def _createSireSystem(molecules):
"""Create a Sire system from a Molecules object or a list of Molecule
objects.
Parameters
----------
molecules : :class:`Molecules <BioSimSpace._SireWrappers.Molecules>` \
[:class:`Molecule <BioSimSpace._SireWrappers.Molecule>`]
Returns
-------
system : Sire.System.System
A Sire system object.
"""
# Create an empty Sire System.
system = _SireSystem.System("BioSimSpace System")
# Create a new "all" molecule group.
molgrp = _SireMol.MoleculeGroup("all")
# Add the molecules to the group.
if type(molecules) is _Molecules:
molgrp.add(molecules._sire_object)
else:
for mol in molecules:
molgrp.add(mol._sire_object)
# Add the molecule group to the system.
system.add(molgrp)
return system
def _createSireSystem(molecules):
"""Create a Sire system from a list of molecules.
Parameters
----------
molecules : [:class:`Molecule <BioSimSpace._SireWrappers.Molecule>`]
Returns
-------
system : Sire.System.System
A Sire system object.
"""
# Create an empty Sire System.
system = _SireSystem.System("BioSimSpace System")
# Create a new "all" molecule group.
molgrp = _SireMol.MoleculeGroup("all")
# Add the molecules to the group.
for mol in molecules:
molgrp.add(mol._sire_molecule)
# Add the molecule group to the system.
system.add(molgrp)
return system
def molecule(self, index=0, gui=True):
"""View a specific molecule.
Parameters
----------
index : int
The molecule index.
gui : bool
Whether to display the gui.
"""
# Make sure we're running inside a Jupyter notebook.
if not _is_notebook:
return None
# Check that the index is an integer.
if type(index) is not int:
raise TypeError("'index' must be of type 'int'")
# Get the latest system from the process.
if self._is_process:
system = self._handle.getSystem()._getSireObject()
# No system.
if system is None:
return
else:
system = self._handle
# Extract the molecule numbers.
molnums = system.molNums()
# Make sure the index is valid.
if index < 0 or index > len(molnums):
raise ValueError("Molecule index is out of range!")
# Create a new system and add a single molecule.
s = _SireSystem.System("BioSimSpace System")
m = _SireMol.MoleculeGroup("all")
m.add(system[molnums[index]])
s.add(m)
# Create and return the view.
return self._create_view(s, gui=gui)
def run(self, molecule, work_dir=None, queue=None):
"""Run the parameterisation protocol.
Parameters
----------
molecule : BioSimSpace._SireWrappers.Molecule
The molecule to apply the parameterisation protocol to.
work_dir : str
The working directory.
queue : queue.Queue
The thread queue is which this method has been run.
Returns
-------
molecule : BioSimSpace._SireWrappers.Molecule
The parameterised molecule.
"""
if type(molecule) is not _Molecule:
raise TypeError(
"'molecule' must be of type 'BioSimSpace._SireWrappers.Molecule'"
)
if type(work_dir) is not None and type(work_dir) is not str:
raise TypeError("'work_dir' must be of type 'str'")
if type(queue) is not None and type(queue) is not _queue.Queue:
raise TypeError("'queue' must be of type 'queue.Queue'")
# Set work_dir to the current directory.
if work_dir is None:
work_dir = _os.getcwd()
# Create the file prefix.
prefix = work_dir + "/"
# Create a copy of the molecule.
new_mol = molecule.copy()
# Use the net molecular charge passed as an option.
if self._net_charge is not None:
charge = self._net_charge
else:
# The user will likely have passed a bare PDB or Mol2 file.
# Antechamber expects the molecule to be uncharged, or integer
# charged (where the charge, or number of electrons, is passed with
# the -nc flag).
# Get the total charge on the molecule.
if "charge" in self._property_map:
_property_map = {"charge": self._property_map["charge"]}
prop = self._property_map["charge"]
else:
_property_map = {"charge": "charge"}
prop = "charge"
# The molecule has a charge property.
if new_mol._getSireObject().hasProperty(prop):
charge = new_mol.charge(property_map=_property_map).magnitude()
# Charge is non-integer, try to fix it.
if abs(round(charge) - charge) > 0:
new_mol._fixCharge(property_map=_property_map)
charge = round(charge)
else:
charge = None
# Only try "formal_charge" when "charge" is missing. Unlikely to have
# both if this is a bare molecule, but the user could be re-parameterising
# an existing molecule.
if charge is None:
# Get the total formal charge on the molecule.
if "formal_charge" in self._property_map:
_property_map = {
"charge": self._property_map["formal_charge"]
}
prop = self._property_map["charge"]
else:
_property_map = {"charge": "formal_charge"}
prop = "formal_charge"
if new_mol._getSireObject().hasProperty(prop):
charge = new_mol.charge(
property_map=_property_map).magnitude()
# Compute the formal charge ourselves to check that it is consistent.
formal_charge = _formalCharge(molecule).magnitude()
if charge != formal_charge:
_warnings.warn(
"The formal charge on the molecule is %d "
"but we estimate it to be %d" %
(charge, formal_charge))
else:
msg = (
"The molecule has no 'charge' or 'formal_charge' information, and "
"no 'net_charge' option has been passed. You can use the "
"'BioSimSpace.Parameters.formalCharge' function to compute the "
"formal charge")
raise _ParameterisationError(msg)
# Create a new system and molecule group.
s = _SireSystem.System("BioSimSpace System")
m = _SireMol.MoleculeGroup("all")
# Add the molecule.
m.add(new_mol._getSireObject())
s.add(m)
# Write the system to a PDB file.
try:
pdb = _SireIO.PDB2(s)
pdb.writeToFile(prefix + "antechamber.pdb")
except Exception as e:
msg = "Failed to write system to 'PDB' format."
if _isVerbose():
raise IOError(msg) from e
else:
raise IOError(msg) from None
# Generate the Antechamber command.
command = ("%s -at %d -i antechamber.pdb -fi pdb " +
"-o antechamber.mol2 -fo mol2 -c %s -s 2 -nc %d") % (
_protocol._antechamber_exe, self._version,
self._charge_method.lower(), charge)
with open(prefix + "README.txt", "w") as file:
# Write the command to file.
file.write("# Antechamber was run with the following command:\n")
file.write("%s\n" % command)
# Create files for stdout/stderr.
stdout = open(prefix + "antechamber.out", "w")
stderr = open(prefix + "antechamber.err", "w")
# Run Antechamber as a subprocess.
proc = _subprocess.run(command,
cwd=work_dir,
shell=True,
stdout=stdout,
stderr=stderr)
stdout.close()
stderr.close()
# Antechamber doesn't return sensible error codes, so we need to check that
# the expected output was generated.
if _os.path.isfile(prefix + "antechamber.mol2"):
# Run parmchk to check for missing parameters.
command = ("%s -s %d -i antechamber.mol2 -f mol2 " +
"-o antechamber.frcmod") % (_protocol._parmchk_exe,
self._version)
with open(prefix + "README.txt", "a") as file:
# Write the command to file.
file.write("\n# ParmChk was run with the following command:\n")
file.write("%s\n" % command)
# Create files for stdout/stderr.
stdout = open(prefix + "parmchk.out", "w")
stderr = open(prefix + "parmchk.err", "w")
# Run parmchk as a subprocess.
proc = _subprocess.run(command,
cwd=work_dir,
shell=True,
stdout=stdout,
stderr=stderr)
stdout.close()
stderr.close()
# The frcmod file was created.
if _os.path.isfile(prefix + "antechamber.frcmod"):
# Now call tLEaP using the partially parameterised molecule and the frcmod file.
# tLEap will run in the same working directory, using the Mol2 file generated by
# Antechamber.
# Try to find a force field file.
if self._version == 1:
ff = _protocol._find_force_field("gaff")
else:
ff = _protocol._find_force_field("gaff2")
# Write the LEaP input file.
with open(prefix + "leap.txt", "w") as file:
file.write("source %s\n" % ff)
file.write("mol = loadMol2 antechamber.mol2\n")
file.write("loadAmberParams antechamber.frcmod\n")
file.write("saveAmberParm mol leap.top leap.crd\n")
file.write("quit")
# Generate the tLEaP command.
command = "%s -f leap.txt" % _protocol._tleap_exe
with open(prefix + "README.txt", "a") as file:
# Write the command to file.
file.write(
"\n# tLEaP was run with the following command:\n")
file.write("%s\n" % command)
# Create files for stdout/stderr.
stdout = open(prefix + "tleap.out", "w")
stderr = open(prefix + "tleap.err", "w")
# Run tLEaP as a subprocess.
proc = _subprocess.run(command,
cwd=work_dir,
shell=True,
stdout=stdout,
stderr=stderr)
stdout.close()
stderr.close()
# tLEaP doesn't return sensible error codes, so we need to check that
# the expected output was generated.
if _os.path.isfile(prefix +
"leap.top") and _os.path.isfile(prefix +
"leap.crd"):
# Load the parameterised molecule.
try:
par_mol = _Molecule(
_IO.readMolecules([
prefix + "leap.top", prefix + "leap.crd"
])._getSireObject()[_SireMol.MolIdx(0)])
except Exception as e:
msg = "Failed to read molecule from: 'leap.top', 'leap.crd'"
if _isVerbose():
raise IOError(msg) from e
else:
raise IOError(msg) from None
# Make the molecule 'mol' compatible with 'par_mol'. This will create
# a mapping between atom indices in the two molecules and add all of
# the new properties from 'par_mol' to 'mol'.
new_mol._makeCompatibleWith(
par_mol,
property_map=self._property_map,
overwrite=True,
verbose=False)
# Record the forcefield used to parameterise the molecule.
new_mol._forcefield = ff
else:
raise _ParameterisationError("tLEaP failed!")
else:
raise _ParameterisationError("Parmchk failed!")
else:
raise _ParameterisationError("Antechamber failed!")
if queue is not None:
queue.put(new_mol)
return new_mol
def molecules(self, indices=None, gui=True):
"""View specific molecules.
Parameters
----------
indices : [int], range
A list of molecule indices.
gui : bool
Whether to display the gui.
"""
# Make sure we're running inside a Jupyter notebook.
if not _is_notebook:
return None
# Return a view of the entire system.
if indices is None:
return self.system(gui=gui)
# Convert single indices to a list.
if isinstance(indices, range):
indices = list(indices)
elif type(indices) is not list:
indices = [indices]
# Check that the indices is a list of integers.
if not all(isinstance(x, int) for x in indices):
raise TypeError("'indices' must be a 'list' of type 'int'")
# Get the latest system from the process.
if self._is_process:
system = self._handle.getSystem()._getSireObject()
# No system.
if system is None:
return
else:
system = self._handle
# Extract the molecule numbers.
molnums = system.molNums()
# Create a new system.
s = _SireSystem.System("BioSimSpace System")
m = _SireMol.MoleculeGroup("all")
# Loop over all of the indices.
for index in indices:
if index < 0 or index > len(molnums):
raise ValueError("Molecule index is out of range!")
# Add the molecule.
m.add(system[molnums[index]])
# Add all of the molecules to the system.
s.add(m)
# Create and return the view.
return self._create_view(s, gui=gui)
def saveMolecules(filebase, system, fileformat, property_map={}):
"""Save a molecular system to file.
Parameters
----------
filebase : str
The base name of the output file.
system : :class:`System <BioSimSpace._SireWrappers.System>`, \
:class:`Molecule< BioSimSpace._SireWrappers.Molecule>`
[ BioSimSpace._SireWrappers.Molecule ]
The molecular system.
fileformat : str, [str]
The file format (or formats) to save to.
property_map : dict
A dictionary that maps system "properties" to their user
defined values. This allows the user to refer to properties
with their own naming scheme, e.g. { "charge" : "my-charge" }
Returns
-------
files : [str]
The list of files that were generated.
Examples
--------
Load a molecular system from AMBER coordinate and topology files then
try to save it to all supported file formats.
>>> import BioSimSpace as BSS
>>> system = BSS.IO.readMolecules(["ala.rst7", "ala.prm7"])
>>> for format in BSS.IO.fileFormats():
... try:
... BSS.IO.saveMolecules("test", system, format)
... except:
... print("Could not convert to format: '%s'" % format)
Load a molecular system from AMBER coordinate and topology files then
try to save it to GROMACS format, mapping and un-mapping the charge
property along the way.
>>> import BioSimSpace as BSS
>>> system = BSS.IO.readMolecules(["ala.rst7", "ala.prm7"], property_map={"charge" : "my-charge"})
>>> BSS.IO.saveMolecules("test", system, ["gro87", "grotop"], property_map={"charge" : "my-charge"})
"""
global _has_gmx_warned
if _gromacs_path is None and not _has_gmx_warned:
_warn(
"BioSimSpace.IO: Please install GROMACS (http://www.gromacs.org) "
"for GROMACS topology file support.")
_has_gmx_warned = True
# Check that the filebase is a string.
if type(filebase) is not str:
raise TypeError("'filebase' must be of type 'str'")
# Check that that the system is of the correct type.
# A System object.
if type(system) is _System:
pass
# A Molecule object.
elif type(system) is _Molecule:
system = [system]
# A list of Molecule objects.
elif type(system) is list and all(
isinstance(x, _Molecule) for x in system):
pass
# Invalid type.
else:
raise TypeError(
"'system' must be of type 'BioSimSpace.SireWrappers.System', "
"'BioSimSpace._SireWrappers.Molecule, or a list of "
"'BiSimSpace._SireWrappers.Molecule' types.")
# Check that fileformat argument is of the correct type.
# Convert to a list if a single string is passed.
# We split on ',' since the user might pass system.fileFormats() as the argument.
if type(fileformat) is str:
fileformat = fileformat.split(",")
# Lists and tuples are okay!
elif type(fileformat) is list:
pass
elif type(fileformat) is tuple:
pass
# Invalid.
else:
raise TypeError(
"'fileformat' must be a 'str' or a 'list' of 'str' types.")
# Make sure all items in list or tuple are strings.
if not all(isinstance(x, str) for x in fileformat):
raise TypeError(
"'fileformat' must be a 'str' or a 'list' of 'str' types.")
# Make a list of the matched file formats.
formats = []
# Make sure that all of the formats are valid.
for format in fileformat:
try:
f = _formats_dict[format.replace(" ", "").upper()][0]
formats.append(f)
except KeyError:
raise ValueError("Unsupported file format '%s'. Supported formats "
"are: %s." % (format, str(_formats)))
# Validate the map.
if type(property_map) is not dict:
raise TypeError("'property_map' must be of type 'dict'")
# Copy the map.
_property_map = property_map.copy()
# Add the GROMACS topology file path.
if _gromacs_path is not None and ("GROMACS_PATH" not in _property_map):
_property_map["GROMACS_PATH"] = _gromacs_path
# We have a list of molecules. Create a new system and add each molecule.
if type(system) is list:
# Create a Sire system and molecule group.
s = _SireSystem.System("BioSimSpace System")
m = _SireMol.MoleculeGroup("all")
# Add all of the molecules to the group.
for molecule in system:
m.add(molecule._getSireMolecule())
# Add the molecule group to the system.
s.add(m)
# Wrap the system.
system = _System(s)
# Get the directory name.
dirname = _os.path.dirname(filebase)
# If the user has passed a directory, make sure that is exists.
if _os.path.basename(filebase) != filebase:
# Create the directory if it doesn't already exist.
if not _os.path.isdir(dirname):
_os.makedirs(dirname, exist_ok=True)
# Store the current working directory.
dir = _os.getcwd()
# Change to the working directory for the process.
# This avoid problems with relative paths.
if dirname != "":
_os.chdir(dirname)
# A list of the files that have been written.
files = []
# Save the system using each file format.
for format in formats:
# Add the file format to the property map.
_property_map["fileformat"] = _SireBase.wrap(format)
# Write the file.
try:
file = _SireIO.MoleculeParser.save(system._getSireSystem(),
filebase, _property_map)
files += file
except:
if dirname != "":
_os.chdir(dir)
raise IOError("Failed to save system to format: '%s'" %
format) from None
# Change back to the original directory.
if dirname != "":
_os.chdir(dir)
# Return the list of files.
return files
def _run_pdb2gmx(self, molecule, work_dir):
"""Run using pdb2gmx.
Parameters
----------
molecule : BioSimSpace._SireWrappers.Molecule
The molecule to apply the parameterisation protocol to.
work_dir : str
The working directory.
"""
# A list of supported force fields, mapping to their GROMACS ID string.
# GROMACS supports a sub-set of the AMBER force fields.
supported_ff = {
"ff99": "amber99",
"ff99SB": "amber99sb",
"ff03": "amber03"
}
if self._forcefield not in supported_ff:
raise _IncompatibleError(
"'pdb2gmx' does not support the '%s' force field." %
self._forcefield)
# Create a new system and molecule group.
s = _SireSystem.System("BioSimSpace System")
m = _SireMol.MoleculeGroup("all")
# Add the molecule.
m.add(molecule._getSireObject())
s.add(m)
# Create the file prefix.
prefix = work_dir + "/"
# Write the system to a PDB file.
try:
pdb = _SireIO.PDB2(s, self._property_map)
pdb.writeToFile(prefix + "input.pdb")
except Exception as e:
msg = "Failed to write system to 'PDB' format."
if _isVerbose():
raise IOError(msg) from e
else:
raise IOError(msg) from None
# Generate the pdb2gmx command.
command = "%s pdb2gmx -f input.pdb -o output.gro -p output.top -ignh -ff %s -water none" \
% (_gmx_exe, supported_ff[self._forcefield])
with open(prefix + "README.txt", "w") as file:
# Write the command to file.
file.write("# pdb2gmx was run with the following command:\n")
file.write("%s\n" % command)
# Create files for stdout/stderr.
stdout = open(prefix + "pdb2gmx.out", "w")
stderr = open(prefix + "pdb2gmx.err", "w")
# Run pdb2gmx as a subprocess.
proc = _subprocess.run(command,
cwd=work_dir,
shell=True,
stdout=stdout,
stderr=stderr)
stdout.close()
stderr.close()
# Check for the expected output.
if _os.path.isfile(prefix +
"output.gro") and _os.path.isfile(prefix +
"output.top"):
return ["output.gro", "output.top"]
else:
raise _ParameterisationError("pdb2gmx failed!")
def _run_tleap(self, molecule, work_dir):
"""Run using tLEaP.
Parameters
----------
molecule : BioSimSpace._SireWrappers.Molecule
The molecule to apply the parameterisation protocol to.
work_dir : str
The working directory.
"""
# Create a new system and molecule group.
s = _SireSystem.System("BioSimSpace System")
m = _SireMol.MoleculeGroup("all")
# Add the molecule.
m.add(molecule._getSireObject())
s.add(m)
# Create the file prefix.
prefix = work_dir + "/"
# Write the system to a PDB file.
try:
pdb = _SireIO.PDB2(s, self._property_map)
pdb.writeToFile(prefix + "leap.pdb")
except Exception as e:
msg = "Failed to write system to 'PDB' format."
if _isVerbose():
raise IOError(msg) from e
else:
raise IOError(msg) from None
# Try to find a force field file.
ff = _find_force_field(self._forcefield)
# Write the LEaP input file.
with open(prefix + "leap.txt", "w") as file:
file.write("source %s\n" % ff)
file.write("mol = loadPdb leap.pdb\n")
file.write("saveAmberParm mol leap.top leap.crd\n")
file.write("quit")
# Generate the tLEaP command.
command = "%s -f leap.txt" % _tleap_exe
with open(prefix + "README.txt", "w") as file:
# Write the command to file.
file.write("# tLEaP was run with the following command:\n")
file.write("%s\n" % command)
# Create files for stdout/stderr.
stdout = open(prefix + "tleap.out", "w")
stderr = open(prefix + "tleap.err", "w")
# Run tLEaP as a subprocess.
proc = _subprocess.run(command,
cwd=work_dir,
shell=True,
stdout=stdout,
stderr=stderr)
stdout.close()
stderr.close()
# tLEaP doesn't return sensible error codes, so we need to check that
# the expected output was generated.
if _os.path.isfile(prefix +
"leap.top") and _os.path.isfile(prefix +
"leap.crd"):
return ["leap.top", "leap.crd"]
else:
raise _ParameterisationError("tLEaP failed!")
