Tables for
Volume F
Crystallography of biological macromolecules
Edited by E. Arnold, D. M. Himmel and M. G. Rossmann

International Tables for Crystallography (2012). Vol. F, ch. 16.1, pp. 425-426   | 1 | 2 |

Section 16.1.11. Substructure solution for native sulfurs and halide soaks

G. M. Sheldrick,a C. J. Gilmore,b H. A. Hauptman,c C. M. Weeks,c* R. Millerc and I. Usónd

aLehrstuhl für Strukturchemie, Georg-August-Universität Göttingen, Tammannstrasse 4, D-37077 Göttingen, Germany,bDepartment of Chemistry, University of Glasgow, Glasgow G12 8QQ, UK,cHauptman–Woodward Medical Research Institute, Inc., 700 Ellicott Street, Buffalo, NY 14203–1102, USA, and dInstitució Catalana de Recerca i Estudis Avançats at IBMB-CSIC, Barcelona Science Park. Baldiri Reixach 15, 08028 Barcelona, Spain
Correspondence e-mail:

16.1.11. Substructure solution for native sulfurs and halide soaks

| top | pdf |

In the past, experimental phasing usually involved either the preparation of selenomethionine derivatives or the incorporation of heavy-metal ions by soaking crystals with a low concentration of the metal salt for several hours. The first of these methods required time in the wet lab and did not work well for all expression systems; the second had a low success rate. The improved quality of modern diffraction data collected from cryo-cooled crystals makes it now possible to exploit the weak anomalous signal from the native sulfur atoms or from halide ions introduced by soaking with a high concentration of a halide (iodide or bromide) for a few seconds immediately before cryocooling the crystal (Dauter et al., 2000[link], 2001[link]; Usón et al., 2003[link]). The success of these approaches is also made possible by the ability of modern, dual-space, substructure-solution programs to locate correctly a large number of sites, possibly with varying occupancies, using the SAD and SIRAS approaches.

In selenomethionine SAD and MAD phasing and in sulfur SAD phasing, the variation of the occupancies (refined in the final two cycles in the case of SHELXD) provides a very good indication as to whether the structure has been solved. Fig.[link](a) shows the phasing of elastase with sulfur SAD; a sharp drop in the relative occupancy after the 12th site confirms the expected presence of 12 sulfur atoms. For an iodide soak of the same protein (Fig.[link]b), the relative occupancies show a gradual fall with peak number. Since the number of sites is difficult to estimate in advance for a halide soak and SHELXD needs to know this number approximately (within say 20%), it may be necessary to make several trials with different numbers of expected sites. From experience, the best number to use is the one that causes the occupancies to fall to about 0.2 relative to the strongest peak. Usually, subsequent refinements of the occupancies show that all the sites are partially occupied for halide soaks.


Figure | top | pdf |

Relative occupancy against peak number for SHELXD substructure solutions of elastase. (a) Sulfur-SAD experiment showing the presence of the 12 expected sulfur atoms. (b) Iodide soak. Subsequent analysis showed that the peaks with relative occupancies less than 0.2 are mainly noise. These figures were made with HKL2MAP (Pape & Schneider, 2004[link]).

When the anomalous signal does not extend beyond about 2.0 Å, the two sulfur atoms of a disulfide bridge coalesce to a single maximum, often referred to as a supersulfur atom. At low resolution, this increases the signal-to-noise ratio for such sites in the dual-space procedure, but tends to impede phase extension to higher resolution (e.g. when density modification is applied to the native data with the starting phases estimated using these super­sulfur atoms). An efficient way around this problem is to fit dumbbells rather than single atoms in the peak-search part of the dual-space recycling (Debreczeni et al., 2003[link]); this dramatically improves the quality of the higher-resolution starting phases.

Because the weak anomalous signal is swamped by the noise at higher resolution in such SAD experiments, it is often essential to truncate the resolution of the anomalous difference data before searching for the substructure. For MAD experiments, it is customary to truncate the data to the resolution at which the correlation coefficient between the signed anomalous differences falls below 30% (Schneider & Sheldrick, 2002[link]). The same criterion can be used for SAD experiments if two independent data sets (e.g. from two different crystals) are available. As a compromise, the signed anomalous differences can be divided randomly into two sets, and then the correlation coefficient between them can be calculated. However, since these sets are not completely independent, a higher threshold (say 40%) might be advisable. An alternative criterion is to truncate the data at the point where the ratio of the mean absolute anomalous difference to its mean standard deviation falls below ~1.3, but this requires rather precise estimates of the standard deviations. In borderline cases, especially when multiple CPUs are available, it is probably safer simply to run the substructure solution for a range of different resolution cutoffs in parallel, and this is already implemented in several of the automated phasing pipelines. Sometimes good solutions are only obtained in a rather limited resolution cutoff range. A good starting value for sulfur SAD is the diffraction limit plus 0.5 Å.


Dauter, Z., Dauter, M. & Rajashankar, K. R. (2000). Novel approach to phasing proteins: derivatization by short cryo-soaking with halides. Acta Cryst. D56, 232–237.
Dauter, Z., Li, M. & Wlodawer, A. (2001). Practical experience with the use of halides for phasing macromolecular structures: a powerful tool for structural genomics. Acta Cryst. D57, 239–249.
Debreczeni, J. É., Girmann, B., Zeeck, A., Krätzner, R. & Sheldrick, G. M. (2003). Structure of viscotoxin A3: disulfide location from weak SAD data. Acta Cryst. D59, 2125–2132.
Schneider, T. R. & Sheldrick, G. M. (2002). Substructure solution with SHELXD. Acta Cryst. D58, 1772–1779.
Usón, I., Schmidt, B., von Bülow, R., Grimme, S., von Figura, K., Dauter, M., Rajashankar, K. R., Dauter, Z. & Sheldrick, G. M. (2003). Locating the anomalous scatterer substructures in halide and sulfur phasing. Acta Cryst. D59, 57–66.

to end of page
to top of page