GeoDiff: a Geometric Diffusion Model for Molecular Conformation Generation

Abstract

Predicting molecular conformations from molecular graphs is a fundamental problem in cheminformatics and drug discovery. Recently, significant progress has been achieved with machine learning approaches, especially with deep generative models. Inspired by the diffusion process in classical non-equilibrium thermodynamics where heated particles will diffuse from original states to a noise distribution, in this paper, we propose a novel generative model named GeoDiff for molecular conformation prediction. GeoDiff treats each atom as a particle and learns to directly reverse the diffusion process (i.e., transforming from a noise distribution to stable conformations) as a Markov chain. Modeling such a generation process is however very challenging as the likelihood of conformations should be roto-translational invariant. We theoretically show that Markov chains evolving with equivariant Markov kernels can induce an invariant distribution by design, and further propose building blocks for the Markov kernels to preserve the desirable equivariance property. The whole framework can be efficiently trained in an end-to-end fashion by optimizing a weighted variational lower bound to the (conditional) likelihood. Experiments on multiple benchmarks show that GeoDiff is superior or comparable to existing state-of-the-art approaches, especially on large molecules.

arXiv commentPublished as a conference paper at ICLR 2022 (https://openreview.net/forum?id=PzcvxEMzvQC)

Cited by in corpus (19)

• Regression Transformer: Concurrent sequence regression and generation for molecular language modeling
• DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models
• Time Series Diffusion Method: A Denoising Diffusion Probabilistic Model for Vibration Signal Generation
• A generative artificial intelligence framework based on a molecular diffusion model for the design of metal-organic frameworks for carbon capture
• A Diffusion model for POI recommendation
• Generative AI and Process Systems Engineering: The Next Frontier
• De novo Design of Polymer Electrolytes with High Conductivity using GPT-based and Diffusion-based Generative Models
• Roadmap on machine learning glassy dynamics
• Accelerating Material Design with the Generative Toolkit for Scientific Discovery
• Sampling with flows, diffusion and autoregressive neural networks: A spin-glass perspective
• CREMP: Conformer-rotamer ensembles of macrocyclic peptides for machine learning
• Normalizing flows as an enhanced sampling method for atomistic supercooled liquids
• Riemannian Denoising Diffusion Probabilistic Models
• Reflection-Equivariant Diffusion for 3D Structure Determination from Isotopologue Rotational Spectra in Natural Abundance
• Physics-informed generative model for drug-like molecule conformers
• Unraveling the Potential of Diffusion Models in Small Molecule Generation
• Learning dissection trajectories from expert surgical videos via imitation learning with equivariant diffusion
• Zero Shot Molecular Generation via Similarity Kernels
• General Binding Affinity Guidance for Diffusion Models in Structure-Based Drug Design