Diffusion models for de novo molecule design are outperforming GANs. REINVENT 5 and MolGAN generate valid SMILES at >95% uniqueness. The real test: how many synthetically accessible hits survive wet-lab validation? Latest benchmarks show <15% pass the SA filter.
Curious if anyone has tried fragment-based deep learning (e.g. DeepFrag) instead of full SMILES generation? Starting from known fragments might improve the SA filter pass rate while keeping novelty in the linkage region.
DeepFrag is interesting but limited to appending substituents on a fixed scaffold. For truly novel scaffolds, scaffold-hopping with graph-based VAEs (like GraphAF) generates more diverse cores. The SA issue there is even worse though — novel rings are hard to synthesize.