AlloPool is a deep learning framework that infers protein allostery from molecular dynamics simulations
- 1Interfaculty Institute of Bioengineering, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland
- 2Ludwig Institute for Cancer Research Lausanne, Lausanne, Switzerland
Abstract
Recent advances in artificial intelligence have transformed protein structure prediction and design. However, protein function is governed not only by static structures but also by the conformational dynamics that allow proteins to access distinct functional states. Predicting these dynamic transitions, central to many biological processes, remains challenging due to the scarcity of high-resolution experimental data, which limits the training of machine-learning models for dynamic and energetic property prediction. Here, we present AlloPool, a graph neural network (GNN)-based framework that interprets molecular dynamics simulations by iteratively pruning residue–residue interactions to uncover minimal, time-resolved interaction networks that govern protein conformational dynamics and structural responses to chemical or mechanical perturbations, collectively known as allostery. By integrating temporal attention with graph aggregation, AlloPool learns evolving interaction graphs from equilibrium and non-equilibrium molecular dynamics simulations, enabling accurate reconstruction of dynamic trajectories and the interaction networks that drive conformational transitions. Validated across diverse dynamic protein systems, including binding domains, mechanosensors, signaling receptors, and enzymes, AlloPool maps allosteric communication pathways, predicts the effects of ligand binding, mechanical forces, and mutations, discovers transient dynamic states and outperforms existing machine-learning approaches in dynamic trajectory reconstruction. This advance provides a general framework for interpreting protein dynamics with broad implications for drug discovery, synthetic biology, and protein engineering.
Protein function depends on dynamic conformational changes that remain difficult to predict despite advances in AI-based structure modeling. This study develops a graph neural network-based framework called AlloPool that accurately extracts allosteric communication networks from molecular dynamics trajectories.
Related
- Development of A Modeling Framework for Characterizing Protein–Polymer Bioconjugates with Enhanced Sampling Molecular Dynamics Simulations
- GONet: A Molecular Interaction-Guided Graph Learning Framework for Multi-Omics Cancer Classification
- AET5: A transcriptome-guided molecular generation framework with contrastive self-supervised learning
- Edge-Aware Graph Attention Networks for Interpreting Biophysical Mechanisms from Molecular Dynamics Simulations
- Isoform-selective modulation of peroxisome proliferator-activated receptors by quercetin and apigenin revealed through molecular simulations
BibTeX
@article{marfoglia2026allopool,
title = {{AlloPool is a deep learning framework that infers protein allostery from molecular dynamics simulations}},
author = {Matthieu Marfoglia and Miguel A. Pedraza-Joya and Lucas Guirardel and Aisima Chatzi Souleiman and Patrick Barth},
journal = {PLOS Biology},
year = {2026},
month = sep,
doi = {10.1371/journal.pbio.3004002},
url = {https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3004002},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - AlloPool is a deep learning framework that infers protein allostery from molecular dynamics simulations
AU - Matthieu Marfoglia
AU - Miguel A. Pedraza-Joya
AU - Lucas Guirardel
AU - Aisima Chatzi Souleiman
AU - Patrick Barth
PY - 2026
DA - 2026/09/21
JO - PLOS Biology
DO - 10.1371/journal.pbio.3004002
UR - https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3004002
AB - Recent advances in artificial intelligence have transformed protein structure prediction and design. However, protein function is governed not only by static structures but also by the conformational dynamics that allow proteins to access distinct functional states. Predicting these dynamic transitions, central to many biological processes, remains challenging due to the scarcity of high-resolution experimental data, which limits the training of machine-learning models for dynamic and energetic property prediction. Here, we present AlloPool, a graph neural network (GNN)-based framework that interprets molecular dynamics simulations by iteratively pruning residue–residue interactions to uncover minimal, time-resolved interaction networks that govern protein conformational dynamics and structural responses to chemical or mechanical perturbations, collectively known as allostery. By integrating temporal attention with graph aggregation, AlloPool learns evolving interaction graphs from equilibrium and non-equilibrium molecular dynamics simulations, enabling accurate reconstruction of dynamic trajectories and the interaction networks that drive conformational transitions. Validated across diverse dynamic protein systems, including binding domains, mechanosensors, signaling receptors, and enzymes, AlloPool maps allosteric communication pathways, predicts the effects of ligand binding, mechanical forces, and mutations, discovers transient dynamic states and outperforms existing machine-learning approaches in dynamic trajectory reconstruction. This advance provides a general framework for interpreting protein dynamics with broad implications for drug discovery, synthetic biology, and protein engineering. Protein function depends on dynamic conformational changes that remain difficult to predict despite advances in AI-based structure modeling. This study develops a graph neural network-based framework called AlloPool that accurately extracts allosteric communication networks from molecular dynamics trajectories.
SN - 1545-7885
ER -