Equivariant Graph Neural Networks for 3D Macromolecular Structure

By Bowen Jing, Stephan Eismann, Pratham N. Soni, Ron O. Dror. 2021

129 citations recorded by the source corpus.

Representing and reasoning about 3D structures of macromolecules is emerging as a distinct challenge in machine learning. Here, we extend recent work on geometric vector perceptrons and apply equivariant graph neural networks to a wide range of tasks from structural biology. Our method outperforms all reference architectures on three out of eight tasks in the ATOM3D benchmark, is tied for first on two others, and is competitive with equivariant networks using higher-order representations and spherical harmonic convolutions. In addition, we demonstrate that transfer learning can further improve performance on certain downstream tasks. Code is available at https://github.com/drorlab/gvp-pytorch.

Subfields: Computational Theory and Mathematics, Materials Chemistry, Molecular Biology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Alexander Rives
2. Robert Verkuil
3. Jiahua Rao
4. Adam Lerer
5. Jason Liu
6. Tom Sercu
7. Zeming Lin
8. Brian L. Hie

In Materials Chemistry

1. Alexander Rives
2. Jiahua Rao
3. Tom Sercu
4. Shuangjia Zheng

In Molecular Biology

1. Brian L. Hie
2. Alexander Rives
3. Robert Verkuil
4. Jiahua Rao
5. Adam Lerer
6. Chloe Hsu
7. Jason Liu
8. Tom Sercu

Citations in the corpus

• TANKBind: Trigonometry-Aware Neural NetworKs for Drug-Protein Binding Structure Prediction (2022, methodology): “We then apply the geometric vector perceptrons (GVP) [18, 37] to embed the protein and arrive at feature h ∈ Rn×s after graph propagation, where n is the number of nodes and s is the embedding size.”
• Learning inverse folding from millions of predicted structures (2022, methodology): “We compare zero-shot predictions using inverse folding models to supervised and transfer learning methods (Townshend et al., 2020; Jing et al., 2021a) on the Atom3D test set.”
• Pocket2Mol: Efficient Molecular Sampling Based on 3D Protein Pockets (2022, unknown): “We adopt the geometric vector perceptrons (Jing et al., 2021) and the vector-based neural network (Deng et al., 2021) to achieve E(3)-equivariance.”