News from China
Updates every hour. Last Updated: 18-Nov-2025 14:11 ET (18-Nov-2025 19:11 GMT/UTC)
Novel cross-linker developed for concurrent enrichment and quantitative analysis of protein interactions in limited cells
Dalian Institute of Chemical Physics, Chinese Academy SciencesPeer-Reviewed Publication
- Journal
- Angewandte Chemie International Edition
AI designs over 7,000 new proteins to help speed up medicine and research
Higher Education PressPeer-Reviewed Publication
Scientists from Chongqing University and Zhejiang University used AI to design more than 7,000 brand-new proteins that dissolve easily, stay stable under heat, and are ready for lab testing—helping drug and diagnostic companies work faster and cut early-stage development costs.
- Journal
- Frontiers of Computer Science
AI speeds up accurate prediction of molecular properties for drug discovery
Higher Education PressPeer-Reviewed Publication
Shandong University developed an advanced AI framework that predicts molecular properties in seconds with high accuracy and minimal computational resources, dramatically accelerating and democratizing early‐stage drug discovery.
- Journal
- Frontiers of Computer Science
Finding new medicines just got faster: AI maps RNA–drug links with 96% accuracy
Higher Education PressPeer-Reviewed Publication
Scientists at Shaanxi Normal University have developed an AI-powered dual-channel model that predicts miRNA–drug interactions with up to 96% accuracy—validated on public datasets and real-world drugs—to accelerate and economize the discovery of novel therapeutic targets.
- Journal
- Frontiers of Computer Science
Clustering-based approach accelerates AI learning in robotics and gaming
Higher Education PressPeer-Reviewed Publication
Researchers from Nanjing University and UC Berkeley have unveiled a clustering-based reinforcement learning framework that balances novelty and reward to accelerate and enhance AI exploration across robotics, gaming, and real-world applications.
- Journal
- Frontiers of Computer Science
Metasurface thermal emitters enabled chip scale mid infrared spectroscopic sensing: An 'instant-camera-like' miniature mid-infrared spectral sensing platform
Chinese Society for Optical EngineeringPeer-Reviewed Publication
Mid-infrared spectroscopy, with its molecular fingerprint recognition capability, plays a crucial role in environmental monitoring, biomedical diagnosis, and industrial chemical analysis. However, traditional spectrometers suffer from drawbacks such as large size, system complexity, high cost, and operational difficulty. The research team led by Prof. Qin Chen at Jinan University has developed a chip-scale mid-infrared spectral sensing technology that pioneers a light source-side regulation strategy in contrast to dispersion element and photodetector regulartions in literature. By employing metasurface arrays as wavelength-selective thermal emission sources, the system enables "instant-camera"-style substance sensing through thermal imaging encoding/decoding, moreover, achieving an exceptional angular tolerance exceeding ±40 degrees. This innovative approach successfully integrates three key components -- light source, collimation unit, and dispersive element -- into a single platform, offering a novel solution for portable mid-infrared spectral detection.
- Journal
- PhotoniX
- Funder
- National Natural Science Foundation of China, National Natural Science Foundation of China, Basic and Applied Basic Research Foundation of Guangdong Province, Basic and Applied Basic Research Foundation of Guangdong Province
New method filters noisy data for safer AI application
Higher Education PressPeer-Reviewed Publication
Researchers have developed a self-tuning AI framework that dynamically filters noisy graph data to boost reliability and accuracy across industries from healthcare to finance.
- Journal
- Frontiers of Computer Science
Smart algorithms that prevent digital traffic jams—especially when public emergencies strike
Higher Education PressPeer-Reviewed Publication
Researchers from Jilin University and the University of North Carolina have developed an energy-efficient, stability-boosting data-offloading method that uses advanced optimization algorithms to slash delays and power use in mobile crowdsensing, paving the way for smoother, greener smart-city services.
- Journal
- Frontiers of Computer Science
Promising solution for detecting money-laundering and collusion in transaction webs
Higher Education PressPeer-Reviewed Publication
Researchers at Tongji University and the Shanghai AI Lab show that graph-based neural networks can uncover hidden money-laundering rings and collusion networks in financial transactions far more effectively than traditional methods, offering a clear roadmap for real-world implementation and stronger fraud defenses across banking, insurance, and regulatory systems.
- Journal
- Frontiers of Computer Science