News from Science China Press

Traditional air quality models struggle with changing emissions, while pure AI systems operate as opaque "black boxes" that can break the laws of physics. To solve this, researchers developed ProNet, a framework that embeds AI directly into atmospheric equations. Tested across 338 cities, ProNet outperformed experienced human forecasters and accurately predicted extreme events—like severe dust storms—without producing "AI hallucinations." Crucially, it also acts as a scientific diagnostic tool, using routine hourly data to uncover and correct hidden flaws in existing chemical models.

Using satellite observations, researchers tracked water storage recovery after 468 severe droughts in large river basins worldwide. Most events were followed by recovery, but persistent deficits clustered in drylands and some inland basins. Recovery trajectories varied across basins and events, reflecting water availability after drought. Recurrent droughts and intensive human water use were associated with prolonged deficits.

Carbonate cerium and neodymium isotope records from South China show that low-oxygen waters invaded shallow habitats twice during the Late Ordovician Mass Extinction, with a brief recovery between the two pulses. Earth-system models suggest that rapid climate swings reorganized continental weathering and nutrient delivery, amplifying marine oxygen loss and ecological stress.

Opto-electronic integration organically combines the high-speed modulation and transmission of light with the precise control of electricity, serving as a core enabler for the practical deployment of photonic quantum devices. A recent review article published in *National Science Review* by a joint research team from the University of Science and Technology of China, Peking University, National University of Defense Technology, and Nanjing University systematically summarizes this field. The research team reviews the progress of on-chip opto-electronic integrated technologies, introduces their applications in quantum light sources, quantum frequency conversion, quantum communication, and quantum computing, analyzes existing challenges, and prospects future development pathways. This work helps further optimize on-chip photonic quantum hardware schemes and provides references for the subsequent engineering development of scalable photonic quantum systems.

Early warning alone cannot stop an outbreak. A new Perspective in Science Bulletin examines why signals from emerging infectious diseases can fail to trigger timely public health action, and proposes a practical pathway linking rapid verification, context-sensitive risk assessment, predefined action triggers and operational readiness.

Spinal muscular atrophy (SMA) is a severe neuromuscular disease caused by insufficient SMN protein, yet the critical downstream mechanisms underlying its pathogenesis remain unclear. We previously identified that splicing of the conserved minor intron of Pcyt2 (Phosphocholine cytidylyltransferase 2) depends on sufficient SMN protein. Here, we reveal that simply replacing the minor intron of Pcyt2 with a major intron is sufficient to reverse SMA-associated phenotypes. Based on our discovery that Pcyt2 is specifically expressed in certain neuroglial subtypes, we further show that restoring Pcyt2 in these subtypes significantly alleviates SMA-associated phenotypes in both Drosophila and mouse models. These findings identify Pcyt2 minor intron mis-splicing as a key driver of SMA, offering opportunities for new therapeutic and diagnostic strategies.