Global Intelligence Database

Results for "Physics & Astronomy"

Front Page Only
FOUND 323 MATCHES
#311Medicine & HealthFront Page1 December 2025

Beyond the Scales: The Fight to Halt Cachexia

Affecting up to 80% of cancer patients, cachexia remains a stubborn adversary with limited treatment options. A recent summit in Washington signalled a regulatory pivot: success is no longer measured by muscle mass alone, but by the restoration of daily function and independence.

By Fioretti, von Haehling, Coats, Butler, Del Fabbro, Skipworth, Laird, Anker

#312Environmental ScienceFront Page18 November 2025

Bintan’s Fishermen: Rich Waters but Fragile Lives

On Indonesia’s Bintan Island, small-scale fishermen face a stark paradox: whilst they possess abundant natural resources and equipment, they suffer from severe financial and social instability. A new assessment reveals that low education levels and reliance on debt are undermining their resilience against climate change and pollution.

By Sodri, Rahawarin, Sakina, Sianipar

#313Genetics & Molecular BiologyFront Page10 November 2025

The Mechanical Secret of Cell Death: Chromatin's Biphasic Journey

New research reveals that the nucleus undergoes distinct, mechanically regulated changes during necrosis, a form of cell death. Chromatin, the cell's genetic material, first slows down, then speeds up, a process controlled by the cytoskeleton and later by nuclear swelling and DNA fragmentation. This discovery challenges the view of necrosis as a passive event, showcasing it as an actively programmed process with significant mechanical oversight.

By Wei, Luo, Jiang, Wang, Dou, Li

#314Computer Science & AIFront Page10 November 2025

Unpacking the 'Black Box': Making Neural Network Potentials Understandable

Researchers have developed a method to interpret complex neural network potentials by breaking them down into simpler, n-body interactions. This breakthrough, utilizing explainable AI tools, allows scientists to understand the underlying physics learned by these potentials without sacrificing their predictive power. The approach demonstrates that well-trained neural networks can indeed capture fundamental physical interactions in complex systems.

By Bonneau, Lederer, Templeton, Rosenberger, Giambagli, Müller, Clementi

Page 32 of 33