Advanced wastewater treatment technologies are widely used to remove pollutants and pathogens before treated water is discharged or reused. However, a new study suggests that some of these processes may also create an overlooked environmental risk.

Salk scientists use DNA origami technique to build dye-cycling ORBIT, a method for tracking protein-DNA interactions for hours at single base-pair resolution. Using dye-cycling ORBIT, the team measured the rotation of RNA polymerase during DNA transcription. The method will allow scientists to study molecular movements that were previously impossible to see or study, adding a new dimension to life science research.

How many sharks should a coral reef have? It sounds like a simple question, but almost nowhere on Earth is there a pristine, untouched reef. So there is no yardstick against which to measure shark decline or recovery.

Making two-dimensional (2D) multiferroics, materials that host both ferromagnetic and ferroelectric orders, is technologically challenging because thinning bulk materials results in defects and degradation of functional properties. The realization of 2D magnets and 2D ferroelectrics in van der Waals materials offers a different route: By combining the two, it may be possible to engineer a 2D material that exhibits both orders. Wang et al. successfully implemented this strategy to fabricate a van der Waals heterostructure consisting of the ferromagnet Fe₃GaTe₂ and ferroelectric CuInP₂S₆. The researchers applied small voltages across the heterostructure to modulate the magnetic coercivity of the ferromagnetic layer at room temperature, indicating the presence of magnetoelectric coupling. —Jelena Stajic

Scientists at SLAC National Accelerator Laboratory, alongside two researchers from Penn State, have demonstrated the importance of generating highly reproducible experimental data when building AI models for catalyst investigations, a framework that can help guide the reactions that do things like turn carbon dioxide into valuable fuel.

Researchers have developed a novel AI model that can analyze a routine whole histopathology image and simultaneously predict cancer subtype, specific genetic mutations, and survival outcomes across 32 different solid cancers rather than focusing on a single cancer type. The findings from the study in The American Journal of Pathology, published by Elsevier, highlight the potential of computational pathology to connect routine diagnostic imaging with molecular oncology.