Boosting Protein Graph Representations through Static-Dynamic Fusion

Pengkang Guo, Bruno Correia, Pierre Vandergheynst, Daniel Probst
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:20777-20792, 2025.

Abstract

Machine learning for protein modeling faces significant challenges due to proteins’ inherently dynamic nature, yet most graph-based machine learning methods rely solely on static structural information. Recently, the growing availability of molecular dynamics trajectories provides new opportunities for understanding the dynamic behavior of proteins; however, computational methods for utilizing this dynamic information remain limited. We propose a novel graph representation that integrates both static structural information and dynamic correlations from molecular dynamics trajectories, enabling more comprehensive modeling of proteins. By applying relational graph neural networks (RGNNs) to process this heterogeneous representation, we demonstrate significant improvements over structure-based approaches across three distinct tasks: atomic adaptability prediction, binding site detection, and binding affinity prediction. Our results validate that combining static and dynamic information provides complementary signals for understanding protein-ligand interactions, offering new possibilities for drug design and structural biology applications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-guo25b, title = {Boosting Protein Graph Representations through Static-Dynamic Fusion}, author = {Guo, Pengkang and Correia, Bruno and Vandergheynst, Pierre and Probst, Daniel}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {20777--20792}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/guo25b/guo25b.pdf}, url = {https://proceedings.mlr.press/v267/guo25b.html}, abstract = {Machine learning for protein modeling faces significant challenges due to proteins’ inherently dynamic nature, yet most graph-based machine learning methods rely solely on static structural information. Recently, the growing availability of molecular dynamics trajectories provides new opportunities for understanding the dynamic behavior of proteins; however, computational methods for utilizing this dynamic information remain limited. We propose a novel graph representation that integrates both static structural information and dynamic correlations from molecular dynamics trajectories, enabling more comprehensive modeling of proteins. By applying relational graph neural networks (RGNNs) to process this heterogeneous representation, we demonstrate significant improvements over structure-based approaches across three distinct tasks: atomic adaptability prediction, binding site detection, and binding affinity prediction. Our results validate that combining static and dynamic information provides complementary signals for understanding protein-ligand interactions, offering new possibilities for drug design and structural biology applications.} }
Endnote
%0 Conference Paper %T Boosting Protein Graph Representations through Static-Dynamic Fusion %A Pengkang Guo %A Bruno Correia %A Pierre Vandergheynst %A Daniel Probst %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-guo25b %I PMLR %P 20777--20792 %U https://proceedings.mlr.press/v267/guo25b.html %V 267 %X Machine learning for protein modeling faces significant challenges due to proteins’ inherently dynamic nature, yet most graph-based machine learning methods rely solely on static structural information. Recently, the growing availability of molecular dynamics trajectories provides new opportunities for understanding the dynamic behavior of proteins; however, computational methods for utilizing this dynamic information remain limited. We propose a novel graph representation that integrates both static structural information and dynamic correlations from molecular dynamics trajectories, enabling more comprehensive modeling of proteins. By applying relational graph neural networks (RGNNs) to process this heterogeneous representation, we demonstrate significant improvements over structure-based approaches across three distinct tasks: atomic adaptability prediction, binding site detection, and binding affinity prediction. Our results validate that combining static and dynamic information provides complementary signals for understanding protein-ligand interactions, offering new possibilities for drug design and structural biology applications.
APA
Guo, P., Correia, B., Vandergheynst, P. & Probst, D.. (2025). Boosting Protein Graph Representations through Static-Dynamic Fusion. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:20777-20792 Available from https://proceedings.mlr.press/v267/guo25b.html.

Related Material