SubGDiff: A Subgraph Diffusion Model to Improve Molecular Representation Learning
This work addresses molecular representation learning for AI-based drug discovery, presenting an incremental improvement by enhancing diffusion models with substructural information.
The paper tackles the problem of molecular representation learning by addressing the limitation of existing diffusion models that treat atoms independently, introducing SubGDiff which incorporates molecular substructure information into the diffusion process. The result is superior performance demonstrated through extensive downstream tasks, though no concrete numbers are provided in the abstract.
Molecular representation learning has shown great success in advancing AI-based drug discovery. The core of many recent works is based on the fact that the 3D geometric structure of molecules provides essential information about their physical and chemical characteristics. Recently, denoising diffusion probabilistic models have achieved impressive performance in 3D molecular representation learning. However, most existing molecular diffusion models treat each atom as an independent entity, overlooking the dependency among atoms within the molecular substructures. This paper introduces a novel approach that enhances molecular representation learning by incorporating substructural information within the diffusion process. We propose a novel diffusion model termed SubGDiff for involving the molecular subgraph information in diffusion. Specifically, SubGDiff adopts three vital techniques: i) subgraph prediction, ii) expectation state, and iii) k-step same subgraph diffusion, to enhance the perception of molecular substructure in the denoising network. Experimentally, extensive downstream tasks demonstrate the superior performance of our approach. The code is available at https://github.com/youjibiying/SubGDiff.