Deciphering interaction fingerprints from protein molecular surfaces using geometric deep learning

By Pablo Gainza, Freyr Sverrisson, F. Monti, E. Rodolà, Davide Boscaini, Michael M. Bronstein, Bruno E. Correia. 2019

804 citations recorded by the source corpus.

Predicting interactions between proteins and other biomolecules solely based on structure remains a challenge in biology. A high-level representation of protein structure, the molecular surface, displays patterns of chemical and geometric features that fingerprint a protein’s modes of interactions with other biomolecules. We hypothesize that proteins participating in similar interactions may share common fingerprints, independent of their evolutionary history. Fingerprints may be difficult to grasp by visual analysis but could be learned from large-scale datasets. We present MaSIF (molecular surface interaction fingerprinting), a conceptual framework based on a geometric deep learning method to capture fingerprints that are important for specific biomolecular interactions. We showcase MaSIF with three prediction challenges: protein pocket-ligand prediction, protein–protein interaction site prediction and ultrafast scanning of protein surfaces for prediction of protein–protein complexes. We anticipate that our conceptual framework will lead to improvements in our understanding of protein function and design. MaSIF, a deep learning-based method, finds common patterns of chemical and geometric features on biomolecular surfaces for predicting protein–ligand and protein–protein interactions.

Subfields: Computational Theory and Mathematics, Molecular Biology, Graph Learning, Computational Biology & Bioinformatics

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 Molecular Biology

In Graph Learning

In Computational Biology & Bioinformatics

Citations in the corpus

TANKBind: Trigonometry-Aware Neural NetworKs for Drug-Protein Binding Structure Prediction (2022, methodology): “To implicitly model side-chain flexibility, we choose a residue-level representation ignoring the finer details of protein structure, separating our method from other methods that use all-atoms or surface vertexes representation Jiang et al. [2021], Gainza et al. [2020].”
EquiBind: Geometric Deep Learning for Drug Binding Structure Prediction (2022, background): “Gainza et al. (2020); Sverrisson et al. (2021) leverage geometric deep learning and mesh convolutional neural networks (CNN) to embed protein surface patches into ﬁngerprints and allow for fast scanning and binding site identiﬁcation, removing the need for handcrafted or expensive pre-computed…”
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 [12, 13, 15, 23, 27, 31].”
Multi-Scale Representation Learning on Proteins (2022, methodology): “Interaction between proteins and ligands, on the other hand, is controlled by molecular surface contacts [Gainza et al., 2020].”
Geometric deep learning on molecular representations (2021, background): “Protein-protein interaction prediction and ligand-pocket fingerprinting [18].”
P HYSICS - INFORMED DEEP NEURAL NETWORK FOR RIGID - BODY PROTEIN DOCKING (2022, methodology): “To this end, we build upon our previous works on protein surface fingerprints Gainza et al. (2020); Sverrisson et al. (2021): we use a point neural network to predict the locations of the docking sites (highlighted in red) on both protein surfaces, before relying on a probabilistic method to…”
Therapeutics Data Commons: Machine Learning Datasets and Tasks for Drug Discovery and Development (2021, background): “…(Zitnik, Agrawal & Leskovec 2018), drug repurposing for emerging diseases (Gysi et al. 2020), protein folding and design (Jumper et al. 2020, Gao et al. 2020), and biomolecular interactions (Zitnik et al. 2015, Agrawal et al. 2018, Huang, Xiao, Glass, Zitnik & Sun 2020, Gainza et al. 2020).”
Graph Representation Learning in Biomedicine and Healthcare (2021, positive): “…and of geometric features (such as distance-dependent curvature) to predict protein-pocket– ligand interactions and protein–protein interactions 113 ; to generate intramolecular and intermolecular residue contact graphs to predict intramolecular and intermolecular energies, binding affinities…”
Improving Graph Neural Network Expressivity via Subgraph Isomorphism Counting (2020, background): “In particular, Graph Neural Networks (GNNs), a family of neural architectures designed for irregularly structured data, have been successfully applied to problems ranging from social networks and recommender systems [1] to bioinformatics [2], [3], chemistry [4], [5], [6] and physics [7], [8], to nam”
GeoMol: Torsional Geometric Generation of Molecular 3D Conformer Ensembles (2021, background): “…it is crucial to understand how a molecule binds to a speciﬁc target protein; this process heavily depends on the 3D structures of the two components, both in terms of geometric (shape matching) and chemical (hydrophobic/hydrophilic) interactions [Gainza et al., 2020, Sverrisson et al., 2020].”
Topsy-Turvy: integrating a global view into sequence-based PPI prediction (2022, background): “Loss terms that quantify protein functional similarity (Ghersi and Singh, 2014) or interface similarity (Budowski-Tal et al., 2018; Gainza et al., 2020) could be added to the framework to further inform predictions.”
Protein Representation Learning by Geometric Structure Pretraining (2022, methodology): “Another potential direction is to extract features from protein surfaces (Gainza et al., 2020; Sverrisson et al., 2021; Dai & Bailey-Kellogg, 2021).”
Artificial Intelligence For Natural Product Drug Discovery (2023, background): “…predicting drug–target binding affinity 113 , SMILES-based approaches for de novo drug-like molecule generation 114,115 , graph-based de novo molecular generation 116 , and property prediction 117,118 and surface mesh-based approaches for protein pocket-conditioned molecular representations 119 .”
GNNGuard: Defending Graph Neural Networks against Adversarial Attacks (2020, result): “While deep learning on graphs has shown exciting results in a variety of applications [6, 23, 35–37], little attention has been paid to the robustness of such models, in contrast to an abundance of research for image (e.”