Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction
Abstract Structure-based machine learning algorithms have been utilized to predict the properties of protein-protein interaction (PPI) complexes, such as binding affinity, which is critical for understanding biological mechanisms and disease treatments. While most existing algorithms represent PPI c...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-11-01
|
| Series: | Nature Communications |
| Online Access: | https://doi.org/10.1038/s41467-024-53583-w |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850194783154208768 |
|---|---|
| author | Yang Yue Shu Li Yihua Cheng Lie Wang Tingjun Hou Zexuan Zhu Shan He |
| author_facet | Yang Yue Shu Li Yihua Cheng Lie Wang Tingjun Hou Zexuan Zhu Shan He |
| author_sort | Yang Yue |
| collection | DOAJ |
| description | Abstract Structure-based machine learning algorithms have been utilized to predict the properties of protein-protein interaction (PPI) complexes, such as binding affinity, which is critical for understanding biological mechanisms and disease treatments. While most existing algorithms represent PPI complex graph structures at the atom-scale or residue-scale, these representations can be computationally expensive or may not sufficiently integrate finer chemical-plausible interaction details for improving predictions. Here, we introduce MCGLPPI, a geometric representation learning framework that combines graph neural networks (GNNs) with MARTINI molecular coarse-grained (CG) models to predict PPI overall properties accurately and efficiently. Extensive experiments on three types of downstream PPI property prediction tasks demonstrate that at the CG-scale, MCGLPPI achieves competitive performance compared with the counterparts at the atom- and residue-scale, but with only a third of computational resource consumption. Furthermore, CG-scale pre-training on protein domain-domain interaction structures enhances its predictive capabilities for PPI tasks. MCGLPPI offers an effective and efficient solution for PPI overall property predictions, serving as a promising tool for the large-scale analysis of biomolecular interactions. |
| format | Article |
| id | doaj-art-58efcb3b577b42d5bc3ea273c6d79174 |
| institution | OA Journals |
| issn | 2041-1723 |
| language | English |
| publishDate | 2024-11-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Nature Communications |
| spelling | doaj-art-58efcb3b577b42d5bc3ea273c6d791742025-08-20T02:13:55ZengNature PortfolioNature Communications2041-17232024-11-0115111410.1038/s41467-024-53583-wIntegration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property predictionYang Yue0Shu Li1Yihua Cheng2Lie Wang3Tingjun Hou4Zexuan Zhu5Shan He6School of Computer Science, The University of Birmingham, EdgbastonMacao Polytechnic UniversitySchool of Computer Science, The University of Birmingham, EdgbastonBone Marrow Transplantation Center of the First Affiliated Hospital, Institute of Immunology, Zhejiang University School of MedicineCollege of Pharmaceutical Sciences, Zhejiang UniversityNational Engineering Laboratory for Big Data System Computing Technology, Shenzhen UniversitySchool of Computer Science, The University of Birmingham, EdgbastonAbstract Structure-based machine learning algorithms have been utilized to predict the properties of protein-protein interaction (PPI) complexes, such as binding affinity, which is critical for understanding biological mechanisms and disease treatments. While most existing algorithms represent PPI complex graph structures at the atom-scale or residue-scale, these representations can be computationally expensive or may not sufficiently integrate finer chemical-plausible interaction details for improving predictions. Here, we introduce MCGLPPI, a geometric representation learning framework that combines graph neural networks (GNNs) with MARTINI molecular coarse-grained (CG) models to predict PPI overall properties accurately and efficiently. Extensive experiments on three types of downstream PPI property prediction tasks demonstrate that at the CG-scale, MCGLPPI achieves competitive performance compared with the counterparts at the atom- and residue-scale, but with only a third of computational resource consumption. Furthermore, CG-scale pre-training on protein domain-domain interaction structures enhances its predictive capabilities for PPI tasks. MCGLPPI offers an effective and efficient solution for PPI overall property predictions, serving as a promising tool for the large-scale analysis of biomolecular interactions.https://doi.org/10.1038/s41467-024-53583-w |
| spellingShingle | Yang Yue Shu Li Yihua Cheng Lie Wang Tingjun Hou Zexuan Zhu Shan He Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction Nature Communications |
| title | Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction |
| title_full | Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction |
| title_fullStr | Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction |
| title_full_unstemmed | Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction |
| title_short | Integration of molecular coarse-grained model into geometric representation learning framework for protein-protein complex property prediction |
| title_sort | integration of molecular coarse grained model into geometric representation learning framework for protein protein complex property prediction |
| url | https://doi.org/10.1038/s41467-024-53583-w |
| work_keys_str_mv | AT yangyue integrationofmolecularcoarsegrainedmodelintogeometricrepresentationlearningframeworkforproteinproteincomplexpropertyprediction AT shuli integrationofmolecularcoarsegrainedmodelintogeometricrepresentationlearningframeworkforproteinproteincomplexpropertyprediction AT yihuacheng integrationofmolecularcoarsegrainedmodelintogeometricrepresentationlearningframeworkforproteinproteincomplexpropertyprediction AT liewang integrationofmolecularcoarsegrainedmodelintogeometricrepresentationlearningframeworkforproteinproteincomplexpropertyprediction AT tingjunhou integrationofmolecularcoarsegrainedmodelintogeometricrepresentationlearningframeworkforproteinproteincomplexpropertyprediction AT zexuanzhu integrationofmolecularcoarsegrainedmodelintogeometricrepresentationlearningframeworkforproteinproteincomplexpropertyprediction AT shanhe integrationofmolecularcoarsegrainedmodelintogeometricrepresentationlearningframeworkforproteinproteincomplexpropertyprediction |