Supplementary data, Table S8. Vh:VBS3 crystallographic refinement statistics against SeMet and native data. The radiation sensitivity of the crystals and the presence of regions affected by varying degrees of disorder made the phasing and model refinement challenging. Disordered regions were difficult to parameterize in terms of multiple conformations at the given resolution which led to relatively high B-factor and R-factor statistics for the single-conformation models used, and limited the scope for reliably assigning bound water molecules. These technical difficulties do not however affect the certainty of our structural results. Bulk solvent corrections were applied during the refinement. A good quality electron density map incorporating purely experimental phases to 4 Å resolution was produced by SAD phasing using data collated from two SeMet substituted crystals, supplemented with isomorphous replacement phasing from partially substituted PtCl42- and Ta6Br122+ derivatives, and followed by density modification. A Ramachandran plot analysis by PROCHECK9 indicates that 88.2% (SeMet) or 84.6% (native) of all residues lie in most favorable regions and the remaining 11.8% (SeMet) or 15.4% (native) lie in additional allowed regions. This structure analysis also showed that all stereochemical parameters are better than expected at the given resolution.

SeMet / Native
Resolution range (Å) / 15 - 3.92 / 15 - 2.7
No. of Reflections (Working set) / 2,376 / 8,618
No. of Reflections (Test set) / 536 / 433
R-factorb / 0.300 / 0.244
last shell / 0.275 (4.03 Å - 3.92 Å) / 0.248 (2.78 Å - 2.7 Å)
R / 0.319 / 0.335
last shell / 0.307 / 0.298
No. of amino acid residues / 279 / 284
No. of protein atoms / 2,172 / 2,216
Average B-factor (Protein) [Å2] / 126.6 / 111.9
R.m.s.d. from ideal geometry
Covalent bond lengths (Å) / 0.010 / 0.010
Bond angles (º) / 1.487 / 1.263

bR-factor =

where <½Fcalc½> is the expectation of ½Fcalc½ under the error model used in maximum-likelihood refinement.

cThe free R-factor is a cross-validation residual calculated by using 5% of the native data (or 22% of the SeMet to ensure to have enough reflections), which were randomly chosen and excluded from the refinement.

2