Zero Shot Molecular Generation via Similarity Kernels
This work addresses the challenge of accelerating chemical discovery by enabling zero-shot molecular generation with shape control, though it is incremental as it builds on existing diffusion models and force field techniques.
The authors tackled the problem of understanding the behavior of learned scores in diffusion models for molecular generation, finding that the score transitions from a restorative potential to a quantum-mechanical force, and introduced SiMGen, a zero-shot method using similarity kernels and pretrained force field descriptors to generate molecules with control over shape, achieving competitive performance without additional training.
Generative modelling aims to accelerate the discovery of novel chemicals by directly proposing structures with desirable properties. Recently, score-based, or diffusion, generative models have significantly outperformed previous approaches. Key to their success is the close relationship between the score and physical force, allowing the use of powerful equivariant neural networks. However, the behaviour of the learnt score is not yet well understood. Here, we analyse the score by training an energy-based diffusion model for molecular generation. We find that during the generation the score resembles a restorative potential initially and a quantum-mechanical force at the end. In between the two endpoints, it exhibits special properties that enable the building of large molecules. Using insights from the trained model, we present Similarity-based Molecular Generation (SiMGen), a new method for zero shot molecular generation. SiMGen combines a time-dependent similarity kernel with descriptors from a pretrained machine learning force field to generate molecules without any further training. Our approach allows full control over the molecular shape through point cloud priors and supports conditional generation. We also release an interactive web tool that allows users to generate structures with SiMGen online (https://zndraw.icp.uni-stuttgart.de).