Interpreting chemical crosslinks: Score-based approaches and deep neural networks
Abstract
Chemical cross-linking coupled with mass spectrometry (XL-MS) has become a powerful tool for probing residue-level proximities within macromolecular assemblies. By providing sparse but informative distance restraints, XL-MS can be integrated with electron microscopy and domain-level high-resolution structures to model the architecture of protein complexes. Unlike X-ray crystallography, electron microscopy, or solid-state Nuclear Magnetic Resonance (NMR), XL-MS can be applied under near-physiological conditions, scaled to large modular systems, and performed at higher throughput. In this review, we highlight recent advances in the field, with particular emphasis on the impact of AI-driven structure prediction. As an illustration, we describe a hybrid protocol that combines the Integrative Modeling Platform (IMP) with the deep neural network Chai-1 to dock and refine the helicase Dbp10 on a transient ribosome biogenesis intermediate using XL-MS restraints. • Chemical cross-linking coupled with mass spectrometry (XL-MS) maps residue proximities within protein complexes. • The interpretation of XL-MS data requires probabilistic restraints that account for structural flexibility and data noise. • Deep neural networks (DNNs) incorporate distance information into pair representations to guide structure prediction. • We demonstrated that score-based methods can be effectively combined with DNNs.
Related
- Edge-Aware Graph Attention Networks for Interpreting Biophysical Mechanisms from Molecular Dynamics Simulations
- Serum IgM crosslinks hepatitis B subviral particles into supramolecular immune complexes
- Patient classification using single-cell RNA-sequencing data with deep learning convolutional neural networks
- Representation learning for low-data protein fitness prediction using sequence–structure embeddings and graph neural networks
- A comparative analysis of graph neural networks and classical machine learning for carbapenem-resistance prediction and genomic interpretation in Acinetobacter baumannii
BibTeX
@article{chen2026interpreting,
title = {{Interpreting chemical crosslinks: Score-based approaches and deep neural networks}},
author = {Xingyu Chen and Kai Steffen Stroh and Jan Erzberger and Florian Stengel and Riccardo Pellarin},
journal = {Current Opinion in Structural Biology},
year = {2026},
month = nov,
doi = {10.1016/j.sbi.2026.103237},
url = {https://hal.science/hal-05850652v1},
note = {CC BY},
}
RIS
TY - JOUR
TI - Interpreting chemical crosslinks: Score-based approaches and deep neural networks
AU - Xingyu Chen
AU - Kai Steffen Stroh
AU - Jan Erzberger
AU - Florian Stengel
AU - Riccardo Pellarin
PY - 2026
DA - 2026/11/03
JO - Current Opinion in Structural Biology
DO - 10.1016/j.sbi.2026.103237
UR - https://hal.science/hal-05850652v1
AB - Chemical cross-linking coupled with mass spectrometry (XL-MS) has become a powerful tool for probing residue-level proximities within macromolecular assemblies. By providing sparse but informative distance restraints, XL-MS can be integrated with electron microscopy and domain-level high-resolution structures to model the architecture of protein complexes. Unlike X-ray crystallography, electron microscopy, or solid-state Nuclear Magnetic Resonance (NMR), XL-MS can be applied under near-physiological conditions, scaled to large modular systems, and performed at higher throughput. In this review, we highlight recent advances in the field, with particular emphasis on the impact of AI-driven structure prediction. As an illustration, we describe a hybrid protocol that combines the Integrative Modeling Platform (IMP) with the deep neural network Chai-1 to dock and refine the helicase Dbp10 on a transient ribosome biogenesis intermediate using XL-MS restraints. • Chemical cross-linking coupled with mass spectrometry (XL-MS) maps residue proximities within protein complexes. • The interpretation of XL-MS data requires probabilistic restraints that account for structural flexibility and data noise. • Deep neural networks (DNNs) incorporate distance information into pair representations to guide structure prediction. • We demonstrated that score-based methods can be effectively combined with DNNs.
SN - 0959-440X
ER -