EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Amortized template matching of molecular conformations from cryoelectron microscopy images using simulation-based inference

Lars Dingeldein, David Silva-Sánchez, Luke Evans iD, Edoardo D'Imprima iD, Nikolaus Grigorieff iD, Roberto Covino iD, Pilar Cossio iD

Proceedings of the National Academy of Sciences · 10.06.2025 · 122 (23)

DOI: 10.1073/pnas.2420158122 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 04.06.2025. Verfasst von Lars Dingeldein und 6 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Physics, Faculty of Physics, Goethe University Frankfurt, Frankfurt am Main 60438, Germany. Hauptthemen laut MEDLINE: Cryoelectron Microscopy, Molecular Conformation, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 17-mal zitiert (Crossref) und nennt 64 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Simons Foundation (SF), Howard Hughes Medical Institute (HHMI), Goethe-Universität Frankfurt am Main (Goethe-Universität).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Characterizing the conformational ensemble of biomolecular systems is key to understand their functions. Cryoelectron microscopy (cryo-EM) captures two-dimensional snapshots of biomolecular ensembles, giving in principle access to thermodynamics. However, these images are very noisy and show projections of the molecule in unknown orientations, making it very difficult to identify the biomolecule’s conformation in each individual image. Here, we introduce cryo-EM simulation-based inference (cryoSBI) to infer the conformations of biomolecules and the uncertainties associated with the inference from individual cryo-EM images. CryoSBI builds on simulation-based inference, a merger of physics-based simulations and probabilistic deep learning, allowing us to use Bayesian inference even when likelihoods are too expensive to calculate. We begin with an ensemble of conformations, templates from experiments, and molecular modeling, serving as structural hypotheses. We train a neural network approximating the Bayesian posterior using simulated images from these templates and then use it to accurately infer the conformation of the biomolecule from each experimental image. Training is only done once on simulations, and after that, it takes just a few milliseconds to make inference on an image, making cryoSBI suitable for arbitrarily large datasets and direct analysis on micrographs. CryoSBI eliminates the need to estimate particle pose and imaging parameters, significantly enhancing the computational speed compared to explicit likelihood methods. Importantly, we obtain interpretable machine learning models by integrating physics-based approaches with deep neural networks, ensuring that our results are transparent and reliable. We illustrate and benchmark cryoSBI on synthetic data and showcase its promise on experimental single-particle cryo-EM data.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cryoelectron MicroscopyMolecular ConformationImage Processing, Computer-AssistedBayes TheoremModels, MolecularDeep LearningNeural Networks, ComputerComputer Simulation

Stichwörter der Autor:innen

Bayesian inferencebiophysicscryo-EMmodelingtemplate matching

Autor:innen und Institutionen

  • Institute of Physics, Faculty of Physics, Goethe University Frankfurt, Frankfurt am Main 60438, Germany.
  • Frankfurt Institute for Advanced Studies, Frankfurt am Main 60438, Germany.
  • Department of Applied Mathematics, Yale University, New Haven, CT 06520.
  • Center for Computational Mathematics, Flatiron Institute, New York, NY 10010.
  • Istituto di Ricovero e Cura a Carattere Scientifico Humanitas Research Hospital, Unit of Correlative Light Emission Microscopy Core, Rozzano, Milan 20089, Italy.
  • RNA Therapeutics Institute, University of Massachusetts Chan Medical School, Worcester, MA 01605.

ORCID-Profile: Luke Evans, Edoardo D'Imprima, Nikolaus Grigorieff, Roberto Covino, Pilar Cossio

Förderung

  • Simons Foundation (SF)
  • Howard Hughes Medical Institute (HHMI)
  • Goethe-Universität Frankfurt am Main (Goethe-Universität)
  • LOEWE Zentrum AdRIA (AdRIA)

Zitieren

APA

Dingeldein, L., Silva-Sánchez, D., Evans, L., D'Imprima, E., Grigorieff, N., Covino, R., & Cossio, P. (2025). Amortized template matching of molecular conformations from cryoelectron microscopy images using simulation-based inference. Proceedings of the National Academy of Sciences, 122(23). https://doi.org/10.1073/pnas.2420158122

Vancouver

Dingeldein L, Silva-Sánchez D, Evans L, D'Imprima E, Grigorieff N, Covino R, et al. Amortized template matching of molecular conformations from cryoelectron microscopy images using simulation-based inference. Proceedings of the National Academy of Sciences. 2025;122(23). doi:10.1073/pnas.2420158122

BibTeX
@article{Dingeldein2025,
  author = {Dingeldein, Lars and Silva-Sánchez, David and Evans, Luke and D'Imprima, Edoardo and Grigorieff, Nikolaus and Covino, Roberto and Cossio, Pilar},
  title = {Amortized template matching of molecular conformations from cryoelectron microscopy images using simulation-based inference},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2025},
  volume = {122},
  number = {23},
  doi = {10.1073/pnas.2420158122}
}
Kostenlos starten →