Deep residual networks for crystallography trained on synthetic data

Mendez, D.; Holton, J.M.; Lyubimov, A.Y.; Hollatz, S.; Mathews, I.I.; Cichosz, A.; Martirosyan, V.; Zeng, T.; Stofer, R.; Liu, R.; Song, J.; McPhillips, S.; Soltis, M.; Cohen, A.E.

doi:10.1107/S2059798323010586

Acta Crystallographica Section D
Acta Crystallographica
Section D STRUCTURAL BIOLOGY

IUCr IT WDC

home archive editors for authors for readers submit subscribe open access

view article

Figure 8
Resonet resolution prediction for rotational data sets recorded at the SSRL Structural and Molecular Biology (SMB) beamlines. The horizontal axis is sorted according to collection time. The red and blue markers are the minimum and average inferred resolution per shot (across the four quadrants), respectively, and the green markers are resolution estimates from DIALS. Each data set is labeled by a resolution determined using AIMLESS (shown at the top of each subplot and indicated by the dashed line) as the point where the overall signal-to-noise ratio of the integrated intensity dipped below 1.5. The yellow-shaded regions labeled I and II in subplots A, S and W correspond to regions where maximum composite images were computed to identify the cause of systematic variation in the Resonet-predicted resolution (see Fig. 9

for the corresponding maximum composite images).

STRUCTURAL
BIOLOGY

ISSN: 2059-7983

Volume 80| Part 1| January 2024| Pages 26-43

https://doi.org/10.1107/S2059798323010586

Open

access

Follow Acta Cryst. D

Search IUCr Journals		doi		Advanced search
Author		volume	page