news

Researchers at EPFL have made a breakthrough in understanding how neural network-based generative models perform against traditional data sampling techniques in complex systems, unveiling both challenges and opportunities for AI's future in data generation. The world of artificial intelligence (AI) has recently seen significant advancements in generative models, a type of machine-learning algorithms that "learn" patterns from set of data in order to generate new, similar sets of data.

EPFL engineers have created a device that can efficiently convert heat into electrical voltage at temperatures lower than that of outer space. The innovation could help overcome a significant obstacle to the advancement of quantum computing technologies, which require extremely low temperatures to function optimally.

Scientists at EPFL unveil DiffPALM, an innovative AI method that enhances the prediction of protein interactions and our understanding of biological processes potentially relevant to medical applications. Proteins are the building blocks of life, involved in virtually every biological process. Understanding how proteins interact with each other is crucial for deciphering the complexities of cellular functions, and has significant implications for drug development and the treatment of diseases.

Researchers at EPFL have discovered that by shining different wavelengths of light on a material called magnetite, they can change its state, making it more or less conducive to electricity. This could lead to the development of innovative materials for electronics. Magnetite is the oldest and strongest natural magnet.

Scientists have discovered that in indoor spaces, droplets containing the flu virus will remain infectious for longer when they also contain certain types of bacteria found in our respiratory tract. This finding provides important insight into how respiratory infections are transmitted and can enhance estimates of exposure risk.

Researchers have succeeded in directing floating objects around an aquatic obstacle course using only soundwaves. Their novel, optics-inspired method holds great promise for biomedical applications such as noninvasive targeted drug delivery. In 2018, Arthur Ashkin won the Nobel Prize in Physics for inventing optical tweezers : laser beams that can be used to manipulate microscopic particles.

Scientists at EPFL have developed advanced atom-thin graphene membranes with pyridinic-nitrogen at pore edges, showing unprecedented performance in CO2 capture. It marks a significant stride toward more efficient carbon capture technologies. As the world battles climate change, the need for efficient and cost-effective carbon capture technologies is more urgent than ever.

Behavioral analysis can provide a lot of information about the health status or motivations of a living being. A new technology developed at EPFL makes it possible for a single deep learning model to detect animal motion across many species and environments. This -foundational model-, called SuperAnimal, can be used for animal conservation, biomedicine, and neuroscience research.

Researchers have created a deep learning pipeline for designing soluble analogues of key protein structures used in pharmaceutical development, sidestepping the prohibitive cost of extracting these proteins from cell membranes. Many drug and antibody discovery pathways focus on intricately folded cell membrane proteins: when molecules of a drug candidate bind to these proteins, like a key going into a lock, they trigger chemical cascades that alter cellular behavior.

Combining advanced molecular mapping technologies and AI, researchers have published an open-source 'atlas' in Nature, providing a comprehensive understanding of spinal cord injury biology in mice and paving the way for new therapies. Scientists at EPFL have achieved a significant research milestone in the field of spinal cord injuries-mapping out the cellular and molecular dynamics of paralysis in unprecedented detail with their open-source project 'Tabulae Paralytica' .

In a new study, scientists have uncovered the complex genetic and environmental interactions influencing metabolic health and have validated their findings in actual human data. Metabolic syndrome (MetS) is a health condition characterized by a group of risk factors: high blood pressure, high blood sugar, unhealthy cholesterol levels, and abdominal fat.

Scientists, through the VELOCE project, have clocked the speed of Cepheid stars - "standard candles" that help us measure the size of the universe - with unprecedented precision, offering exciting new insights about them. Photo: RS Puppis , one of the most luminous Cepheid variable star s, rhythmically brightens and dims over a six-week cycle.

A fully edible robot could soon end up on our plate if we overcome some technical hurdles, say scientists involved in RoboFood - an project which aims to marry robots and food. Robots and food have long been distant worlds: Robots are inorganic, bulky, and non-disposable; food is organic, soft, and biodegradable.

Scientists at EPFL have successfully miniaturized a powerful erbium-based biber laser on a silicon-nitride photonic chip. Since typical erbium-based fiber lasers are large and difficult to scale down, the breakthrough promises major advances in optical communications and sensing technologies. Lasers have revolutionized the world since the 60's and are now indispensable in modern applications, from cutting-edge surgery and precise manufacturing to data transmission across optical fibers.

Researchers at EPFL have discovered how networks of neurons in fruit flies transform simple brain signals into coordinated actions. This sheds light on the neural mechanisms underlying complex behaviors for potential application in robotics. Understanding how animals, including humans, transform brain signals into coordinated movements is a fundamental question in neuroscience.

Cancer treatments often struggle with balancing efficacy and side effects. A new study by scientists offers a promising solution using antibody-peptide inhibitor conjugates to target specific cell types and block the activity of cancer-promoting enzymes called cathepsins. Tumor cells often hijack normal physiological processes to support their growth, exploiting proteins that are in charge of essential cell functions.

A group of EPFL and scientists have mapped the biodiversity in forests worldwide. Their data, when combined with climate projections, reveal trends that could support ecosystem conservation and restoration efforts. According to the latest figures from the Food and Agriculture Organization of the United Nations, forests cover just over 4 billion hectares of the Earth's surface, or one-third of its total land.

A new study by scientists shows that age-related decline in motor ability can be countered in fruit flies by enhancing the expression of the protein Trio, suggesting potential treatments for age-related movement decline. As we age, we suffer a noticeable decline in motor ability, which affects our quality of life and independence.

Scientists, in collaboration with researchers at EMBL Heidelberg, have discovered that a group of marine protists (eukaryotic organisms) closely related to animals use open or closed mitosis based on their life cycle stages, suggesting that the way animal cells perform cell division evolved long before animals themselves.

Scientists at EPFL have devised and tested out a new, gamma-noise method for monitoring nuclear reactors non-invasively and from a distance. The new method, tested out on EPFL's CROCUS nuclear reactor, can improve nuclear safety and treaty compliance. Monitoring nuclear reactors around the world to ensure that they comply with international treaties is essential for safety.