TreeRef-BFN generates versatile 3D molecules from variable structures
TreeRef-BFN: Equivariance-Free De Novo Molecule Generation based on 2D Topology and Internal 3D Geometry
Artificial IntelligenceMachine Learning
Summary
Making new 3D molecules from scratch is hard because you have to decide their size, shape, and connections all at once. The authors created TreeRef-BFN, a method that represents molecules as trees capturing both their connections and small 3D details. This lets the method build molecules of different sizes and shapes more flexibly, without retraining for each task. It can quickly generate realistic and stable molecules with varied properties, useful for designing chemicals or drugs.
What this means in practice
- •For pharmaceutical chemists: Generate new 3D drug-like molecules with complex structures for early-stage drug discovery without retraining models for each new scaffold.
- •For materials scientists: Design novel molecules with desired 3D shapes and properties by completing molecular fragments flexibly using a single pretrained generator.
Authors
Ruiqing Sun, Sen Yang, Dawei Feng, Bo Ding, Yijie Wang, Huaimin Wang
Abstract
De novo 3D molecular generation jointly models molecular size, topology, and geometry. Most methods pre-sample molecular size and generate Cartesian coordinates, limiting variable-size conditional tasks such as fragment completion and scaffold decoration while often relying on equivariant architectures. Internal-coordinate methods avoid rigid-body redundancy but typically require a known molecular graph or autoregressive construction, which may accumulate errors. We propose TreeRef, a tree-based molecular representation that assigns molecular topology and topology-dependent local 3D geometry to a naturally variable-size tree. RingRef nodes encode ring closures while preserving the tree structure, while Null nodes allow molecular size to emerge directly from node occupancy. Based on TreeRef, we develop TreeRef-BFN, a Bayesian Flow Network with a standard Transformer backbone that globally couples these locally defined variables and jointly generates discrete molecular variables and continuous local geometry. A single pretrained TreeRef-BFN supports unconditional generation and variable-size structure-conditioned 3D generation through masking alone, without retraining. Empirical studies demonstrate strong chemical validity, molecular stability, and diversity, accurate local geometric distributions, fast sampling, and competitive property-conditioned generation, establishing TreeRef-BFN as an efficient and flexible framework for 3D molecular generation.