Huairou Science City Sci-Tech Innovation Update (Issue No. 20)

Date:2025-12-18 Source:Huairou Science City

PKU’s School of Physics Uses AI to Empower Atomic Force Microscopy and Reveals “Amorphous Ice Layer” Hidden Before Ice Melting

Recently, the research team led by Professor XU Limei, Professor JIANG Ying, and Distinguished Researcher HONG Jiani from the International Center for Quantum Materials, School of Physics, Peking University, has developed an AI-AFM combined characterization and analysis method. Through this method, they discovered that ice forms a previously unknown “amorphous ice layer” before melting. Although this ice layer has a disordered structure, it still maintains solid characteristics. This discovery not only redefines the “melting” process of ice but also provides a new physical perspective for understanding the growth, reconstruction, and interface reactions of ice. This research provides an atomic-level tool for exploring complex disordered interfaces, phase transition behaviors, and material defects, and is expected to bring significant applications in fields such as catalytic interfaces, functional materials, and biomolecular research. The relevant results have been published in Physical Review X.

Source: School of Physics, Peking University

UCAS Makes New Progress in Research on Plant Stem Cell Fate Determination

Recently, the research team led by YANG Weibing, a doctoral supervisor at the University of Chinese Academy of Sciences (UCAS) and a researcher at the CAS Center for Excellence in Molecular Plant Sciences, has found that the dynamic changes in cell wall structure can be regarded as a core switch regulating stem cell fate, governing the transition between states such as stem cell division and differentiation. This study reveals the mechanism of plant stem cell fate determination and discovers a brand-new gene expression regulation mode—mRNA nuclear retention. It indicates that the cell wall not only plays the role of a “exoskeleton” for plant cells but also acts as a “commander,” influencing and guiding the fate of plant stem cells. Based on the “precise cell wall design” strategy, it is expected to enhance the activity of crop meristems and yield potential, providing theoretical support and technical approaches for breeding high-yield and high-efficiency crops. The relevant research results have been published in Science.

Source: Center for Excellence in Molecular Plant Sciences, Chinese Academy of Sciences

Institute of Tibetan Plateau Research Reveals the Dual-Drive Mechanism of MKK3 in Barley Seed Dormancy

Recently, the Group of Alpine Paleoecology and Human Adaptation from the Institute of Tibetan Plateau Research (ITP), CAS, in collaboration with dozens of top international scientific research institutions, has jointly solved the problem of what determines “how long barley seeds sleep.” The research team found that the MKK3 (Mitogen-Activated Protein Kinase 3) gene shapes the dormancy rhythm of barley in different climatic regions around the world through a “copy number + kinase activity” dual-drive mechanism. This achievement provides operable molecular modules for crop stress-resistant breeding and supports agricultural sustainable development under global climate change conditions. This research received technical support from the Earth System Science Numerical Simulator Facility (EarthLab). The relevant research results have been published in Science.

Source: Institute of Tibetan Plateau Research, Chinese Academy of Sciences

Institute of Atmospheric Physics and Collaborators Quantitatively Evaluate the Synergistic Impact of High Temperature and Drought on the Probability of Extreme Wildfires

Recently, the State Key Laboratory of Earth System Numerical Modeling and Application, Institute of Atmospheric Physics (IAP), CAS, in collaboration with domestic and foreign teams, has quantitatively evaluated the synergistic impact of high temperature and drought on the probability of extreme wildfires. The study shows that extreme drought increases the probability of events with fire intensity and burned area similar to the 2025 Los Angeles fire by 54% and 75% respectively; while the combined effect of high temperature and drought further increases the above probabilities by 149% (intensity) and 210% (burned area). This study deepens the understanding of the formation mechanism of wildfires in the Los Angeles area and reveals the amplifying effect of high temperature-drought compound events on fire risks. This research received technical support from the Earth System Science Numerical Simulator Facility (EarthLab). The relevant results have been published in Advances in Atmospheric Sciences.

Source: Earth System Science Numerical Simulator Facility (EarthLab)

Institute of Mechanics Proposes AI-Enhanced Parameter Inversion Method for Complex Spatial Gravitational Wave Signals

Recently, the research team from the Gravitational Wave Experiment Center, Institute of Mechanics, Chinese Academy of Sciences (CAS), has proposed a hybrid computational framework called “Flow-Matching Markov Chain Monte Carlo (FM-MCMC).” This framework creatively integrates flow matching technology from generative AI with the traditional parallel tempering MCMC algorithm. By leveraging AI models to quickly capture the global characteristics of the parameter space, the framework successfully overcomes the local extremum trap that is difficult to surmount with traditional methods. It significantly reduces the parameter inference time from several days to hours (on a single GPU), achieving an order-of-magnitude improvement in computational efficiency. More importantly, it ensures the accuracy and reliability of physical parameter inference, providing a feasible solution for the automated processing of spatial gravitational wave observation data and demonstrating the enormous potential of “AI + scientific computing” in solving inversion problems of complex physical systems.

Source: Institute of Mechanics, Chinese Academy of Sciences