CAGenMol-2 enables versatile drug molecule design from one model
One Sequence, Many Decodings: CAGenMol-2 Recasts Drug Design as Masked Molecular Inference
Machine LearningArtificial Intelligence
Summary
Designing new drug molecules usually requires many different specialized tools. The authors present CAGenMol-2, a single AI model that can understand molecules, predict properties, design new molecules, and optimize parts of them all at once. By selectively revealing or hiding different pieces of information during use, this one model can perform many tasks without needing separate setups. They also introduce AdaFO, a new method that improves molecule design success without needing gradients, leading to better drug candidates. This approach works well with preserving important molecule structures and helps guide drug discovery workflows.
What this means in practice
- •For drug discovery teams: Use one AI model to perform multiple drug molecule design tasks without retraining or fine-tuning for each.
- •For biotechnology engineers: Apply iterative local molecular optimization with AdaFO to improve targeted drug candidates efficiently.
Authors
Yanting Li, Enyan Dai, Lei Wang, Wen-Cai Ye, Li Liu
Abstract
Drug design couples property evaluation, conditional generation, structure-based design, and local optimization, yet machine learning systems typically address these capabilities with separate task-specific models. We introduce CAGenMol-2, a masked diffusion molecular language model that represents molecules, continuous scalar properties, and 3D protein pockets within a single wrapped sequence. Within this pretrained interface, downstream operations are selected by which sequence regions are observed or masked at inference, allowing one checkpoint to perform property prediction, property- and pocket-conditioned generation, and partial-constraint design without task-specific architectures or backbone fine-tuning. We further propose Adaptive Fragment Optimization (AdaFO), a gradient-free mask-and-refill search that turns the masked decoder into an iterative local molecular optimizer. On CrossDocked2020, AdaFO increases Success Rate from 30.2\% to 70.8\%, the best reported under this protocol, while largely preserving drug-likeness and diversity. Finally, scaffold-preserving directional editing and CRBN/VHL case studies demonstrate its use in compound design workflows spanning local molecular editing, structure-based prioritization, and downstream simulation-based screening.