Fast end-to-end learning on protein surfaces

Sverrisson, Freyr; Feydy, Jean; Correia, Bruno E.; Bronstein, Michael M.

doi:10.1109/cvpr46437.2021.01502

Cited by 76 publications

(79 citation statements)

References 31 publications

(25 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Protein Surface Representation Structure-based encoders of protein molecular surfaces have been successfully applied to predicting protein-protein interaction sites and identifying potential binding partners in protein docking (Gainza et al, 2019;Sverrisson et al, 2021;Somnath et al, 2021). Molecular surface interaction fingerprinting (MaSIF) (Gainza et al, 2019) pioneered learning geometric surface descriptors for solving protein interaction-related tasks.…”

Section: Related Workmentioning

confidence: 99%

“…Protein Pockets To retrieve the protein pocket, we first generate a surface of the protein and compute embeddings for each point on the surface using dMaSIF (Sverrisson et al, 2021). For each ligand atom, we select the closest point on the protein surface.…”

Section: Dataset Preparationmentioning

confidence: 99%

“…In Table 1, we report the fractions of pockets from the query set for which the methods returned correct ligands in top-1, top-5, top-10, top-20, and top-50. Our best performing method operates on pockets built with neighborhood radius r = 2 Å and using a dMaSIF-search model trained (for 33 epochs) with the standard parameters (Sverrisson et al, 2021) except subsampling (set to 150), resolution (set to 0.7 Å), and embedding size (set to 16). We report metrics computed after the first and last stages of the search process.…”

Section: Ligand Searchmentioning

confidence: 99%

“…In case of dMaSIF-ligand and dMaSIF-affinity, we slightly adjusted the code for new purposes. dMaSIF-search models were trained on the same dataset as in Sverrisson et al (2021). dMaSIF-ligand was trained on the dataset that was used for training MaSIF-ligand Gainza et al (2019).…”

Section: A Ligand Search Parametersmentioning

confidence: 99%

“…It computes geometric and chemical features on a triangulated surface mesh and aggregates local information using a geometric deep learning architecture based on geodesic convolution operators (Bronstein et al, 2017). Differential MaSIF (dMaSIF) (Sverrisson et al, 2021) is a derived method that sidesteps the expensive mesh generation and feature pre-computation steps and creates a lightweight point cloud representation of the molecular surface solely from raw atom coordinates and types. Feature vectors at each surface point are updated by applying approximate geodesic convolutions resulting in the final embedding vectors that can be further used in various downstream tasks.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Decoding Surface Fingerprints for Protein-Ligand Interactions

Igashov

Jamasb

Sadek

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Small molecules have been the preferred modality for drug development and therapeutic interventions. This molecular format presents a number of advantages, e.g. long half-lives and cell permeability, making it possible to access a wide range of therapeutic targets. However, finding small molecules that engage “hard-to-drug” protein targets specifically and potently remains an arduous process, requiring experimental screening of extensive compound libraries to identify candidate leads. The search continues with further optimization of compound leads to meet the required potency and toxicity thresholds for clinical applications. Here, we propose a new computational workflow for high-throughput fragment-based screening and binding affinity prediction where we leverage the available protein-ligand complex structures using a state-of-the-art protein surface embedding framework (dMaSIF). We developed a tool capable of finding suitable ligands and fragments for a given protein pocket solely based on protein surface descriptors, that capture chemical and geometric features of the target pocket. The identified fragments can be further combined into novel ligands. Using the structural data, our ligand discovery pipeline learns the signatures of interactions between surface patches and small pharmacophores. On a query target pocket, the algorithm matches known target pockets and returns either potential ligands or identifies multiple ligand fragments in the binding site. Our binding affinity predictor is capable of predicting the affinity of a given protein-ligand pair, requiring only limited information about the ligand pose. This enables screening without the costly step of first docking candidate molecules. Our framework will facilitate the design of ligands based on the target’s surface information. It may significantly reduce the experimental screening load and ultimately reveal novel chemical compounds for targeting challenging proteins.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Dataset Preparationmentioning

confidence: 99%

Section: Ligand Searchmentioning

confidence: 99%

Section: A Ligand Search Parametersmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Decoding Surface Fingerprints for Protein-Ligand Interactions

Igashov

Jamasb

Sadek

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

Structure‐Based Drug Discovery with Deep Learning**

et al. 2023

View full text Add to dashboard Cite

Artificial intelligence (AI) in the form of deep learning has promise for drug discovery and chemical biology, for example, to predict protein structure and molecular bioactivity, plan organic synthesis, and design molecules de novo. While most of the deep learning efforts in drug discovery have focused on ligand‐based approaches, structure‐based drug discovery has the potential to tackle unsolved challenges, such as affinity prediction for unexplored protein targets, binding‐mechanism elucidation, and the rationalization of related chemical kinetic properties. Advances in deep‐learning methodologies and the availability of accurate predictions for protein tertiary structure advocate for a renaissance in structure‐based approaches for drug discovery guided by AI. This review summarizes the most prominent algorithmic concepts in structure‐based deep learning for drug discovery, and forecasts opportunities, applications, and challenges ahead.

show abstract

Flexible protein–protein docking with a multitrack iterative transformer

Chu,

Ruffolo,

Harmalkar

et al. 2024

Protein Science

View full text Add to dashboard Cite

Conventional protein‐protein docking algorithms usually rely on heavy candidate sampling and re‐ranking, but these steps are time‐consuming and hinder applications that require high‐throughput complex structure prediction, e.g., structure‐based virtual screening. Existing deep learning methods for protein‐protein docking, despite being much faster, suffer from low docking success rates. In addition, they simplify the problem to assume no conformational changes within any protein upon binding (rigid docking). This assumption precludes applications when binding‐induced conformational changes play a role, such as allosteric inhibition or docking from uncertain unbound model structures. To address these limitations, we present GeoDock, a multi‐track iterative transformer network to predict a docked structure from separate docking partners. Unlike deep learning models for protein structure prediction that input multiple sequence alignments (MSAs), GeoDock inputs just the sequences and structures of the docking partners, which suits the tasks when the individual structures are given. GeoDock is flexible at the protein residue level, allowing the prediction of conformational changes upon binding. On the DIPS test set, GeoDock achieves a 43% top‐1 success rate, outperforming all other tested methods. However, in the standard DIPS train/test splits, we discovered contamination of close homologs in the training set. After decontaminating the training set, the success rate is 31%. On the DB5.5 test set and a benchmark dataset of antibody‐antigen complexes, GeoDock outperforms the deep learning models trained using the same dataset but falls behind most of the conventional methods and AlphaFold‐Multimer. GeoDock attains an average inference speed of under one second on a single GPU, enabling its application in large‐scale structure screening. Although binding‐induced conformational changes are still a challenge owing to limited training and evaluation data, our architecture sets up the foundation to capture this backbone flexibility. Code and a demonstration Jupyter notebook are available at https://github.com/Graylab/GeoDock.This article is protected by copyright. All rights reserved.

show abstract

Fast end-to-end learning on protein surfaces

Cited by 76 publications

References 31 publications

Decoding Surface Fingerprints for Protein-Ligand Interactions

Decoding Surface Fingerprints for Protein-Ligand Interactions

Structure‐Based Drug Discovery with Deep Learning**

Flexible protein–protein docking with a multitrack iterative transformer

Contact Info

Product

Resources

About