Historical models of diffraction have built upon the deep understanding of the scattering physics. This approach works very well in the context of small molecule crystals. However, large biological molecules like proteins and nucleic acids can be very challenging to treat from first principles. A recent breakthrough (Dalton et al., 2022) demonstrated how a deep-learning model could be trained using Bayesian inference to discover an appropriate physical model which is bespoke for each sample. This pioneering work had notable limitations which prevented it from treating the large data sets generated at the LCLS and other free-electron lasers. A new project, ABISMAL, is being developed to surpass these limitations and bring the power of neural networks to LCLS in order to understand the motions of biological molecules.
Graphical representation of the ABISMAL model.
Schematic representation of the ABISMAL implementation.
GitHub Repository
References
2022
Crystallography
A unifying Bayesian framework for merging X-ray diffraction data
Kevin M. Dalton, Jack B. Greisman, and Doeke R. Hekstra
Novel X-ray methods are transforming the study of the functional dynamics of biomolecules. Key to this revolution is detection of often subtle conformational changes from diffraction data. Diffraction data contain patterns of bright spots known as reflections. To compute the electron density of a molecule, the intensity of each reflection must be estimated, and redundant observations reduced to consensus intensities. Systematic effects, however, lead to the measurement of equivalent reflections on different scales, corrupting observation of changes in electron density. Here, we present a modern Bayesian solution to this problem, which uses deep learning and variational inference to simultaneously rescale and merge reflection observations. We successfully apply this method to monochromatic and polychromatic single-crystal diffraction data, as well as serial femtosecond crystallography data. We find that this approach is applicable to the analysis of many types of diffraction experiments, while accurately and sensitively detecting subtle dynamics and anomalous scattering.
