MS02-P07 Ensembling for Molecular Replacement: Making the most of your Distant Homologues Ronan Keegan (CCP4, STFC Rutherford Appleton Laboratory, Didcot, United Kingdom) Daniel Rigden (Institute of Integrative Biology, University of Liverpool, Liverpool, United Kingdom) Stuart McNicholas (YSBL, University of York, York, United Kingdom) Eugene Krissinel (CCP4, STFC Rutherford Appleton Laboratory, Didcot, United Kingdom) Jens Thomas (Institute of Integrative Biology, University of Liverpool, Liverpool, United Kingdom) Adam Simpkin (Synchrotron Soleil, Paris, France) Felix Simkovic (Institute of Integrative Biology, University of Liverpool, Liverpool, United Kingdom) Martyn Winn (CCP4, STFC Rutherford Appleton Laboratory, Didcot, United Kingdom) Keith Wilson (YSBL, University of York, York, United Kingdom)email: ronan.keegan@stfc.ac.ukIn the process of structure determination from a macromolecular crystallography diffraction experiment, the most successful approach to solving the well-documented phase problem has been and remains Molecular Replacement (MR). Typically, MR exploits the structural similarity between proteins that are evolutionarily related to derive the necessary phases from a known structure for an unknown target. The majority of proteins in the Protein Data Bank (PDB) have been phased in this way and, as this database expands, the chance of finding a suitable homologue for use in MR increases. Despite its success, the method requires a high degree of structural similarity between the homologue and the target in order for it to act as a suitable proxy for the target’s phases. Given the availability of several structurally-similar homologues, an ensemble of such homologues can help to reduce this sensitivity. In addition, the structural alignment of these models to generate an ensemble can reveal common structural motifs or cores that are likely to be also present in the target. We present here two recent developments in the CCP4 suite designed to generate and exploit ensembles for use in MR. MrBUMP [1], an automated MR pipeline, can retrieve, align and truncate sets of known homologues to produce a set of ensembles. Alternatively, where there may be only one suitable homologue available, AMPLE [2], originally designed to exploit the use of ab initio generated search models in MR, can also now make use of the CONCOORD [3] application to produce a set of ensembles from a single known homologue. In addition, we present a related development in the molecular graphics application, CCP4mg. Through integration with the MrBUMP application, CCP4mg can assist in the generation of ensemble models by enabling the user to visually inspect and interactively edit the structural models.
 
References:

[1] Keegan, R.M. et al. (2018). Acta Cryst. D74, 167-182

[2] Rigden, D.J. et al. (2018). Acta Cryst. D74, 183-193

[3] de Groot, B.L. et al. (1997). Proteins 29: 240-251
Keywords: Molecular Replacement, Structure Solution, Macromolecular Crystallography