Experimental phasing with SHELXC/D/E: combining chain tracing with density modification

Abstract

The programs SHELXC, SHELXD and SHELXE are designed to provide simple, robust and efficient experimental phasing of macromolecules by the SAD, MAD, SIR, SIRAS and RIP methods and are particularly suitable for use in automated structure-solution pipelines. This paper gives a general account of experimental phasing using these programs and describes the extension of iterative density modification in SHELXE by the inclusion of automated protein main-chain tracing. This gives a good indication as to whether the structure has been solved and enables interpretable maps to be obtained from poorer starting phases. The autotracing algorithm starts with the location of possible seven-residue alpha-helices and common tripeptides. After extension of these fragments in both directions, various criteria are used to decide whether to accept or reject the resulting poly-Ala traces. Noncrystallographic symmetry (NCS) is applied to the traced fragments, not to the density. Further features are the use of a 'no-go' map to prevent the traces from passing through heavy atoms or symmetry elements and a splicing technique to combine the best parts of traces (including those generated by NCS) that partly overlap.

Figure 1

Results of the search for (a) seven-residue α-helices and (b) common tripeptides using the density obtained by SHELXE density modification for the 2.75 Å MAD test data for GerE (Ducros et al., 2001; PDB code 1fse). Δ is defined as the average distance to the true atomic site; distances greater than 2.5 Å were replaced by 2.5 Å before calculating the average; f(ρ′) is defined in the text.

Figure 2

Since most main-chain amide N—H groups take part in hydrogen bonds, the density at a point found by extrapolating the N—H vector to 2.9 Å from the N atom provides an indication as to whether the amide has been positioned correctly.

Figure 3

Splicing of two chains that almost coincide for part of the backbone. Firstly, the point is found at which the chains fit best, cutting each chain into two parts (P and R or Q and S). The better of P and Q (according to the figure of merit defined in the text) is spliced onto the better of R and S and the other two partial chains are discarded.

Figure 4

The improvement in model quality for cycles of density modification followed by autotracing for the fibronectin test structure starting from sulfur-SAD phases. The colour indicates the deviation of the C^α atoms from their true positions. Each row represents the protein from the N- to the C-terminus. In the first cycle, 41% was traced with C^α atoms within 1.0 Å, 33% within 0.5 Å and 4% incorrectly traced. After three cycles the figures were 94, 87 and 0%, respectively.

Figure 5

Autotracing quality for the GerE test structure using the same conventions as in Fig. 4 (a) for phasing using only the 2.75 Å MAD data and (b) after phase extension to the 2.15 Å native data.

Figure 6

The C^α trace of one molecule from the MAD phasing of GerE at 2.7 Å with SHELXE (blue) compared with PDB entry 1fse (red). Some terminal residues are missing but otherwise the fit is good. This figure was prepared using PyMOL (DeLano, 2002 ▶).

Figure 7

Autotracing quality after the first cycle for the GerE test structure (PDB code 1fse) using the 2.75 Å MAD data only and the same conventions as in Fig. 4 without and with NCS. Without NCS 55% of the C^α atoms were within 1.0 Å of their true positions and 35% were within 0.5 Å, with 6% wrongly traced. When the sixfold NCS was taken into account, the figures were 74, 49 and 3%, respectively.