ATOM3D: Tasks On Molecules in Three Dimensions
This work provides a standardized benchmark and toolkit for researchers developing 3D molecular machine learning methods, aiming to accelerate progress in computational biology and chemistry.
This paper introduces ATOM3D, a benchmark collection and toolkit for 3D molecular machine learning, addressing the lack of systematic benchmarks and unified tools in the biomolecular domain. The authors demonstrate that 3D learning methods consistently outperform 1D and 2D representations, with specific architectures like 3D convolutional networks, graph networks, and equivariant networks showing varying strengths across tasks.
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 .