Graph Neural Networks (GNN) with hierarchical explainability provide accurate multi-target screening in drug discovery.
Combining 3D molecular generation with optimization frameworks for improved drug design.
Incorporating decomposed priors improves diffusion model performance for drug design.