Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention Networks
Computational prediction of ligand–target interactions is a crucial part of modern drug discovery as it helps to bypass high costs and labor demands of in vitro and in vivo screening. As the wealth of bioactivity data accumulates, it provides opportunities for the development of deep learning (DL) m...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Molecules |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1420-3049/27/16/5114 |
| _version_ | 1827626346643193856 |
|---|---|
| author | Mohit Pandey Mariia Radaeva Hazem Mslati Olivia Garland Michael Fernandez Martin Ester Artem Cherkasov |
| author_facet | Mohit Pandey Mariia Radaeva Hazem Mslati Olivia Garland Michael Fernandez Martin Ester Artem Cherkasov |
| author_sort | Mohit Pandey |
| collection | DOAJ |
| description | Computational prediction of ligand–target interactions is a crucial part of modern drug discovery as it helps to bypass high costs and labor demands of in vitro and in vivo screening. As the wealth of bioactivity data accumulates, it provides opportunities for the development of deep learning (DL) models with increasing predictive powers. Conventionally, such models were either limited to the use of very simplified representations of proteins or ineffective voxelization of their 3D structures. Herein, we present the development of the PSG-BAR (Protein Structure Graph-Binding Affinity Regression) approach that utilizes 3D structural information of the proteins along with 2D graph representations of ligands. The method also introduces attention scores to selectively weight protein regions that are most important for ligand binding. Results: The developed approach demonstrates the state-of-the-art performance on several binding affinity benchmarking datasets. The attention-based pooling of protein graphs enables identification of surface residues as critical residues for protein–ligand binding. Finally, we validate our model predictions against an experimental assay on a viral main protease (Mpro)—the hallmark target of SARS-CoV-2 coronavirus. |
| first_indexed | 2024-03-09T12:52:28Z |
| format | Article |
| id | doaj.art-f60c71814bfc4a21be54e803f827c010 |
| institution | Directory Open Access Journal |
| issn | 1420-3049 |
| language | English |
| last_indexed | 2024-03-09T12:52:28Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Molecules |
| spelling | doaj.art-f60c71814bfc4a21be54e803f827c0102023-11-30T22:04:23ZengMDPI AGMolecules1420-30492022-08-012716511410.3390/molecules27165114Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention NetworksMohit Pandey0Mariia Radaeva1Hazem Mslati2Olivia Garland3Michael Fernandez4Martin Ester5Artem Cherkasov6Vancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC V6T 1Z2, CanadaVancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC V6T 1Z2, CanadaVancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC V6T 1Z2, CanadaVancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC V6T 1Z2, CanadaVancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC V6T 1Z2, CanadaSchool of Computing Science, Simon Fraser University, Burnaby, BC V5A 1S6, CanadaVancouver Prostate Centre, Department of Urologic Sciences, University of British Columbia, Vancouver, BC V6T 1Z2, CanadaComputational prediction of ligand–target interactions is a crucial part of modern drug discovery as it helps to bypass high costs and labor demands of in vitro and in vivo screening. As the wealth of bioactivity data accumulates, it provides opportunities for the development of deep learning (DL) models with increasing predictive powers. Conventionally, such models were either limited to the use of very simplified representations of proteins or ineffective voxelization of their 3D structures. Herein, we present the development of the PSG-BAR (Protein Structure Graph-Binding Affinity Regression) approach that utilizes 3D structural information of the proteins along with 2D graph representations of ligands. The method also introduces attention scores to selectively weight protein regions that are most important for ligand binding. Results: The developed approach demonstrates the state-of-the-art performance on several binding affinity benchmarking datasets. The attention-based pooling of protein graphs enables identification of surface residues as critical residues for protein–ligand binding. Finally, we validate our model predictions against an experimental assay on a viral main protease (Mpro)—the hallmark target of SARS-CoV-2 coronavirus.https://www.mdpi.com/1420-3049/27/16/5114deep learningdrug–target interactiongraph attention networkcomputer-aided drug discoveryprotein–ligand bindingvirtual screening |
| spellingShingle | Mohit Pandey Mariia Radaeva Hazem Mslati Olivia Garland Michael Fernandez Martin Ester Artem Cherkasov Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention Networks Molecules deep learning drug–target interaction graph attention network computer-aided drug discovery protein–ligand binding virtual screening |
| title | Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention Networks |
| title_full | Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention Networks |
| title_fullStr | Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention Networks |
| title_full_unstemmed | Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention Networks |
| title_short | Ligand Binding Prediction Using Protein Structure Graphs and Residual Graph Attention Networks |
| title_sort | ligand binding prediction using protein structure graphs and residual graph attention networks |
| topic | deep learning drug–target interaction graph attention network computer-aided drug discovery protein–ligand binding virtual screening |
| url | https://www.mdpi.com/1420-3049/27/16/5114 |
| work_keys_str_mv | AT mohitpandey ligandbindingpredictionusingproteinstructuregraphsandresidualgraphattentionnetworks AT mariiaradaeva ligandbindingpredictionusingproteinstructuregraphsandresidualgraphattentionnetworks AT hazemmslati ligandbindingpredictionusingproteinstructuregraphsandresidualgraphattentionnetworks AT oliviagarland ligandbindingpredictionusingproteinstructuregraphsandresidualgraphattentionnetworks AT michaelfernandez ligandbindingpredictionusingproteinstructuregraphsandresidualgraphattentionnetworks AT martinester ligandbindingpredictionusingproteinstructuregraphsandresidualgraphattentionnetworks AT artemcherkasov ligandbindingpredictionusingproteinstructuregraphsandresidualgraphattentionnetworks |