The advent of the evolutionary scale modeling (ESM) series of protein language models (PLMs) is a significant innovation in the convergence of large language models (LLMs) with protein representation. These models, trained on large amounts of unlabeled protein sequence data, learn the intricate patterns of mutation and conservation that have sculpted protein families through evolutionary history. ESM has become a widely used foundation model family for protein representation and downstream biological tasks.

Over the past decades, combination therapy has become a promising strategy to combat antimicrobial resistance. Compared with monotherapy, drug combinations can enhance efficacy and delay resistance. However, experimental identification of drug–drug interactions (DDIs) remains time-consuming and limits large-scale screening.

Myocardial infarction (MI) remains a major cause of morbidity and mortality worldwide, stemming from the heart’s limited regenerative capacity and formation of noncontractile fibrotic tissue. Current treatments, including pharmacological and surgical interventions, manage symptoms and restore perfusion but fail to promote regeneration.

Structured light is a new frontier in optics because it can be programmed across multiple degrees of freedom—amplitude, phase, spatial patterns, frequency, and polarization. Yet in practice, generating and controlling such light still often relies on bulky, alignment-sensitive optical setups. Researchers led by Prof. Hongtao Lin at Zhejiang University (ZJU), China, have introduced a unified inverse-design method to bring vectorial structured light onto a chip, using topological valley photonic crystals. Their tiny couplers show ultra-low loss and broad bandwidth at telecom wavelengths, offering a practical route to compact photonic chips for communications, sensing, and quantum technologies.

POSTECH Professor Seung-Ki Min’s Research Team Compares Future Extreme Fire Weather Under ‘Net-Zero’ vs. ‘Net-Negative’ Emission Scenarios.

A research team led by Professor Lizhi Xu from the Department of Mechanical Engineering under the Faculty of Engineering at the University of Hong Kong (HKU) has created a new type of ionogel that overcomes this challenge. The material is soft and flexible, yet strong enough to withstand significant mechanical stress, making it ideal for wearable and biomedical applications.

A team led by Professor Ed X. WU, Chair Professor of Biomedical Engineering & Lam Woo Professorship in Biomedical Engineering, and Dr. Alex T. L. LEONG, Research Assistant Professor from the Department of Electrical and Computer Engineering (ECE) under the Faculty of Engineering at the University of Hong Kong (HKU) has achieved a major breakthrough in understanding how the brain processes information through large-scale network changes. Their findings, published in Nature Communications, reveal that the brain can rewire its networks in just seconds after a brief neural signal—challenging the long-held belief that such changes are slow and gradual, based on imaging techniques like functional MRI (fMRI).

A research team from the Faculty of Engineering at the University of Hong Kong (HKU) has recently achieved a significant scientific breakthrough. Researchers have successfully used mechanical stretching technology to dynamically control the emission colour of gallium nitride (GaN) material from "ultraviolet (UV) to blue light." This technological breakthrough provides a new semiconductor material control solution for future advanced power transistors, optoelectronic components, radio frequency components, and micro-LED displays.

Compressed CO₂ energy storage (CCES) system has received widespread attention due to its superior performance. This paper proposes a novel CCES concept based on gas-liquid phase change and cold-electricity cogeneration. Thermodynamic and exergoeconomic analyses are performed under simulation conditions, followed by an investigation of the impacts of various decision parameters on the proposed system. Next, a multi-objective optimization is conducted with the total energy efficiency and total product unit cost as the objective functions. Finally, brief comparisons are made between the proposed system and existing systems. The results indicate that the total energy efficiency of the proposed system reaches 79.21% under the given simulation conditions, outperforming the electrical efficiency of 61.27%. Additionally, the total product unit cost of the system is 25.61 $/GJ. A key component, T1, plays an important role due to its large exergy destruction rate (1.0591 MW) and total investment cost rate (154.85 $/h). Despite this, the exergoeconomic factors of T1 is only 41.08%, indicating that investing in T1 to improve the efficiency is practicable. The analysis shows that a lower CO₂ condensation temperature benefits the proposed system performance. While improving the isentropic efficiencies of the compressors and turbines enhances total energy efficiency, excessive isentropic efficiencies can lead to a significant increase in total product unit cost. Through multi-objective optimization, an optimal favorable operating condition is identified, yielding a compromise result with a total energy efficiency of 111.91% and a total product unit cost of 28.35 $/GJ. The proposed CCES system efficiently delivers both power and cooling energy, demonstrating clear superiorities over previous systems.