This project develops machine-learning methods for XFEL imaging with a focus on Coulomb explosions. Coulomb explosions are the rapid and violent fragmentation of molecules caused by intense internal electrostatic repulsion. Intense laser ionization can be used to drive this phenomenon; in Coulomb Explosion Imagine (CEI) the resulting ion spatial and momentum distributions encode the original molecular geometry [1] and can be used for structure reconstruction and time-resolved, single-molecule imaging. While coincident momentum measurements limit CEI to small molecules due to detector saturation, theory suggests that ion position patterns alone may still distinguish protein conformations [2-3], making CEI feasible for large biomolecules despite complex many-body dynamics. We will use ML to analyse simulations, in particular to create a robust computational method to extract structural information from Coulomb explosions, supporting the development and interpretation of future XFEL experiments. To address the aforementioned challenges, we are developing ML computational to: - Characterize and simulate laser-induced ion explosions of proteins - Preprocess ion signatures from simulated and experimental datasets using unsupervised ML algorithms (PCA, tSNE, UMAP) - Gain structural knowledge from molecular fragmentation ion maps. We employ molecular fragmentation simulations, including those performed with MolDStruct [4] and other in-house developed codes, to generate synthetic ion maps. These maps help address critical challenges, such as determining sample orientation and understanding fragmentation pathways. These results are very recent and we are just now standarour data analysis procedures. References: [1.] Richard, B., et al. (2025) Imaging collective quantum fluctuations of the structure of a complex molecule. Science 389.6760, pp. 650–654 [2.] André, T., De Santis, E., Timneanu, N., & Caleman, C. (2025). Partial Orientation Retrieval of Proteins From Coulomb Explosions. Accepted in Physical Review Letters. [3.] André, T., Dawod, I., Cardoch, S., De Santis, E., Tîmneanu, N., & Caleman, C. (2025). Protein Structure Classification Based on X-Ray-Laser-Induced Coulomb Explosion. Physical Review Letters, 134(12), 128403. [4.] Dawod, I., et. al. (2024); MolDStruct: Modeling the dynamics and structure of matter exposed to ultrafast x-ray lasers with hybrid collisional-radiative/molecular dynamics. J. Chem. Phys. 14 May 2024; 160 (18): 184112