Poster
in
Workshop: Machine Learning for Drug Discovery (MLDD)
MiDi: Mixed Graph and 3D Denoising Diffusion for Molecule Generation
ClĂ©ment Vignac · Nagham Osman · Laura Toni · Pascal Frossard
Abstract:
This work introduces MiDi, a diffusion model for jointly generating molecular graphs and corresponding 3D conformers. In contrast to existing models which derive molecular bonds from the conformation using predefined rules, MiDi streamlines the molecule generation process with an end-to-end differentiable model. Preliminary results demonstrate the benefits of this approach: on the complex GEOM-DRUGS dataset, our model generates significantly better molecular graphs than 3D-based models, and even surpasses specialized algorithms that directly optimize the bond orders for validity.
Chat is not available.