On artificial crystal structure generation for solving the phase problem with deep learning

Melgalvis, D.M.; Rekis, T.

doi:10.1107/S2053273325009428

Acta Crystallographica Section A
Acta Crystallographica
Section A FOUNDATIONS AND ADVANCES

IUCr IT WDC

home archive editors for authors for readers submit subscribe open access

view article

Figure 7
Density map projections (0.5 Å slabs in e₃ summed in 0.05 Å steps) of selected large unit-cell structures phased with the PhAI model trained on $[V\sim{\rm Lognorm}]$ artificial molecule data. (a) CSD ARUZIY, COD 7225798 (Hall et al., 2016

), V = 1264 Å³. (b) CSD PIDDEP (Michalsky et al., 2022

), V = 2201 Å³. (c) CSD NUQHAK, COD 7222071 (Mir et al., 2015

), V = 3203 Å³. (d) CSD DYEETS (Kaneda et al., 1977

), V = 3068 Å³. True atomic positions are marked with crosses. Correlation coefficients r between the phased maps shown and true density maps are indicated.

FOUNDATIONS
ADVANCES

ISSN: 2053-2733

Volume 82| Part 1| January 2026| Pages 32-40

https://doi.org/10.1107/S2053273325009428

Open

access

Follow Acta Cryst. A

Search IUCr Journals		doi		Advanced search
Author		volume	page