The use of artificial intelligence to process diffraction images is challenged by the need to assemble large and precisely designed training data sets. To address this, a codebase called Resonet was developed for synthesizing diffraction data and training residual neural networks on these data. Here, two per-pattern capabilities of Resonet are demonstrated: (i) interpretation of crystal resolution and (ii) identification of overlapping lattices. Resonet was tested across a compilation of diffraction images from synchrotron experiments and X-ray free-electron laser experiments. Crucially, these models readily execute on graphics processing units and can thus significantly outperform conventional algorithms. While Resonet is currently utilized to provide real-time feedback for macromolecular crystallography users at the Stanford Synchrotron Radiation Lightsource, its simple Python-based interface makes it easy to embed in other processing frameworks. This work highlights the utility of physics-based simulation for training deep neural networks and lays the groundwork for the development of additional models to enhance diffraction collection and analysis.
Mendez, Derek, Holton, James M., Lyubimov, Artem Y., et al., "Deep residual networks for crystallography trained on synthetic data," Acta Crystallographica. Section D. Structural Biology 80, no. 1 (2024), https://doi.org/10.1107/S2059798323010586
@article{osti_2274863,
author = {Mendez, Derek and Holton, James M. and Lyubimov, Artem Y. and Hollatz, Sabine and Mathews, Irimpan I. and Cichosz, Aleksander and Martirosyan, Vardan and Zeng, Teo and Stofer, Ryan and Liu, Ruobin and others},
title = {Deep residual networks for crystallography trained on synthetic data},
annote = { The use of artificial intelligence to process diffraction images is challenged by the need to assemble large and precisely designed training data sets. To address this, a codebase called Resonet was developed for synthesizing diffraction data and training residual neural networks on these data. Here, two per-pattern capabilities of Resonet are demonstrated: (i) interpretation of crystal resolution and (ii) identification of overlapping lattices. Resonet was tested across a compilation of diffraction images from synchrotron experiments and X-ray free-electron laser experiments. Crucially, these models readily execute on graphics processing units and can thus significantly outperform conventional algorithms. While Resonet is currently utilized to provide real-time feedback for macromolecular crystallography users at the Stanford Synchrotron Radiation Lightsource, its simple Python-based interface makes it easy to embed in other processing frameworks. This work highlights the utility of physics-based simulation for training deep neural networks and lays the groundwork for the development of additional models to enhance diffraction collection and analysis. },
doi = {10.1107/S2059798323010586},
url = {https://www.osti.gov/biblio/2274863},
journal = {Acta Crystallographica. Section D. Structural Biology},
issn = {ISSN 2059-7983},
number = {1},
volume = {80},
place = {United Kingdom},
publisher = {International Union of Crystallography (IUCr)},
year = {2024},
month = {01}}
Lawrence Berkeley National Laboratory (LBNL), Berkeley, CA (United States); SLAC National Accelerator Laboratory (SLAC), Menlo Park, CA (United States)
Sponsoring Organization:
National Institutes of Health (NIH); National Science Foundation (NSF); USDOE; USDOE Laboratory Directed Research and Development (LDRD) Program; USDOE Office of Science (SC), Basic Energy Sciences (BES); USDOE Office of Science (SC), Basic Energy Sciences (BES). Scientific User Facilities (SUF); USDOE Office of Science (SC), Biological and Environmental Research (BER)
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 810https://doi.org/10.1016/j.nima.2015.11.147
Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, Vol. 942https://doi.org/10.1016/j.nima.2019.06.065