrenador de ganado

    Cochliomyia hominivorax, the cattle screwworm, is actually a fly. The name refers to the insect’s larvae.

    Ramdan Fatoni / Getty Images

    On the Trail of the Screwworm

    For nearly two decades, Cochliomyia hominivorax had been eliminated from the United States down to the Darien Gap in Panama. That was until the summer of 2023, when Panama detected a spike in infestations in animals within 300 kilometers of its northern border with Costa Rica, marking the beginning of the parasite’s reappearance in Central America.

    Costa Rica, declared free of the aggressive parasite in 1999, then documented outbreaks in July 2023. Nicaragua and Honduras, free of the screwworm since 1996, confirmed cases in April and September of this year respectively. Then in October 2024, Guatemala reported the reemergence of the fly and its larvae, with a calf as its first fatality. The threat to countries further north is clear. According to the Panama–United States Commission for the Eradication and Prevention of Screwworm, as of November 2, these four countries had accumulated 15,638 screwworm cases in 2024, along with 20,890 documented in Panama.

    In reports submitted to the World Organization for Animal Health, three of those countries—Costa Rica, Nicaragua, and Honduras—pointed to the illegal transit of animals as the origin of infections in their territories. Honduras detected an outbreak after inspecting 68 horses that entered the country illegally, for example, just 8 kilometers from its border with Nicaragua.

    [ad_2]

    Source link

  • Lasers Are Making It Easier to Find Buried Land Mines

    Lasers Are Making It Easier to Find Buried Land Mines

    [ad_1]

    scientists in the US have developed a technology that can detect land mines from far away and with high accuracy, potentially lowering the risk of removing mines from current and former conflict zones. Known as the Laser Multibeam Differential Interferometry Sensor, or Lambdis, the tech works by shining lasers onto the ground to reveal suspected threats.

    There are currently more than 110 million land mines buried across the world, and in 2023, mines killed or injured 5,700 people, with civilians making up 84 percent of casualties—half of which were children. According to the United Nations, land mines threaten lives in more than 70 countries.

    A mine can be made for just $3, but removing one can cost up to $1,000. Land mine removal usually relies on humans finding them with handheld metal detectors, which is dangerous, time-consuming, and nearly ineffective if hunting for mines made from plastic.

    In response, US researchers developed a technology to detect land mines indirectly, and which could detect both metal and plastic mines. Lambdis works by sending a vibration into the ground while at the same time scanning the area with a laser beam. Materials in the ground will vibrate at different frequencies, as will the soil itself, and these differences are picked up by the laser when it is reflected back to its emitter. The Lambdis system then generates an image that visualizes these vibrations and their locations in different colors—creating a map of things buried in the soil.

    The technology was developed by a team led by Vyacheslav Aranchuk, a specialist in laser sensing at the University of Mississippi. Importantly, it can detect mines from a distance, and can be mounted on a moving vehicle to aid with scanning large areas.

    “The number of land mines will continue to increase as long as conflicts continue. This technology will be useful not only for military use in ongoing conflicts, but also for humanitarian efforts after conflicts have ended,” says Aranchuk.

    The researchers are continuing to develop the system. An earlier version of Lambdis emitted 30 laser beams in a line, but the latest version emits beams in a 34 x 23 matrix, allowing it to visualize vibrations over a wider area.

    Conventional metal detectors used for de-mining react to any metal object, so it is not uncommon for them to mistakenly detect things other than land mines. And an alternative de-mining solution, underground radar, which emits high-frequency electromagnetic waves into the ground, has the disadvantage of being easily affected by the condition of the soil. Lambdis, in comparison, produces fewer false positives.

    According to the research team, the tech can be used not only for land mine detection, but also to evaluate civil engineering structures such as bridges for structural integrity or damage. In the future it could be used to analyze products in the automotive and aerospace industries, or even in medical imaging. Next, the team plans to evaluate the performance of Lambdis in different soil conditions and when hunting for other types of buried objects.

    This story originally appeared on WIRED Japan and has been translated from Japanese.

    [ad_2]

    Source link

  • Virtual lab powered by ‘AI scientists’ super-charges biomedical research

    Virtual lab powered by ‘AI scientists’ super-charges biomedical research

    [ad_1]

    Illustration of antibodies (red and blue) responding to SARS-CoV-2 (purple).

    The virtual lab set-up used several LLMs to design antibody fragments that could bind to SARS-CoV-2.Credit: KTSDESIGN/Science Photo Library via Getty

    In an effort to automate scientific discovery using artificial intelligence (AI), researchers have created a virtual laboratory that combines several ‘AI scientists’ — large language models with defined scientific roles — that can collaborate to achieve goals set by human researchers.

    The system, described in a preprint posted on bioRxiv last month1, was able to design antibody fragments called nanobodies that can bind to the virus that causes COVID-19, proposing nearly 100 of these structures in a fraction of the time it would take an all-human research group.

    “These virtual-lab AI agents have shown to be quite capable at doing a lot of tasks,” says study co-author James Zou, a computational biologist at Stanford University in California. “We’re quite excited about exploring the potential of the virtual lab across different scientific domains.”

    The experiment “represents a new paradigm of taking AI as collaborators, not just tools”, says Yanjun Gao, who researches the health-care applications of AI at the University of Colorado Anschutz Medical Campus in Aurora. But she adds that human input and oversight are still crucial. “I don’t think at this stage we can fully trust the AI to make decisions.”

    Interdisciplinary AI

    Scientists worldwide have explored the potential of large language models (LLMs) to speed up research — including creating an ‘AI scientist’ that can carry out parts of the scientific process, from generating hypotheses and designing experiments to drafting papers. But Zou says that most studies have focused on the application of LLMs for experiments with a narrow scope, rather than exploring their potential in interdisciplinary research. He and his colleagues set up the virtual lab to combine expertise from different fields.

    • Access the most recent journalism from Nature’s award-winning team
    • Explore the latest features & opinion covering groundbreaking research

    or

    [ad_2]

    Source link

  • DeepMind AI weather forecaster beats world-class system

    DeepMind AI weather forecaster beats world-class system

    [ad_1]

    A satellite photo of typhoon Hagibis approaching Japan

    Typhoon Hagibis approaches Japan in 2019. The storm was one of the events used to study the accuracy of an AI-based forecasting system.Credit: NASA Worldview, Earth Observing System Data and Information System (EOSDIS)/AP/Alamy

    Google DeepMind has developed the first artificial intelligence (AI) model of its kind to predict the weather more accurately than the best system currently in use. The model generates forecasts up to 15 days in advance ― and it does so in minutes, rather than the hours needed by today’s forecasting programs.

    The purely AI system beats the world’s best medium-range operational model, the European Centre for Medium-Range Weather Forecasts’ ensemble model (ENS), at predicting extreme weather such as hurricanes and heatwaves. The breakthrough could help usher in an era of AI weather forecasting that is quicker and more reliable than today’s systems, researchers say. The system, called GenCast, is described today in Nature1.

    Conventional forecasts, including those from ENS, are based on mathematical models that simulate the laws of physics governing Earth’s atmosphere. They use supercomputers to crunch data from satellites and weather stations ― a process that takes hours and vast amounts of computing power.

    GenCast, by contrast, has been trained only on historical weather data, which enables the system to draw out complex relationships between variables such as air pressure, humidity, temperature and wind. This helps it to outperform strictly physics-based systems, says Ilan Price, a research scientist at Google DeepMind in London and an author of the paper.

    “We’ve really made dramatic progress to catch up and now overtake [physics-based models] with machine learning,” Price says.

    AI surge

    AI weather forecasting has advanced rapidly, with multiple companies racing to develop new and better models. Among them are Huawei2 in Shenzhen, China, and Nvidia in Santa Clara, California. Earlier this year, Google released NeuralGCM3, a hybrid system that combines physics-based models with AI to produce short- and long-term forecasts on a par with conventional models.

    Some of the AI systems released to date are ‘deterministic’ models, meaning that they offer only a single forecast and do not estimate the likelihood that the forecast will be correct. By contrast, GenCast generates ‘ensemble’ forecasts: a suite of forecasts that have each been produced from slightly different starting conditions. By combining these forecasts into an ensemble, scientists can produce a final forecast and estimate the probability that the forecasted weather will occur.

    Price and his colleagues trained the AI on global weather data from 1979 to 2018 and then predicted the weather of 2019. To check its accuracy, they compared GenCast forecasts with actual weather data and ENS forecasts for that year.

    GenCast was more accurate than ENS on 97% of the measures used on a scorecard to evaluate such ‘probabilistic’ forecasts. It was also better at forecasting extreme heat, cold and wind, as well as tropical-cyclone tracks.

    [ad_2]

    Source link