This exhaustive search algorithm explores the crystallographic occupancy & B factor of ligands. The algorithm is designed to be used as a last stage step to confirm crystallographic occupancy, when the position of the ligand has been refined.
We consider the crystal with a ligand bound to be in a superposition of bound and ground states, specifically we use the framework detailed in Pearce, N. M., Krojer, T. & von Delft, F. 2017. Acta Cryst. D73, 256-266.
A superposed model contians a bound state and a ground state. The bound state is usually the bound ligand and any alternative conformations of the protein induced when the ligand fits. The ground state is the conformation(s) modelled for the protien without any ligand bound. Occupancy of the states are varied such that they sum to unity:i.e. Bound state occupancy 0.1, Ground state occupancy 0.9.
At each value of occupancy the code explores possible values of B factor. Currently the B factor is fixed to be the same across the whole state. The output is thus recorded to a CSV file, which can be used to generate a pdb file at the minima of B factor and occupancy, as well as be plotted.
The project heavily relies on the computational crystallographic toolbox, as distributed in CCP4 as well as tools from PanDDA.
Install a recent version of CCP4.
Please install an up to date version of panddas
Download/clone this code from github. Unzip the download directory and cd into it:
cd "/path/to/download/directory"
Then install using
ccp4-python setup.py install
Scripts accessible are:
exhaustive
This can be run minimally by supplying a pdb and mtz file
exhaustive test/resources/FALZA-x0085/FALZA-x0085.free.mtz test/resources/FALZA-x0085/refine.pdb
Or with additional parameters:
exhaustive test/resources/FALZA-x0085/FALZA-x0085.free.mtz test/resources/FALZA-x0085/refine.pdb exhaustive.options.step=0.2'''
Options for this script are detailed in exhaustive.utils.phil, and can be supplied on command line as shown for the step size in the above example. The input is minimally a pdb file refined using giant.quick_refine, and the corresponding mtz file.
exhaustive_multiple_sampling
As above but runs at step sizes 0.05 across occupancy [0,1], u_iso [0.2, 1.2] and then at the minima +/- 0.1 in step sizes of 0.01. This will increase the program runtime by 10 times, and is appropraite as in tested cases the minima of the difference maps are flat