3DProtDTA: a deep learning model for drug-target affinity prediction based on residue-level protein graphs

Voitsitskyi, Taras; Stratiichuk, Roman; Koleiev, Ihor; Popryho, Leonid; Ostrovsky, Zakhar; Henitsoi, Pavlo; Khropachov, Ivan; Vozniak, Volodymyr; Zhytar, Roman; Nechepurenko, Diana; Yesylevskyy, Semen O.; Nafiiev, Alan; Starosyla, Sergiy A.

doi:10.1039/d3ra00281k

Cited by 11 publications

(11 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use mean-squared errors (MSE), concordance index (CI), and r 2 m to evaluate the performance. Baseline methods include KronRLS, SimBoost, SimCNN-DTA, DeepDTA, WideDTA, AttentionDTA, MATT-DTI, GraphDTA, FusionDTA, BiCompDTA, and their results are taken from the work of Voitsitskyi et al [39] and Kalemati et al [68].…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Multimodal Protein Representation Learning and Target-aware Variational Auto-encoders for Protein-binding Ligand Generation

Ngo,

2023

Preprint

View full text Add to dashboard Cite

Without knowledge of specific pockets, generating ligands based on the global structure of a protein target plays a crucial role in drug discovery as it helps reduce the search space for potential drug-like candidates in the pipeline. However, contemporary methods require optimizing tailored networks for each protein, which is arduous and costly. To address this issue, we introduce TargetVAE, a target-aware variational auto-encoder that generates ligands with desirable properties including high binding affinity and high synthesizability to arbitrary target proteins , guided by a multimodal deep neural network built based on geometric and sequence models, named Protein Multimodal Network (PMN), as the prior for the generative model. PMN unifies different representations of proteins (e.g., primary structure-sequence of amino acids, 3D tertiary structure, and residue-level graph) into a single representation. Our multimodal architecture learns from the entire protein structure and is able to capture their sequential, topological, and geometrical information by utilizing language modeling, graph neural networks and geometric deep learning. We showcase the superiority of our approach by conducting extensive experiments and evaluations, including predicting protein-ligand binding affinity in PBDBind v2020, DAVIS and KIBA datasets as well as the assessment of generative model quality, ligand generation for unseen targets, and docking score computation. Empirical results demonstrate the promising and competitive performance of our proposed approach. Our software package is publicly available at https://github.com/HySonLab/ Ligand Generation

show abstract

Section: Methodsmentioning

confidence: 99%

“…To ensure a fair comparison with the baselines, we train efficient Transformers from scratch on the protein sequences in KIBA and DAVIS, instead of extracting sequence embeddings from ESM. Similarly to previous works [39], we also augment Morgan Fingerprints as additional inputs of the ligands.…”

Section: Implementation Detailsmentioning

confidence: 99%

Multimodal Protein Representation Learning and Target-aware Variational Auto-encoders for Protein-binding Ligand Generation

Ngo,

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…On the other hand, Nguyen et al [38]; Voitsitskyi et al [39] utilize GNNs to learn the representations of the molecular graphs and protein structures, which are superior to the previous methods operating on sequences and handcrafted features. However, the common limitation of these above approaches is that none of them covers a wide range of representations of proteins.…”

Section: Related Work 21 Protein-ligand Binding Predictionmentioning

confidence: 99%

“…The data that support the findings of this study are openly available at the following URL/DOI: https:// github.com/HySonLab/Ligand_Generation. The data that we used for this study is publicly available at https://github.com/vtarasv/3d-prot-dta [39] and https://github.com/HannesStark/EquiBind [17]. We release our data processing pipeline and software along with the installation instructions at https://github.com/ HySonLab/Ligand_Generation.…”

Section: Data Availability Statementmentioning

confidence: 99%

Multimodal protein representation learning and target-aware variational auto-encoders for protein-binding ligand generation

Khang Ngo,

Son Hy

2024

Mach. Learn.: Sci. Technol.

View full text Add to dashboard Cite

Without knowledge of specific pockets, generating ligands based on the global structure of a protein target plays a crucial role in drug discovery as it helps reduce the search space for potential drug-like candidates in the pipeline. However, contemporary methods require optimizing tailored networks for each protein, which is arduous and costly. To address this issue, we introduce TargetVAE, a target-aware variational auto-encoder that generates ligands with desirable properties including high binding affinity and high synthesizability to arbitrary target proteins, guided by a multimodal deep neural network built based on geometric and sequence models, named Protein Multimodal Network (PMN), as the prior for the generative model. PMN unifies different representations of proteins (e.g., primary structure - sequence of amino acids, 3D tertiary structure, and residue-level graph) into a single representation. Our multimodal architecture learns from the entire protein structure and is able to capture their sequential, topological, and geometrical information by utilizing language modeling, graph neural networks, and geometric deep learning. We showcase the superiority of our approach by conducting extensive experiments and evaluations, including predicting protein-ligand binding affinity in the PBDBind v2020 dataset as well as the assessment of generative model quality, ligand generation for unseen targets, and docking score computation. Empirical results demonstrate the promising and competitive performance of our proposed approach. Our software package is publicly available at https://github.com/HySonLab/Ligand_Generation

show abstract

“…To measure the performance of our model in this task, we calculated the average accuracy across the 10-fold cross-validation setup. [51] 0.379 0.871 0.407 0.411 0.782 0.342 SimBoost [52] 0.282 0.872 0.640 0.220 0.836 0.629 SmCNN-DTA [53] 0.319 0.852 0.590 0.274 0.821 0.573 DeepDTAY [54] 0.261 0.878 0.630 0.194 0.863 0.673 WideDTA [55] 0.886 0.202 -0.875 0.179 -AttentionDTA [56] 0.216 0.893 0.670 0.155 0.882 0.755 MATT-DTI [57] 0.227 0.891 0.653 0.150 0.882 0.756 GraphDTA [58] 0.258 0.884 0.656 0.162 0.879 0.736 FusionDTA [59] 0.220 0.903 0.666 0.167 0.891 0.699 BiCompDTA [60] a. Problem statement Predicting the stability of protein mutations is a critical task in understanding the intricate interplay between genetic variations and protein structure and function.…”

Section: Enzyme Identificationmentioning

confidence: 99%

Multimodal Pretraining for Unsupervised Protein Representation Learning

Nguyen,

2023

Preprint

View full text Add to dashboard Cite

In this paper, we introduce a framework of symmetry-preserving multimodal pretraining to learn a unified representation of proteins in an unsupervised manner, encompassing both primary and tertiary structures. Our approach involves proposing specific pretraining methods for sequences, graphs, and 3D point clouds associated with each protein structure, leveraging the power of large language models and generative models. We present a novel way to combining representations from multiple sources of information into a single global representation for proteins. We carefully analyze the performance of our framework in the pretraining tasks. For the fine-tuning tasks, our experiments have shown that our new multimodal representation can achieve competitive results in protein-ligand binding affinity prediction, protein fold classification, enzyme identification and mutation stability prediction. We expect that this work will accelerate future research in proteins. Our source code in PyTorch deep learning framework is publicly available athttps://github.com/HySonLab/Protein_Pretrain.

show abstract

3DProtDTA: a deep learning model for drug-target affinity prediction based on residue-level protein graphs

Abstract: We propose a new deep learning DTA model 3DProtDTA, which utilises AlphaFold structure predictions in conjunction with the graph representation of proteins.

Cited by 11 publications

References 47 publications

Multimodal Protein Representation Learning and Target-aware Variational Auto-encoders for Protein-binding Ligand Generation

Multimodal Protein Representation Learning and Target-aware Variational Auto-encoders for Protein-binding Ligand Generation

Multimodal protein representation learning and target-aware variational auto-encoders for protein-binding ligand generation

Multimodal Pretraining for Unsupervised Protein Representation Learning

Contact Info

Product

Resources

About