Iqbal et al. show that HflK/C conformational dynamics regulate bacterial adaptation to aminoglycoside stress. Using disulfide crosslinking to constrain the closed state, they demonstrate that stabilizing a closed HflK/C assembly impairs stress recovery and reveal a stress-induced conformation with dual openings that may facilitate FtsH-dependent membrane proteolysis.
Nanoprinting imprinting metalenses 100x faster than lithography.
Professor RHO Jun-seok from the Departments of Mechanical Engineering and Chemical Engineering at POSTECH has gained international attention for developing a mass-production process for metalenses and a switchable 2D-3D display technology based on them. The two studies were simultaneously published in the April 30 issue of Nature. This marks the first case in Korea of a researcher publishing two separate papers as corresponding author in the same issue of the journal.
A metalens is a flat optical device that controls light using nanoscale structures rather than curved glass. By replacing bulky glass lenses with engineered surface patterns, optical systems become far thinner and lighter. Because this enables control of light at scales smaller than its wavelength, metamaterials are often regarded as a Nobel Prize–worthy field of research.
The first study addressed a key barrier to commercialization: large-scale manufacturing. Production has so far relied on expensive and complex semiconductor fabrication processes due to the extreme precision required, making it slow, costly, and largely limited to laboratory research. To overcome this, Prof. RHO’s team developed a Roll-to-Roll Nanoimprint process enabling continuous production using a cylindrical roller. Instead of fabricating nanoscale structures one by one on rigid molds, flexible polymer molds were used to imprint patterns onto thin films. This shifts fabrication from a one-at-a-time process to continuous factory-scale production. The team produced over 300 metalenses per second, about 100 times faster than conventional methods, while maintaining consistent performance over a 200-meter process.
As quantum computing moves closer to large-scale deployment, new research is examining its future energy, water, and material demands.
David McCollum, an Oak Ridge National Laboratory distinguished scientist, is leading the project. McCollum is also a joint faculty professor in the Center for Energy, Transportation, and Environmental Policy (CETEP) at the Howard H. Baker Jr. School of Public Policy and Public Affairs at the University of Tennessee, Knoxville. The work aims to inform the rollout of quantum infrastructure over the coming decades. It examines technologies evolving from experimental environments to commercial-scale use. Quantum computing is expected to unlock advances in drug discovery, material science, artificial intelligence, and cybersecurity.
“Quantum computing presents extraordinary opportunities, from accelerating scientific discovery to solving complex optimization problems,” McCollum said. “At the same time, it introduces new questions about the energy, water, and materials required to operate these systems at scale. Our research aims to get ahead of those questions before resource and supply chain constraints start to bite.”
In a small clinical trial, a CAR T-cell therapy—a type of immunotherapy that uses a patient’s own immune cells to fight cancer—shrank tumors in several children and young adults with diffuse midline gliomas. This fast-growing form of brain and spinal cord cancer typically causes death within a year of diagnosis.
In the trial, several participants were still alive 2 years or more after receiving the treatment.
Patients in the trial had a type of diffuse midline gliomas known as H3K27M mutant, a genetic change that is found in about 80% of younger patients with these cancers Exit Disclaimer. Researchers at Stanford University, who led the study, designed the experimental CAR T-cell therapy to target a molecule called GD2 that is produced in large amounts by H3K27M-mutant diffuse midline gliomas.
Manuel Blum (Carnegie Mellon University)
https://simons.berkeley.edu/talks/man…
The Role of TCS in Modern Machine Learning.
We define the Conscious Turing Machine (CTM), a formal global workspace model of consciousness specified as a 7-tuple. Its 10 million processors self-define a multimodal language, Brainish, together with a dictionary of chunks. Each chunk is a 5-tuple that contains and defines a 2-tuple Brainish word.
Our principal contribution is not theorems—though there is one—but theoretical insights into several central puzzles of consciousness. From this formal definition follow a proposed solution to the binding problem, an explanation of how the suffering of pain is generated, and testable predictions derived from the CTM.
A new study shows how the brain abandons outdated strategies and adapts to new rules.
Most people have experienced the feeling: switching from one task to another, only to find the brain momentarily stuck in the old mode of thinking. Sometimes, even after realizing a strategy no longer works, the mind keeps returning to it anyway.
Neuroscientists call the ability to adapt and shift strategies “cognitive flexibility”—a core feature of higher cognition that allows the brain to abandon outdated rules and respond to changing conditions. Impairments in cognitive flexibility are associated with disorders including Attention-Deficit/Hyperactivity Disorder (ADHD), depression, obsessive-compulsive disorder (OCD), schizophrenia, and Alzheimer’s disease.
13 years ago, I sat down with Peter Joseph, musician, filmmaker, and founder of the Zeitgeist Movement.
His argument was simple, and uncomfortable: the system we live under (debt-based money, work-for-survival economics, infinite growth on a finite planet) isn’t broken. It’s working exactly as designed. And it’s running out of runway.
In 2013, this sounded radical. In 2026, it sounds like a weather report.
We covered a lot of ground in 75 minutes: the Resource-Based Economy, the role of Artificial Intelligence in managing scarcity, the schism between Zeitgeist and the Venus Project, sustainability, central planning, and the technological singularity itself.
You don’t have to agree with Peter to take the conversation seriously. I don’t agree with all of it. But the questions he was asking back then are the questions we’re being forced to ask now, except we’re asking them in an era when AI systems can actually do things he could only theorize about.
The technology has caught up with the critique. The philosophy hasn’t caught up with the technology.
A research project led by the Institute for Research in Nutrition and Food Safety (INSA) and the Faculty of Pharmacy and Food Sciences at the University of Barcelona, together with the Molecular Biology Institute of Barcelona (IBMB) of the CSIC (which stands for Consejo Superior de Investigaciones Científicas), has successfully designed and tested a gluten-degrading molecule that is a promising ally in the management of celiac disease, an autoimmune disease whose symptoms are triggered by the consumption of gluten and other prolamins found in cereals.
At present, there is a complete lack of treatment options beyond a diet free from gluten, which is difficult to maintain in Western societies where diets rely heavily on wheat products.
The major breakthrough is that the molecule is effective at very low concentrations and at a pH of 2—the pH of the stomach—a condition that none of the molecules currently available or under development had previously achieved with efficiency. Although some of them are marketed as nutritional supplements, they are not an effective alternative to gluten-free diets.