Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog

Get the latest international news and world events from around the world.

Log in for authorized contributors

Ancestry-Associated Performance Variability of Open-Source AI Models for EGFR Prediction in Lung Cancer

Open-source AI models for LungCancer EGFR mutation prediction showed high accuracy overall but reduced performance in Asian patients and pleural samples, indicating the need for broader validation.

Importance Artificial intelligence (AI) models are emerging as rapid, low-cost tools for predicting targetable genomic alterations directly from routine pathology slides. Although these approaches could accelerate treatment decisions in lung cancer, little is known about whether their performance is consistent across diverse patient populations and tissue contexts.

Objective To evaluate the performance and generalizability of 2 open-source AI pathology models for predicting EGFR mutation status in lung adenocarcinoma (LUAD) across independent cohorts and ancestral subgroups.

Design, Setting, and Participants This cohort study included patients with LUAD from 2 cohorts: Dana-Farber Cancer Institute (DFCI) from June 2013 to November 2023, and a European-based trial (TNM-I) from August 2016 to February 2022. All patients had paired next-generation sequencing data and hematoxylin-eosin–stained whole-slide images. In the DFCI cohort, genetic ancestry was inferred using germline genotype data. Data analyses were performed from July 2025 to September 2025.

If alien signals have already reached Earth, why haven’t we seen them?

For decades, scientists have searched the skies for signs of extraterrestrial technology. A study from EPFL asks a sharp question: if alien signals have already reached Earth without us noticing, what should we realistically expect to detect today?

Since the first SETI experiment in 1960, astronomers have scanned the Milky Way for signs of advanced extraterrestrial civilizations. These searches have covered radio waves, optical flashes, and infrared heat signatures.

So far, they have found nothing confirmed. That silence is often explained by saying we have only searched a tiny part of the cosmic landscape. But what if signals did reach Earth and slipped past us?

Silicon quantum processor detects single-qubit errors while preserving entanglement

Quantum computers are alternative computing devices that process information, leveraging quantum mechanical effects, such as entanglement between different particles. Entanglement establishes a link between particles that allows them to share states in such a way that measuring one particle instantly affects the others, irrespective of the distance between them.

Quantum computers could, in principle, outperform classical computers in some optimization and computational tasks. However, they are also known to be highly sensitive to environmental disturbances (i.e., noise), which can cause quantum errors and adversely affect computations.

Researchers at the International Quantum Academy, Southern University of Science and Technology, and Hefei National Laboratory have developed a new approach to detect these errors in a silicon-based quantum processor. This error detection strategy, presented in a paper published in Nature Electronics, was found to successfully detect quantum errors in silicon qubits, while also preserving entanglement after their detection.

New catalyst unlocks aluminum’s ability to switch between oxidation states

Aluminum’s journey has been remarkable, going from being more expensive than gold to one of the most widely used materials, from beverage cans to window frames and car parts. Scientists from the Southern University of Science and Technology have added a new feather in aluminum’s cap by expanding its use beyond the metallic form. They created a new aluminum-based redox catalyst —carbazolylaluminylene—that can flip back and forth between two oxidation states: Al(I) and Al(III). This catalyst drove chemical transformations long considered exclusive to transition metals.

This unique feature allowed the team to carry out selective aromatic reactions that bring together three separate alkyne molecules and assemble them into a single benzene ring, resulting in a wide range of benzene derivatives. Carbazolylaluminylene also stood out for its remarkable durability, completing up to 2,290 reaction cycles without losing any catalytic activity. The findings are published in Nature.

Long-term radio observations probe a relativistic binary pulsar system

Astronomers have analyzed the data from long-term radio observations of a binary pulsar known as PSR J1906+0746. Results of the new study, published February 5 on the arXiv pre-print server, deliver important information regarding the nature of this system.

Pulsars are highly magnetized, rotating neutron stars emitting a beam of electromagnetic radiation. They are usually detected in the form of short bursts of radio emission; however, some of them are also observed via optical, X-ray, and gamma-ray telescopes.

Researchers uncover MraZ ‘donut’ deformation that triggers bacterial cell division

A research team led by UAB researcher David Reverter has discovered the molecular mechanism that describes in detail the process regulating cell division in bacteria, based on the binding of the MraZ protein to the dcw gene cluster. The research has been published in Nature Communications.

Cell division is a central process in all living organisms and requires the coordinated action of many proteins and other regulatory elements. In most bacteria, this process is encoded in a gene cluster called the dcw operon, which groups all the genes that produce the proteins necessary to carry out cell division and bacterial wall formation.

These sets of genes are activated by proteins that act as transcription factors: they bind to the promoter region of the gene, the DNA sequence that indicates the point to start transcription, just before the first codon (the basic unit of gene information) that codes for the beginning of the protein sequence. One of these transcription factors is MraZ, the first gene of the dcw operon in all bacteria. When activated, the necessary proteins (encoded within the genes of the operon) are produced so that the bacteria can divide. It is, therefore, the transcription factor that controls the activity of the operon responsible for cell division in most bacteria.

Using light to probe fractional charges in a fractional Chern insulator

In some quantum materials, which are materials governed by quantum mechanical effects, interactions between charged particles (i.e., electrons) can prompt the creation of quasiparticles called anyons, which carry only a fraction of an electron’s charge (i.e., fractional charge) and fractional quantum statistics.

A well-known phenomenon characterized by the emergence of anyons is the so-called fractional quantum Hall effect (FQHE). This effect can emerge in two-dimensional (2D) electron gases under strong magnetic fields and is marked by quantum states in which electrons strongly interact with each other.

Recent studies showed that a similar effect can also arise in the absence of magnetic fields, known as fractional quantum anomalous Hall (FQAH) effect, in quantum phases of matter fractional Chern insulators (FCIs). The FQAH effect was realized for the first time using bilayer twisted molybdenum ditelluride (tMoTe₂)—a moiré superlattice that has a characteristic lattice pattern and a slight twist angle between constituent layers.

Physicists explain the exceptional energy-harvesting efficiency of perovskites

Despite being riddled with impurities and defects, solution-processed lead-halide perovskites are surprisingly efficient at converting solar energy into electricity. Their efficiency is approaching that of silicon-based solar cells, the industry standard. In a new study published in Nature Communications, physicists at the Institute of Science and Technology Austria (ISTA) present a comprehensive explanation of the mechanism behind perovskite efficiency that has long perplexed researchers.

How can a device assembled with minimal sophistication rival state-of-the-art technology perfected over decades? Over the past 15 years, materials research has witnessed the rise of lead-halide-based perovskites as prospective next-generation solar-cell materials. The puzzle is that despite similar performance, perovskite solar cells are fabricated using inexpensive solution-based techniques, while the industry-standard silicon cells require ultra-pure single-crystal wafers.

Now, postdoc Dmytro Rak and assistant professor Zhanybek Alpichshev at the Institute of Science and Technology Austria (ISTA) have uncovered the mechanism behind the unique photovoltaic properties of perovskites. Their key finding is that while silicon-based technology relies on the absence of impurities, the opposite is true in perovskites: It is the natural network of structural defects in these materials that enables the long-range charge transport necessary for efficient photovoltaic energy harvesting.

New nanohole-based microscopy monitors electrochemical reactions millisecond by millisecond

Many technological applications, such as sensors and batteries, greatly rely on electrochemical reactions. Improving these technologies depends on understanding how electrochemical reactions work. However, most current methods cannot look at electrochemical reactions in detail.

Scientists at Utrecht University have now developed a new method that overcomes this limitation. This provides a powerful new way to study and improve electrochemical processes. The study is published in PNAS.

Hydrogen production by water electrolysis is one example where electrochemical reactions at electrodes matter for sustainable technology. But the decisive steps happen within just a few nanometers of the electrode surface, which is too small for most conventional methods to resolve.

/* */