Evaluation of MRSAD phasing protocols

Molecular Replacement in combination with Single Anomalous Diffraction (MRSAD) is a crystallographic phasing method that can lead to structure solution starting from weak anomalous signal and/or poor MR search models, where both the SAD and MR methods alone would fail. The advent of MRSAD has been triggered by the need to reduce the MR intrinsic model bias, in particular at low resolution, and by the increasing availability of high-resolution structures for components of larger biological complexes. To explore the capabilities and limitations of currently available MRSAD protocols, we have tested these on known crystal structures where we assume that only a part of the crystallographic unit cell is known and can be placed by molecular replacement. Our model system is Cdc23NTerm, a dimeric protein of which the structure has been recently determined via S-SAD at 3.1Å resolution1 . Also, a monomeric, 1.9Å-resolution structure has been obtained in 2013 by Se-SAD2 . MRSAD has been tested on the low-resolution structure using different search models. The pipeline involves MR- and MRSAD-phasing, followed by model building/density modification with three different programs. To assess the results, the deposited PDB model has been used as a reference. The success of the protocol has been judged on the number of residues built into the electron density map, on the phase quality as evaluated via the Mean Phase Error (MPE) and real-space map correlation coefficients against the reference. Numerically, the improvements from MR to MRSAD are limited to a few degrees in terms of MPE. However, the visual inspection of electron density maps and the analysis of the real space correlation coefficients have shown that the electron densities produced with MRSAD phases are clearly superior to those produced from MR or SAD phases alone. As a next step of the study, we will apply the same strategy to larger macromolecular complexes.