Deep Local Analysis evaluates protein docking conformations with locally oriented cubes

Behbahani, Yasser Mohseni; Crouzet, Simon; Laine, Élodie; Carbone, Alessandra

doi:10.1093/bioinformatics/btac551

Cited by 10 publications

(8 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We formulate the task of protein-ligand interaction prediction as a multirelational link prediction problem in a heterogeneous graph. The formulation is inspired by the observations that heterogeneous multimodal graphs depict various entities and their interactions in a real-world system, and are therefore ideal for coarse-grained modelling of biochemical processes [38, 47]. We subsequently trained a GNN model to predict bioactivities of unseen protein-ligand pairs (Figure 1).…”

Section: Resultsmentioning

confidence: 99%

“…The benefit is particularly prominent in protein-relevant tasks, such as protein representation [zhang2022protein, ingraham2019generative, 29], protein folding [35], protein design [36,37], and protein-protein interaction [38][39][40][41][42][43]. Second, various research groups have developed new techniques to improve the interpretability of GNNs [24,44,45].…”

Section: Introductionmentioning

confidence: 99%

“…protein folding [38], protein design [39,40], and protein-protein interaction [41][42][43][44][45][46][47].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction

Crouzet,

Lieberherr,

Atz

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Protein-ligand interaction (PLI) shapes efficacy and safety profiles of small-molecule drugs. Most existing methods rely on resource-intensive computation to predict PLI based on structural information. Here we show that a light-weight graph neural network (GNN), trained with quantitative PLIs of a small number of proteins and ligands, is able to predict the strength of unseen PLIs. The model has no direct access to structural information of protein-ligand complexes. Instead, the predictive power is provided by encoded knowledge of proteins and ligands, including primary protein sequence, gene expression, protein-protein interaction network, and structural similarities between ligands. Our novel model performs competitively with or better than structure-aware models. Our results suggest that existing PLI-prediction methods may be further improved by using omics data and representation learning techniques that embed biological and chemical knowledge.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction

Crouzet,

Lieberherr,

Atz

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…For instance, DeepRank [17] is a 3D CNN-based deep learning method that first maps the amino acid residues at the protein-protein interface to a 3D grid centered on the interface. More recently, Deep Local Analysis (DLA) [18] extends such 3D grid representations to an ensemble of 3D grids.…”

Section: Related Workmentioning

confidence: 99%

EGGNet, a generalizable geometric deep learning framework for protein complex pose scoring

Wang

Brand

Adolf-Bryfogle

et al. 2023

Preprint

View full text Add to dashboard Cite

Computational prediction of molecule-protein interactions has been key for developing new molecules to interact with a target protein for therapeutics development. Literature includes two independent streams of approaches: (1) predicting protein-protein interactions between naturally occurring proteins and (2) predicting the binding affinities between proteins and small molecule ligands (aka drug target interaction, or DTI). Studying the two problems in isolation has limited computational models' ability to generalize across tasks, both of which ultimately involve non- covalent interactions with a protein target. In this work, we developed Equivariant Graph of Graphs neural Network (EGGNet), a geometric deep learning framework for molecule-protein binding predictions that can handle three types of molecules for interacting with a target protein: (1) small molecules, (2) synthetic peptides and (3) natural proteins. EGGNet leverages a graph of graphs (GoGs) representation constructed from the molecule structures at atomic-resolution and utilizes a multi-resolution equivariant graph neural network (GNN) to learn from such representations. In addition, EGGNet gets inspired by biophysics and makes use of both atom- and residue-level interactions, which greatly improve EGGNet's ability to rank candidate poses from blind docking. EGGNet achieves competitive performance on both a public protein-small molecule binding affinity prediction task (80.2% top-1 success rate on CASF-2016) and an synthetic protein interface prediction task (88.4% AUPR). We envision that the proposed geometric deep learning framework can generalize to many other protein interaction prediction problems, such as binding site prediction and molecular docking, helping to accelerate protein engineering and structure-based drug development.

show abstract

“…For example, PatchBag characterized protein interface regions in terms of geometrical features from small surface units to search for evolutionary and functional relationships between proteins [6]. Deep Local Analysis evaluates the 3D conformational information with locally oriented cubes [46]. Molecular Surface Interaction Fingerprint (MaSIF) adapted a “patch” data representation to predict protein interactions [19].…”

Section: Introductionmentioning

confidence: 99%

PIsToN: Evaluating Protein Binding Interfaces with Transformer Networks

Stebliankin

Shirali

Baral

et al. 2023

Preprint

View full text Add to dashboard Cite

The computational studies of protein binding are widely used to investigate fundamental biological processes and facilitate the development of modern drugs, vaccines, and therapeutics. Scoring functions aim to predict complexes that would be formed by the binding of two biomolecules and to assess and rank the strength of the binding at the interface. Despite past efforts, the accurate prediction and scoring of protein binding interfaces remain a challenge. The physics-based methods are computationally intensive and often have to trade accuracy for computational cost. The possible limitations of current machine learning (ML) methods are ineffective data representation, network architectures, and limited training data. Here, we propose a novel approach called PIsToN (evaluating Protein binding Interfaces with Transformer Networks) that aim to distinguish native-like protein complexes from decoys. Each protein interface is transformed into a collection of 2D images (interface maps), where each im- age corresponds to a geometric or biochemical property in which pixel intensity represents the feature values. Such a data representation provides atomic-level resolution of relevant protein characteristics. To build hybrid machine learning models, additional empirical-based energy terms are computed and provided as inputs to the neural network. The model is trained on thousands of native and computationally-predicted protein complexes that contain challenging examples. The multi-attention transformer network is also endowed with explainability by highlighting the specific features and binding sites that were the most important for the classification decision. The developed PIsToN model significantly outperforms existing state-of-the-art scoring functions on well-known datasets.

show abstract

Deep Local Analysis evaluates protein docking conformations with locally oriented cubes

Cited by 10 publications

References 50 publications

G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction

G-PLIP: Knowledge graph neural network for structure-free protein-ligand bioactivity prediction

EGGNet, a generalizable geometric deep learning framework for protein complex pose scoring

PIsToN: Evaluating Protein Binding Interfaces with Transformer Networks

Contact Info

Product

Resources

About