MolMiner: Towards Controllable, 3D-Aware, Fragment-Based Molecular Design
This addresses the need for flexible and geometry-aware molecular generation in drug discovery, though it appears incremental by combining existing techniques like fragment-based methods and conditioning.
The paper tackles the problem of controllable molecular design by introducing MolMiner, a fragment-based model that dynamically updates 3D geometry and supports conditional generation over twelve properties, achieving calibrated performance and competitive unconditional results.
We introduce MolMiner, a fragment-based, geometry-aware, and order-agnostic autoregressive model for molecular design. MolMiner supports conditional generation of molecules over twelve properties, enabling flexible control across physicochemical and structural targets. Molecules are built via symmetry-aware fragment attachments, with 3D geometry dynamically updated during generation using forcefields. A probabilistic conditioning mechanism allows users to specify any subset of target properties while sampling the rest. MolMiner achieves calibrated conditional generation across most properties and offers competitive unconditional performance. We also propose improved benchmarking methods for both unconditional and conditional generation, including distributional comparisons via Wasserstein distance and calibration plots for property control. To our knowledge, this is the first model to unify dynamic geometry, symmetry handling, order-agnostic fragment-based generation, and high-dimensional multi-property conditioning.