End-to-End Learning on 3D Protein Structure for Interface Prediction

By Raphael J. L. Townshend, Rishi Bedi, Patricia Suriana, Ron O. Dror. 2018

118 citations recorded by the source corpus.

Despite an explosion in the number of experimentally determined, atomically detailed structures of biomolecules, many critical tasks in structural biology remain data-limited. Whether performance in such tasks can be improved by using large repositories of tangentially related structural data remains an open question. To address this question, we focused on a central problem in biology: predicting how proteins interact with one another—that is, which surfaces of one protein bind to those of another protein. We built a training dataset, the Database of Interacting Protein Structures (DIPS), that contains biases but is two orders of magnitude larger than those used previously. We found that these biases significantly degrade the performance of existing methods on gold-standard data. Hypothesizing that assumptions baked into the hand-crafted features on which these methods depend were the source of the problem, we developed the first end-to-end learning model for protein interface prediction, the Siamese Atomic Surfacelet Network (SASNet). Using only spatial coordinates and identities of atoms, SASNet outperforms state-of-the-art methods trained on gold-standard structural data, even when trained on only 3% of our new dataset. Code and data available at https://github.com/drorlab/DIPS.

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

In Materials Chemistry

In Molecular Biology

Citations in the corpus

EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction (2022, methodology): “…other protein representations have been proposed for (graph) DL methods for individual structure prediction (Jing et al., 2020), protein-protein interactions (Dai & Bailey-Kellogg, 2021; Eismann et al., 2020; Townshend et al., 2019), or protein function prediction (Gligorijević et al., 2021).”
Learning from Protein Structure with Geometric Vector Perceptrons (2020, background): “Thanks to their importance and difficulty, such problems, which we broadly refer to as learning from structure, have recently developed into an exciting and promising application area for deep learning (Graves et al., 2020; Ingraham et al., 2019; Pereira et al., 2016; Townshend et al., 2019; Won et ”
Multi-Scale Representation Learning on Proteins (2022, methodology): “, 2018), protein contact prediction (Townshend et al., 2019) and protein-ligand binding affinity tasks (Ragoza et al.”
Independent SE(3)-Equivariant Models for End-to-End Rigid Protein Docking (2021, background): “5 test set, and randomly sample 100 pairs from different protein families from the DIPS test set.”
Hierarchical, rotation‐equivariant neural networks to select structural models of protein complexes (2020, unknown): “voxelization-based network architectures (that only take the atoms at the protein-protein interface into account) and highlights the efficiency of our point-based approach.(29) 4 | DISCUSSION”
Fast end-to-end learning on protein surfaces (2020, background): “Relations between amino acids of different proteins have also been predicted to handle protein-protein interactions [42, 20].”
ATOM3D: Tasks On Molecules in Three Dimensions (2020, methodology): “Protein Interface Prediction (PIP) Amino Acid Interaction DIPS [Townshend et al., 2019] DB5 [Vreven et al.”
Masked Inverse Folding with Sequence Transfer for Protein Representation Learning (2023, background): “…2019), corrupting and reconstructing the structure in addition to the sequence (Mansoor et al., 2021; Chen et al., 2022), encoding surface features (Townshend et al., 2019), contrastive learning (Zhang et al., 2022; Cao et al., 2021), or a graph encoder without sequence decoding (Somnath et al.,…”
A Survey of Deep Learning for Scientific Discovery (2020, background): “Scientiﬁc Examples Graph neural networks have been very popular for several chemistry tasks, such as predicting molecular properties [53, 93, 67, 103], determining protein interfaces [60, 229] and even generating candidate molecules [41, 21].”
Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures (2021, background): “…of the structure (Derevyanko et al., 2018), classiﬁed proteins in enzymes classes (Amidi et al., 2017), predicted the protein-ligand binding afﬁnity (Ragoza et al., 2017) and the binding site (Jiménez et al., 2017), as well as the contact region between two proteins (Townshend et al., 2019).”