In a world where urban traffic congestion and environmental concerns are escalating, innovative solutions are crucial for creating sustainable and efficient transportation systems. A groundbreaking study led by Hongbin Ma from the National Key Lab of Autonomous Intelligent Unmanned Systems at the School of Automation, Beijing Institute of Technology, introduces a pioneering approach that combines Long Range (LoRa) technology with distributed machine learning to optimize network connectivity for green and intelligent transportation systems.

A powerful AI model called Deep Novel Mutation Search (DNMS) predicts virus mutations more accurately and efficiently than traditional, time-consuming lab experiments. Focused on the SARS-CoV-2 spike protein, the model uses a specialized protein language model fine-tuned to understand the virus's specific “language.” DNMS can predict mutations that cause small, functional changes – crucial for viruses like SARS-CoV-2, which evolve through subtle adjustments to maintain function. This approach promises to enhance virus tracking and public health by predicting mutations more accurately and quickly.

Dartmouth researchers conducted the first clinical trial of an AI-powered therapy chatbot and found that, on average, people with diagnosed mental disorders experienced clinically significant improvements in their symptoms over eight weeks, according to results published in NEJM AI, a journal from the publishers of the New England Journal of Medicine. Users engaged with the software, known as Therabot, through a smartphone app and reported that interactions were comparable to working with a mental-health professional. The researchers conclude that while AI-powered therapy is in critical need of clinician oversight, it has the potential to provide real-time support for the many people who lack regular or immediate access to a professional.

A team led by researchers from the Institute of Industrial Science, The University of Tokyo, has discovered brand new interference patterns in twisted two-dimensional tungsten ditelluride lattices. These so-called moiré patterns can be tuned to look like periodic spots or even one-dimensional bands by adjusting the twist angle between layers, and they can drastically alter the physical properties of the material.

Imagine navigating a virtual reality with contact lenses or operating your smartphone under water: This and more could soon be a reality thanks to innovative e-skins. A research team led by the Helmholtz-Zentrum Dresden-Rossendorf (HZDR) has developed an electronic skin that detects and precisely tracks magnetic fields with a single global sensor. This artificial skin is not only light, transparent and permeable, but also mimics the interactions of real skin and the brain, as the team reports in the journal Nature Communications (doi:/10.1038/s41467-025-56805-x).

A recently published article in the journal Weed Science demonstrates just how devasting Palmer amaranth (Amaranthus palmeri) can be for furrow-irrigated rice production, even in small amounts. The findings show that successful, in-furrow rice production greatly hinges on Palmer amaranth-free fields – before, during and after the growing season.