Mohsen Sadeghi
Mohsen Sadeghi

Computational Scientist
AI4Science Researcher
Entrepreneur

Co-Founder and CTO at Neurorium
Guest Researcher at Zuse-Institut Berlin
Guest Researcher at AI4Science group, Freie Universität Berlin

Bio

I am a computational scientist with a specialization in the modeling and simulation of biological processes at the mesoscopic scale. My research plays a key role in quantitative integrative biology, bridging physics-based predictions with biological observables. Alongside this, I'm actively engaged in the development of deep learning techniques, which enable the automated analysis of microscopy and tomography imaging data.

For more details and examples, please visit the Research section.

I am also a co-founder of the science-based startup, Neurorium, dedicated to realizing the computational potential of living neurons.

Born in Tehran, Iran, I attended Sharif University of Technology, where I studied mechanical engineering, and later pursued a PhD in computational nanotechnology. During this phase, my contributions spanned various research areas, ranging from nonlinear mechanical analysis of nanostructures to molecular simulations of nano-confined fluids.

The Publications section enumerates my peer-reviewed works across these multifaceted disciplines.

In my free time, I enjoy swimming, bouldering, and reading. I am also a photography enthusiast, and you can check a selection of my works at my 500px portfolio.

Research

Computational model of synaptic vesicle docking

I joined forces with researchers from Max Delbrück Center and Charité — Universitätsmedizin Berlin, to look at the docking and fusion of synaptic vesicles in mouse hippocampal neurons.

Extremely detailed insights, derived from meticulous cryo-ET experiments and tomographic data analysis, effectively streamlined the construction of a computational model of the process.

My mesoscopic simulations unveiled the kinetics of the process in clear detail, highlighted the role of membrane-curving particles mimicking synaptotagmin C2B domain, and helped test a hypothesis on how the concentration of these particles can adversely affect the docking kinetics.

Read the rest of the details, including how we developed a Markov state model of the docking and fusion process, in the paper published in Nature Communications.

You can also check the press release from Charité — Universitätsmedizin Berlin (in German).

Mesoscopic model of the synaptic vesicle docking. © Mohsen Sadeghi.
Mesoscopic model of the synaptic vesicle docking. © Mohsen Sadeghi.

Mesoscopic model of human cytomegalovirus

In a recent collaboration with researchers from FMP Berlin and Charité Berlin, we built a comprehensive model of the human cytomegalovirus (HCMV) virion. We used information from quantitative proteomics as well as cross-linking mass spectrometry to model a wide range of viral and host proteins inside the virion. Models of glycoproteins (shown in red in the image) as well as the capsid (blue-green particles in the image) are built from cryoEM and/or AlphaFold2 predictions.

Large-scale simulations using this model lead to predictions about the role of pp150 protein in organizing the tegument proteins into a three-layer structure.

Further details of the model and simulations can be found in our Nature Microbiology paper.

You can also read more in the press release from FMP Berlin (in German).

Sectioned view of the coarse-grained model of HCMV. © Mohsen Sadeghi.
Sectioned view of the mesoscopic model of the HCMV. © Mohsen Sadeghi.

Particle-based modeling of biomembranes

For a while, my main research focus was on modeling membrane-involved biological processes. As my main simulation tool, I developed a highly coarse-grained model of bilayer membranes.

This model mimics the mechanics of the membrane via bonded interactions, with a force field optimized against the energy density predicted by the curvature elasticity of the membrane.

I further complemented this membrane model with a hydrodynamic coupling scheme, as detailed in my 2020 Nature Communications paper.

You can see the model in action by checking this Simularium visualization (by Allen Institute) of my simulation of a nanoparticle being wrapped by the membrane.

Finally, I implemented "force field masking" to incorporate membrane-bending peripheral proteins in the model (blue particles in the movie). Predictions about the interplay between flexible peripheral proteins and the dynamic membrane are the basis of several publications, including this Journal of Physical Chemistry Letters paper.

Aggregation of membrane-bending peripheral proteins on the membrane due to implicit membrane-mediated interactions. © Mohsen Sadeghi.
Endocytosis of spherical nanoparticles. © Mohsen Sadeghi.
Endocytosis of a spherical nanoparticle: nanoparticle is being wrapped by the fluid membrane as result of surface adhesion. © Mohsen Sadeghi.

Deep-SXT

In a collaborative research between Freie Universität Berlin and Helmholtz-Zentrum Berlin, we looked at tomograms of eukaryotic cells with nanometer-scale resolutions. These tomograms are obtained through soft x-ray imaging facilitated by a synchrotron light source (cryo-soft x-ray tomography).

I developed a software for semi-supervised deep learning-based segmentation and 3D reconstruction of these tomograms, based on few manual labels provided by experts.

My software, which was developed using TensorFlow, and the trained network weights are freely available under the MIT license. You can download the software from the GitHub repository github.com/noegroup/deep_sxt and follow the instructions for setting up your local pipeline.

Surface reconstruction of a cryo-soft x-ray tomogram using deep learning.
3D reconstruction of a cryo-soft x-ray tomogram using semi-supervised deep learning. © Mohsen Sadeghi.

In-Browser Molecular Dynamics (IBMD)

This is a hobby project of mine from 2014. I implemented a JavaScript-based molecular dynamics code that runs in your browser and has a graphical output of the trajectory as well as numerical and plot outputs of thermodynamic properties such as pressure. All outputs are in sync and reflect the current state of the simulation. In some cases, the simulation can be tweaked by changing the setpoint temperature or pressure, and particles can be added or removed from the simulation box by simply clicking on it.

Check it out here: In-Browser Molecular Dynamics (IBMD).

Publications

My Google Scholar profile

Journal articles

  1. Harris, S. A., Lattanzi, G., Rosa, A., Everaers, R., Papoian, G., Vidal, A., Brown, C., Liverpool, T., Pasos-Trejo, A., Sadeghi, M., Khalid, S., Ramanathan, A., Henrich, O., Dougan, L., Freyberg, Z., Pezeshkian, W., Sept, D., Geiger, B., Beccaria, R., … Carnovale, F., Meso-soup: A Community Approach to Building a Computational Description of the Biological Mesoscale, Biophysical Journal (2026).
  2. von Seggern, B., Sadeghi, M., “Methods for Inferring Interaction Potentials from Cross-Linking Mass Spectrometry Data”, arXiv (2026) 2606.05541.
  3. Kroll, J., Kravčenko, U., Sadeghi, M., Diebolder, C. A., Ivanov, L., Lubas, M., Sprink, T., Schacherl, M., Kudryashev, M., Rosenmund, C., “Dynamic nanoscale architecture of synaptic vesicle fusion in mouse hippocampal neurons”, Nature Communications (2025) 16:11131.
  4. Wehlitz, N., Sadeghi, M., Montefusco, A., Schütte, C., Pavliotis, G. A., Winkelmann, S., “Approximating particle-based clustering dynamics by stochastic PDEs”, SIAM Journal on Applied Dynamical Systems (2025) 24(2):1676661.
  5. Rentsch, J., Bandstra, S., Sezen, B., Sigrist, P. S., Bottanelli, F., Schmerl, B., Shoichet, S. A., Noé, F., Sadeghi, M., Ewers, H., “Sub-membrane actin rings compartmentalize the plasma membrane”, The Journal of Cell Biology (2024) 223 (4): e202310138.
  6. de Jong-Bolm, D., Sadeghi, M., Bao, G., Klaehn, G., Hoff, M., Mittelmeier, L., Basmanav, F. B., Opazo, F., Noé, F., Rizzoli, S. O., “Protein nanobarcodes enable single-step multiplexed fluorescence imaging”, PLoS Biology (2023) 21(12):e3002427.
  7. Bogdanow, B., Gruska, I., Mühlberg, L., Protze, J., Hohensee, S., Vetter, B., Lehmann, M., Sadeghi, M., Wiebusch, L., Liu, F., “Spatially resolved protein map of intact human cytomegalovirus virions”, Nature Microbiology (2023) 8:1732-1747.
  8. Dyhr, M., Sadeghi, M., Moynova, R., Knappe, C., Kepsutlu, B., Werner, S., Schneider, G., McNally, J., Noé, F., Ewers, H., “3D-surface reconstruction of cellular cryo-soft X-ray microscopy tomograms using semi-supervised deep learning”, Proceedings of the National Academy of Sciences (2023) 120 (24): e2209938120.
  9. Galama, M. M., Wu, H., Krämer, A., Sadeghi, M., Noé, F., “Stochastic approximation to MBAR and TRAM: batch-wise free energy estimation”, Journal of Chemical Theory and Computation (2022) 19(3):758-766.
  10. Sadeghi, M., “Formation of membrane invaginations by curvature-inducing peripheral proteins: free energy profiles, kinetics, and membrane-mediated effects”, bioRxiv (2022) 515891.
  11. Sadeghi, M., “Investigating the entropic nature of membrane-mediated interactions driving the aggregation of peripheral proteins”, Soft Matter (2022) 18:3917-3927.
  12. Sadeghi, M., Noé, F., “Thermodynamics and kinetics of aggregation of flexible peripheral membrane proteins”, The Journal of Physical Chemistry Letters (2021) 12:10497-10504.
  13. Sadeghi, M., Noé, F., “Aggregation of flexible membrane-bound proteins: thermodynamic and kinetic insights from large-scale simulations”, European Biophysics Journal with Biophysics Letters (2021) 50 (S1):171.
  14. Sadeghi, M., Noé, F., “Hydrodynamic coupling for particle-based solvent-free membrane models”, The Journal of Chemical Physics (2021) 155:114108.
  15. Sadeghi, M., Noé, F., “Large-scale simulation of biomembranes incorporating realistic kinetics into coarse-grained models”, Nature Communications (2020) 11:2951.
  16. Dimou, E., Cosentino, K., Platonova, E., Ros, U., Sadeghi, M., Kashyap, P., Katsinelos, T., Wegehingel, S., Noé, F., García-Sáez, A. J., Ewers, H., Nickel, W., “Single event visualization of unconventional secretion of FGF2”, The Journal of Cell Biology (2018) 218(2):683-699.
  17. Sadeghi, M., Weikl, T. R., Noé, F., “Particle-based membrane model for mesoscopic simulation of cellular dynamics”, The Journal of Chemical Physics (2018) 148:044901.
  18. Albrecht, D., Winterflood, C. M., Sadeghi, M., Tschager, T., Noé, F., Ewers, H., “Nanoscopic compartmentalization of membrane protein motion at the axon initial segment”, The Journal of Cell Biology (2016) 215(1):37-46.
  19. Sadeghi, M., Parsafar, G. A., “Density-induced molecular arrangements of water inside carbon nanotubes”, Physical Chemistry Chemical Physics (2013) 15:7379-7388.
  20. Sadeghi, M., Parsafar, G. A., “Toward an equation of state for water inside carbon nanotubes”, Journal of Physical Chemistry B (2012) 116:4943-4951.
  21. Niaki, S. A., Mianroodi, J. R., Sadeghi, M., Naghdabadi, R., “Dynamic and static fracture analyses of graphene sheets and carbon nanotubes”, Composite Structures (2012) 94(8):2365-2372.
  22. Montazeri, A., Sadeghi, M., Naghdabadi, R., Rafii-Tabar, H., “Multiscale modeling of the effect of carbon nanotube orientation on the shear deformation properties of reinforced polymer-based composites”, Physics Letters A (2011) 375:1588-1597.
  23. Sadeghi, M., Naghdabadi, R., “Nonlinear vibration analysis of single-layer graphene sheets”, Nanotechnology (2010) 21:105705.
  24. Montazeri, A., Sadeghi, M., Naghdabadi, R., Rafii-Tabar, H., “Computational modeling of the transverse-isotropic elastic properties of single-walled carbon nanotubes”, Computational Materials Science (2010) 49:544-551.
  25. Sadeghi, M., Naghdabadi, R., “Stability analysis of carbon nanotubes using a hybrid atomistic-structural element”, International Journal of Nanomanufacturing (2009) 5(3/4):366-375
  26. Sadeghi, M., Ozmaian, M., Naghdabadi, R., “Stability analysis of carbon nanotubes under electric fields and compressive loading”, Journal of Physics D: Applied Physics (2008) 41:205411.

Books & Chapters

  1. Sadeghi, M., Rosenberger, D., 2024 “Dynamic framework for large-scale modeling of membranes and peripheral proteins” in MIE 701: Biophysical Approaches for the Study of Membrane Structure Part B, ed. Markus Deserno, Tobias Baumgart, (Elsevier: Methods in Enzymology, Academic Press).
  2. Sadeghi, M. 2010 “Nonlinear mechanical analysis of carbon nanostructures: Numerical simulation using the hybrid atomistic-structural method” (Saarbrücken: Lambert Academic Publishing).

Contact

If you:

  • would like to know more about me or what I do,
  • are interested in a collaboration,
  • have a suggestion or remark about my work or this website,
  • are looking to hire me,

I'd love to talk to you!

Feel free to send me an email to either of these addresses:
sadeghi [at] zib.de
mohsen.sadeghi [at] fu-berlin.de

Or find me on LinkedIn.