Abstract
<jats:p>While artificial intelligence has revolutionized the prediction of static protein structures, characterizing their dynamics and interactions with drug candidates remains a computational bottleneck. Here, we introduce FeNNix-Bio1, a foundation machine learning model designed to power accurate, reactive atomistic simulations of biological systems at an unprecedented speed and scalability. Trained exclusively on high-accuracy computational chemistry data, FeNNix-Bio1 accurately captures complex condensed-phase phenomena such as ion solvation and subtle liquid water properties for which it outperforms state-of-the-art specialized force fields. We demonstrate its versatility across a full spectrum of drug design applications, including the calculation of hydration free energies (HFEs), the structure and dynamics of proteins, the simulation of protein-ligand absolute binding free energies (ABFEs), lipid bilayer properties and chemical reactions. Notably, FeNNix-Bio1 provides excellent agreement with experiments for both the HFEs of the more than 600 molecules of the Freesolv dataset (MAE of 0.7 kcal/mol) and for the ABFEs of 30 representative protein-ligand complexes covering three protein targets (MAE of 1.05 kcal/mol). By enabling scalable, quantum-accurate molecular dynamics without the need for manual parametrization, FeNNix-Bio1 bridges the gap between static structure prediction and dynamic biological reality, unlocking next-generation drug design with physics-driven AI.</jats:p>