MG-BERT combines GNN-style local attention with BERT’s masked pretraining on molecular graphs, learning context-sensitive atomic representations that improve ADMET property prediction across 11 benchmark datasets.
AlphaDrug generates drug candidates for specific protein targets by combining an Lmser Transformer (with hierarchical encoder-decoder skip connections) and Monte Carlo tree search guided by docking scores, achieving higher binding affinities than known ligands on 86% of test proteins.
Proposes Augmented Hill-Climb, a hybrid RL strategy for SMILES-based generative models that improves sample efficiency ~45-fold over REINVENT by filtering low-scoring molecules from the loss computation, with diversity filters to prevent mode collapse.
Shows that divergence between optimization and control scores during goal-directed molecular generation is explained by pre-existing disagreement among QSAR models on the training distribution, not by algorithmic exploitation of model-specific biases.
BindGPT formulates 3D molecular design as autoregressive text generation over combined SMILES and XYZ tokens, using large-scale pre-training and reinforcement learning to achieve competitive pocket-conditioned molecule generation.
Winter et al. propose CDDD, a translation-based encoder-decoder that learns continuous molecular descriptors by translating between equivalent chemical representations like SMILES and InChI, pretrained on 72 million compounds.
CogMol uses a SMILES VAE and multi-attribute controlled sampling (CLaSS) to generate novel, target-specific drug molecules for unseen SARS-CoV-2 proteins without model retraining.
Introduces curriculum learning to the REINVENT de novo design platform, decomposing complex drug design objectives into simpler sequential tasks that accelerate agent convergence and improve output quality over standard reinforcement learning.
This survey organizes generative AI for de novo drug design into two themes: small molecule generation (target-agnostic, target-aware, conformation) and protein generation (structure prediction, sequence generation, backbone design, antibody, peptide). It covers four generative model families (VAEs, GANs, diffusion, flow-based), catalogs key datasets and benchmarks, and provides 12 comparative benchmark tables across all subtasks.
Lingo3DMol introduces FSMILES, a fragment-based SMILES representation with local and global coordinates, to generate drug-like 3D molecules in protein pockets via a transformer language model.
Link-INVENT is an RNN-based generative model for molecular linker design that uses reinforcement learning with a flexible scoring function, demonstrated on fragment linking, scaffold hopping, and PROTAC design.
PrefixMol prepends learnable condition vectors to a GPT transformer for SMILES generation, enabling joint control over binding pocket targeting and chemical properties like QED, SA, and LogP.