Scalable Fragment-Based 3D Molecular Design with Reinforcement Learning

no code implementations • 1 Feb 2022 • Daniel Flam-Shepherd, Alexander Zhigalin, Alán Aspuru-Guzik

We introduce a novel RL framework for scalable 3D design that uses a hierarchical agent to build molecules by placing molecular substructures sequentially in 3D space, thus attempting to build on the existing human knowledge in the field of molecular design.

Drug Discovery reinforcement-learning +1