Artificial Intelligence is transforming protein research by predicting structures, learning patterns of evolutionary variation, and exploring designable regions of protein space. AI-derived data reveals new insights into protein organization, folding topology, and functional specialization.
A new framework aims to embed natural laws, scientific goals, and ethical standards into AI-driven molecular design processes. This requires addressing misalignment between AI objectives and real-world scientific and societal requirements.
Researchers introduced QCell, a curated collection of 525,000 new quantum-mechanical calculations for biomolecular fragments. The dataset addresses the limited coverage of nucleic acids, lipids, and carbohydrates, enabling reliable simulations of critical biological processes such as DNA dynamics and membrane behavior.
Researchers have developed AI models to predict molecular electrostatic potentials, enabling rapid and accurate analysis of battery electrolytes. The study reveals that quadrupole moments provide valuable information for recovering the electrostatic landscape from simple point charges.
A new machine learning approach accelerates Raman spectrum prediction for fast-ion conductors, revealing liquid-like ion motion. The method identifies low-frequency Raman signatures associated with high ionic mobility and relaxational host-lattice dynamics.