CoarsenConf: Equivariant Coarsening with Aggregated Attention for Molecular Conformer Generation

TL;DR

CoarsenConf is a novel equivariant hierarchical variational autoencoder that uses coarse-graining and aggregated attention to efficiently generate high-quality molecular conformers, improving accuracy over previous models.

Contribution

The paper introduces CoarsenConf, a new model combining equivariant coarse-graining with aggregated attention for improved molecular conformer generation.

Findings

Generates more accurate conformers than prior models

Effective in downstream property prediction tasks

Enhances structural exploration in drug discovery

Abstract

Molecular conformer generation (MCG) is an important task in cheminformatics and drug discovery. The ability to efficiently generate low-energy 3D structures can avoid expensive quantum mechanical simulations, leading to accelerated virtual screenings and enhanced structural exploration. Several generative models have been developed for MCG, but many struggle to consistently produce high-quality conformers. To address these issues, we introduce CoarsenConf, which coarse-grains molecular graphs based on torsional angles and integrates them into an SE(3)-equivariant hierarchical variational autoencoder. Through equivariant coarse-graining, we aggregate the fine-grained atomic coordinates of subgraphs connected via rotatable bonds, creating a variable-length coarse-grained latent representation. Our model uses a novel aggregated attention mechanism to restore fine-grained coordinates from the coarse-grained latent representation, enabling efficient generation of accurate conformers. Furthermore, we evaluate the chemical and biochemical quality of our generated conformers on multiple downstream applications, including property prediction and oracle-based protein docking. Overall, CoarsenConf generates more accurate conformer ensembles compared to prior generative models.