- Python2.7
- NumPy
- SciPy
- cvxopt
- IBM ILOG CPLEX Optimization Studio (Community Edition)
Optional for MTZ to CCP4 file conversion
- CCTBX or Phenix
Optional requirements used for installation
- pip
- conda
- git
If you have access to the conda package manager, installing all dependencies
is straightforward
conda install -c conda-forge numpy scipy cvxopt
conda install -c ibmdecisionoptimization cplex
If you prefer the more traditional pip tool, the requirements can be installed
as follows
pip install numpy scipy cvxopt
Next, obtain a copy of CPLEX, the Community Edition will do, by registering and downloading it from the IBM website. After installing and unpacking the software, install the CPLEX Python interface
cd <CPLEX_ROOT>/cplex/python/2.7/x86_64_<PLATFORM>
python setup.py install
where <CPLEX_ROOT> is the directory where you installed CPLEX, and <PLATFORM> is
a platform dependent string, such as linux for Linux systems and osx for
macOSX.
You are now all set now to install qfit_ligand. Installation of qfit_ligand is
as simple as
git clone https://github.com/excitedstates/qfit_ligand
cd qfit_ligand
python setup.py install
The main program that comes with installing qfit_ligand is the eponymously named
qfit_ligand command line tool. It has two calling interfaces
qfit_ligand <CCP4-map> <resolution> <PDB-of-ligand> -r <PDB-of-receptor>
qfit_ligand <CCP4-map> <resolution> <PDB-of-ligand-and-receptor> --selection <chain>,<resi>
where <CCP4-map> is a 2mFo-DFc electron density map in CCP4 format, and
<resolution> is its corresponding resolution in angstrom. In the first
command,<PDB-of-ligand> is a PDB file containing solely the ligand and
<PDB-of-receptor> a PDB file containing the receptor (and other ligands).
In the second command, <PDB-of-ligand-and-receptor is a PDB file containing
both the ligand and receptor, and --selection requires the chain and residue
id of the ligand as a comma separated value, e.g. A,1234. Note that the
receptor (and other ligands) are used to determine the scaling factor of the
density map and used for collision detection during conformer sampling.
To see all options, type
qfit_ligand -h
The main options are -s to give the angular stepsize in degree, and -b to
provide the number of degrees of freedom that are sampled simultaneously.
Reasonably values are -s 1 -b 1, -s 6 -b 2, and -s 24 -b 3. Decreasing
-s and especially increasing -b further will be RAM memory intensive and
will take significantly longer.
Other useful options are -d to store the results in a directory to your
chosing, and if -v to be more verbose about the output, especially when using
it interactively in the shell.
The output of qfit_ligand consists of the following files:
- multiconformer.pdb: Final occupancy weighted ligand multiconformer model.
- conformer_N.pdb: Conformers found before the final rescoring round, where N is an integer.
- qfit_ligand.log: Logging file of run.
To combine the output multiconformer ligand with the receptor use the following command
qfit_combine <multiconformer-ligand-pdbs> -r <receptor> --remove -o <output>
where <multiconformer-ligand-pdbs> is one or more PDB files of the ligand you
want to recombine with the receptor, <receptor> the PDB file of the receptor,
the --remove flag is optionally used the remove the ligand in the current
<receptor> file, and <output> is the file name of the combined PDB
structure.
If you have access to CCTBX/Phenix use phenix.mtz2map to convert a MTZ file
to CCP4. Make sure it outputs the 2mFo-DFc map. Read the documentation for
available options.
The code is licensed under the MIT licence (see LICENSE).
The spacegroups.py module is based on the SpaceGroups.py module of the
pymmlib package, originally licensed under the Artistic License 2.0. See the
licenses directory for a copy of the original source code and its full license.
The elements.py is licensed under MIT, Copyright (c) 2005-2015, Christoph
Gohlke. See file header.