ATOM3D: Tasks On Molecules in Three Dimensions

TL;DR

ATOM3D introduces a comprehensive benchmark suite and toolkit for three-dimensional molecular learning, demonstrating that 3D deep learning models outperform traditional 2D approaches across various biomolecular tasks.

Contribution

The paper provides the first systematic benchmark datasets and a unified toolkit for 3D molecular learning, highlighting the importance of architecture choice and offering open-source resources.

Findings

3D convolutional networks excel at complex geometries

Graph networks perform well with positional data

Equivariant networks show significant promise

Abstract

Computational methods that operate on three-dimensional molecular structure have the potential to solve important questions in biology and chemistry. In particular, deep neural networks have gained significant attention, but their widespread adoption in the biomolecular domain has been limited by a lack of either systematic performance benchmarks or a unified toolkit for interacting with molecular data. To address this, we present ATOM3D, a collection of both novel and existing benchmark datasets spanning several key classes of biomolecules. We implement several classes of three-dimensional molecular learning methods for each of these tasks and show that they consistently improve performance relative to methods based on one- and two-dimensional representations. The specific choice of architecture proves to be critical for performance, with three-dimensional convolutional networks excelling at tasks involving complex geometries, graph networks performing well on systems requiring detailed positional information, and the more recently developed equivariant networks showing significant promise. Our results indicate that many molecular problems stand to gain from three-dimensional molecular learning, and that there is potential for improvement on many tasks which remain underexplored. To lower the barrier to entry and facilitate further developments in the field, we also provide a comprehensive suite of tools for dataset processing, model training, and evaluation in our open-source atom3d Python package. All datasets are available for download from https://www.atom3d.ai .