def __init__(self, engine="GROMACS", box_length_complex=9, box_length_morph=4, ion_conc=0.154, shell=0,
neutralise=True, centre=True, protein_ff=_PC.AMBERDEFAULTPROTEINFF, ligand_ff=_PC.AMBERDEFAULTLIGANDFF,
water_ff=_PC.AMBERDEFAULTWATERFF, protein=None, ligand_ref=None, morphs=None, workdir=None):
# work directory
self.workdir = _fileio.Dir(workdir) if workdir else _fileio.Dir(".")
# writing options
self.engine = engine
# solvation options
self.box_length_complex = box_length_complex
self.box_length_morph = box_length_morph
self.ion_conc = ion_conc
self.shell = shell
self.neutralise = neutralise
self.centre = centre
# parametrisation options
self.params = _parametrise.Params(protein_ff=protein_ff, ligand_ff=ligand_ff, water_ff=water_ff)
# molecular systems
if not isinstance(protein, Protein):
self.protein = Protein(protein, ligand_ref=ligand_ref)
else:
self.protein = protein
self.morphs = morphs
self._complex_template = None
self.systems_prep = {}
def __init__(self,
code=None,
pdb_file=None,
ligands=None,
ligand_ref=None,
fasta_file=None,
complex_template=None,
name=None,
workdir=None):
self.name = name if name is not None else code
self.workdir = _fileio.Dir(workdir) if workdir else _fileio.Dir(
self.name)
with self.workdir:
self.code = code
self.complex_template = complex_template
if pdb_file:
self.pdb = pdb_file
else:
self.pdb = complex_template
self.fasta = fasta_file
self.ligands = ligands
self.cofactors = []
# remove ligands from PDB file and non-ligands from SDF files
self.filter(missing_residues="all",
chains="all",
waters="all",
simple_anions="all",
complex_anions="all",
simple_cations="all",
complex_cations="all",
ligands="all",
cofactors="all")
self.ligand_ref = ligand_ref
def __init__(self, name, gro_files, top_files, work_dir=None, **lambda_dict):
try:
len_arr, = {len(arr) for arr in lambda_dict.values() if len(arr)}
except ValueError:
raise ValueError("Need lists of the same non-zero size")
if not isinstance(gro_files, list):
gro_files = [gro_files] * len_arr
else:
assert len(gro_files) == len_arr, "Number of GRO files does not match number of lambda values"
if not isinstance(top_files, list):
top_files = [top_files] * len_arr
else:
assert len(top_files) == len_arr, "Number of TOP files does not match number of lambda values"
if work_dir is None:
work_dir = _os.getcwd()
self.files = [{} for _ in range(len_arr)]
for f, gro_file, top_file in zip(self.files, gro_files, top_files):
f["gro"] = _os.path.abspath(gro_file)
f["top"] = _os.path.abspath(top_file)
self._workdir = _fileio.Dir("%s/%s" % (work_dir, name))
self.lambda_dict = lambda_dict
self.protocols = []
self.mbar_data = []
def saveAsGromacs(filebase, system):
"""
Saves the input object to a GRO and a TOP file.
Parameters
----------
filebase : str
Base of the output filenames.
system : BioSimSpace.System or parmed.structure.Structure
Input object to be saved.
Returns
-------
files : [str, str]
A list containing the absolute paths of the TOP and GRO files.
"""
name = _os.path.basename(_tempfile.TemporaryDirectory().name)
with _fileio.Dir(dirname=name, temp=True) as dir:
if "BioSimSpace" in str(type(system)):
_BSS.IO.saveMolecules(filebase, system, "Gro87,GroTop")
_pmd.load_file(filebase + ".gro87").save(filebase + ".gro",
combine="all")
_shutil.move(filebase + ".gro",
_os.path.join(dir.workdirname, filebase + ".gro"))
_shutil.move(filebase + ".grotop",
_os.path.join(dir.workdirname, filebase + ".top"))
elif isinstance(system, (_pmd.Structure, _PC.Morph.Morph)):
system.save(filebase + ".gro")
system.save(filebase + ".top")
_shutil.move(filebase + ".gro",
_os.path.join(dir.workdirname, filebase + ".gro"))
_shutil.move(filebase + ".top",
_os.path.join(dir.workdirname, filebase + ".top"))
else:
raise TypeError(
"Passed object to save as GROMACS not recognised. Please pass BioSimSpace or ParmEd "
"objects only.")
return [
_os.path.abspath(filebase + ".top"),
_os.path.abspath(filebase + ".gro")
]
def saveSystems(self, systems=None):
"""
Saves all systems.
Parameters
----------
systems : dict or None
Input systems to be saved. None means use self.systems_prep.
"""
if systems is None: systems = self.systems_prep
# TODO support other engines
_logging.info("Saving solvated complexes as GROMACS...")
with self.workdir:
for name, (morph, complexes) in systems.items():
with _fileio.Dir(name):
_IO.GROMACS.saveAsGromacs("morph", morph)
if len(complexes) == 1:
_IO.GROMACS.saveAsGromacs("complex_final", complexes[0])
else:
for j, complex in enumerate(complexes):
_IO.GROMACS.saveAsGromacs("complex_final%d" % j, complex)
def __init__(self,
input,
parametrised_files=None,
name=None,
protonated=False,
minimise=None,
workdir="."):
self.name = name
self.workdir = _fileio.Dir(workdir)
self.minimise = minimise
# always set protonated to False and if a valid protonated file is given it is automatically set to True
self.protonated_filename = None
with self.workdir:
if parametrised_files:
self.molecule = _rdkit.openAsRdkit(parametrised_files,
removeHs=False,
minimise=False,
template=input)
self.protonated_filename = _rdkit.saveFromRdkit(
self.molecule, "{}.pdb".format(name))
elif isinstance(input, str):
if _os.path.exists(input):
if protonated:
self.protonated_filename = input
else:
self.molecule = _rdkit.openAsRdkit(input,
minimise=minimise)
else:
self.string = input
elif isinstance(input, _rdchem.Mol):
self.molecule = input
else:
raise TypeError(
"Need a SMILES, InChI string, filename or an RDKit object as an input"
)
self.parametrised_files = parametrised_files
self.minimise = False
def generateMBARData(self, n_cores=None, n_nodes=1, cont=True):
"""
Manually generates input data for MBAR using gmx mdrun -rerun using N
topologies and N trajectories generated from N simulations (N squared
energy evaluations are needed). Doesn't work with constrained
simulations. For most cases one should instead use the XVG files
readily generated by GROMACS during the simulation.
Parameters
----------
n_cores : int or None, optional
Number of cores used. Default: the same as the number of replicas.
n_nodes : int, optional
Number of nodes used.
cont : bool, optional
Whether to overwrite existing files or to continue from the last
one.
Returns
-------
mbar_data : numpy.ndarray
Data which is to be passed to pymbar.
"""
# GROMACS might generate a bunch of warnings when we apply a non-dummy Hamiltonian to a trajectory with dummies
# due to clashes and backup too many structures. Here we suppress this backing up
gmx_suppress_dump = _os.environ[
"GMX_SUPPRESS_DUMP"] if "GMX_SUPPRESS_DUMP" in _os.environ.keys(
) else None
_os.environ["GMX_SUPPRESS_DUMP"] = "1"
self.mbar_data = []
_logging.info("Generating Energy files...")
with self._workdir:
with _fileio.Dir("MBAR"):
for i, file in enumerate(self.files):
self.mbar_data += [[]]
gro, top = file["gro"], file["top"]
for j, file in enumerate(self.files):
run = True
trr = file["trr"]
filebase = "Energy_%d_%d" % (i, j)
# we only overwrite the last generated energies, because they might be incomplete
if cont:
filebase_next = "Energy_%d_%d" % (
i + (j + 1) // len(self.files),
(j + 1) % len(self.files))
files_current = _glob.glob(
"%s/%s.xvg" % (_os.getcwd(), filebase))
files_next = _glob.glob(
"%s/%s.xvg" % (_os.getcwd(), filebase_next))
if len(files_current) and len(files_next):
run = False
if run:
# use the last protocol or create a default one
if len(self.protocols):
protocol = self.protocols[-1]
else:
protocol = _Protocol.Protocol(
use_preset="default", **self.lambda_dict)
protocol.init_lambda_state = i
protocol.skip_positions = 0
protocol.skip_velocities = 0
protocol.skip_forces = 0
protocol.write_derivatives = False
protocol.random_velocities = False
protocol.constraint = "no"
protocol.__setattr__("continuation", "yes")
protocol.__setattr__("dhdl-print-energy",
"potential")
protocol.__setattr__("calc-lambda-neighbors", "0")
protocol.__setattr__("calc-lambda-neighbors", "0")
mdp = protocol.write("GROMACS", filebase=filebase)
tpr = filebase + ".tpr"
grompp_command = "%s grompp -maxwarn 10 -f '%s' -c '%s' -p '%s' -o '%s'" % (
_PC.GROMACSEXE, mdp, gro, top, tpr)
_runexternal.runExternal(grompp_command,
procname="gmx grompp")
if n_cores is None:
mdrun_command = "%s mdrun -s '%s' -rerun '%s' -deffnm '%s'" % (
_PC.GROMACSEXE, tpr, trr, filebase)
else:
ppn = n_cores // n_nodes
mdrun_command = "{0} -np {1} --map-by ppr:{2}:node {3} mdrun -s '{4}' -rerun '{5}' " \
"-deffnm {6}".format(_PC.MPIEXE, n_cores, ppn, _PC.GROMACSMPIEXE, tpr,
trr, filebase)
_runexternal.runExternal(mdrun_command,
procname="gmx mdrun")
self.mbar_data[i] += list(
_MDAnalysis.auxiliary.XVG.XVGReader(
filebase + ".xvg")._auxdata_values[:, 1])
# restore the original environment variable
del _os.environ["GMX_SUPPRESS_DUMP"]
if gmx_suppress_dump is not None:
_os.environ["GMX_SUPPRESS_DUMP"] = gmx_suppress_dump
return self.mbar_data
def runSimulation(self,
name,
multi=False,
multidir=False,
single_lambda=None,
use_mpi=False,
mdrun_mpi=False,
gpu_id=None,
use_preset=None,
replex=None,
plumed_file=None,
n_cores_per_process=None,
n_nodes=1,
n_processes=None,
dlb=False,
gmx_kwargs=None,
**protocol_params):
"""
Runs a simulation in GROMACS.
Parameters
----------
name : str
The name of the simulation.
multi : bool, optional
Whether to run the simulations in parallel, using -multi.
multidir : bool, optional
Whether to run the simulations in parallel, using -multidir. Overrides multi.
single_lambda : None, int, optional
An integer runs a simulation at a single lambda value and overrides multi and replex.
None runs all lambda values.
use_mpi : bool, optional
Whether to use ProtoCaller.GROMACSEXE or GROMACSMPIEXE.
mdrun_mpi : bool, optional
Whether to call mdrun as "mdrun" or as "mdrun_mpi".
gpu_id : str, optional
Which GPU id to be used, if applicable.
use_preset : str, Protocaller.Protocol.Protocol, None
Which default preset to use. One of: "minimisation", "equilibration_nvt", "equilibration_npt", "production"
"vacuum". You can alternatively pass a custom protocol here, in which case **protocol_params will not be
used.
replex : int or None, optional
Attempts replica exchange after replex number of steps using PLUMED. None means no replica exchange.
Overrides use_mpi and multi if not None.
plumed_file : str, optional
If replex is True, it uses this file as a configuration for PLUMED, otherwise an empty file is used.
n_cores_per_process : int or None, optional
Number of cores used per process. Default: let GROMACS decide.
n_nodes : int, optional
Number of physical nodes used.
n_processes : int, optional
Number of processes. Default: the same as the number of simulation.
dlb : bool
Whether to enable dynamic load balancing. Default is False due to some possible instabilities with
gmx_mpi mdrun in some cases.
gmx_kwargs : dict
Additional arguments to be passed to mdrun. The keys of the dictionary need to be the name of the option,
e.g. "cpt" for checkpoint interval, while the values need to be the value of the option if it permits one
or None if it doesn't. If the values contain "{}" while the user is running the lambda windows in serial,
this will be replaced by the lambda number.
protocol_params
Keyword arguments passed to ProtoCaller.Protocol.Protocol.
"""
# perform some checks and initialise some default values
if n_processes is None:
n_processes = 1 if single_lambda else self.lambda_size
if n_nodes > 1 or multi or multidir:
use_mpi = True
if single_lambda is not None:
replex, multi, multidir = None, False, False
if replex is not None:
use_mpi, multi = True, True
if mdrun_mpi:
MDRUNEXE = _os.path.dirname(_PC.GROMACSMPIEXE) + "/mdrun_mpi"
elif use_mpi:
MDRUNEXE = _PC.GROMACSMPIEXE + " mdrun"
else:
MDRUNEXE = _PC.GROMACSEXE + " mdrun"
if use_mpi or mdrun_mpi:
base_mdrun_command = f"{_PC.MPIEXE} -np {n_processes} --map-by ppr:{n_processes // n_nodes}:node {MDRUNEXE}"
else:
base_mdrun_command = MDRUNEXE
gmx_kwargs = {} if gmx_kwargs is None else gmx_kwargs
gmx_kwargs["dlb"] = "yes" if dlb else "no"
if n_cores_per_process is not None:
gmx_kwargs["ntomp"] = n_cores_per_process
if gpu_id is not None:
gmx_kwargs["gpu_id"] = gpu_id
# this is due to incompatibility between GROMACS 2019+ and multi
multidir = True if self._gmx_version(
_PC.GROMACSMPIEXE) >= 2019 and multi else multidir
# initialise the protocol
if isinstance(use_preset, _Protocol.Protocol):
protocol = use_preset
else:
protocol = _Protocol.Protocol(use_preset=use_preset,
**protocol_params,
**self.lambda_dict)
self.protocols += [protocol]
# run the simulations
_logging.info(f"Running {name}...")
with self._workdir:
with _fileio.Dir(name, overwrite=False):
# run single lambda if needed
if single_lambda is not None:
it = [single_lambda]
else:
it = range(self.lambda_size)
# call grompp for each lambda
for i in it:
filebase = name if multidir else f"{name}_{i}"
dirname = f"Lambda_{i}" if multidir else "."
with _fileio.Dir(dirname):
protocol.init_lambda_state = i
self.files[i]["mdp"] = _os.path.abspath(
protocol.write(engine="GROMACS",
filebase=filebase))
grompp_args = {
"-f": "mdp",
"-c": "gro",
"-p": "top",
"-t": "cpt",
}
grompp_command = f"{_PC.GROMACSEXE} grompp -maxwarn 10 -o {filebase}.tpr"
for grompp_arg, filetype in grompp_args.items():
if filetype in self.files[i].keys():
grompp_command += f" {grompp_arg} '{self.files[i][filetype]}'"
_runexternal.runExternal(grompp_command,
procname="gmx grompp")
# call the simulations consecutively for each lambda if multi is not specified
if not (multi or multidir):
for i in it:
filebase = f"{name}_{i}"
gmx_kwargs["s"] = f"'{filebase}.tpr'"
gmx_kwargs["deffnm"] = f"'{filebase}'"
# run GROMACS
mdrun_command = self._dict_to_arguments(
base_mdrun_command, gmx_kwargs, i)
_runexternal.runExternal(mdrun_command,
procname="gmx mdrun")
# update files after the run
self._update_files(i, filebase)
# alternatively, run all simulations in parallel
else:
filebase = name if multidir else name + "_"
gmx_kwargs["s"] = f"'{filebase}.tpr'"
gmx_kwargs["deffnm"] = f"'{filebase}'"
if multidir:
gmx_kwargs["multidir"] = " ".join(f"Lambda_{i}"
for i in it)
else:
gmx_kwargs["multi"] = f"{self.lambda_size}"
# run replica exchange with PLUMED if needed
if replex is not None:
if plumed_file is None:
plumed_file = "plumed.dat"
open(plumed_file, "a").close()
gmx_kwargs["plumed"] = f"'{plumed_file}'"
gmx_kwargs["replex"] = replex
gmx_kwargs["hrex"] = None
# run GROMACS
mdrun_command = self._dict_to_arguments(
base_mdrun_command, gmx_kwargs, i)
_runexternal.runExternal(mdrun_command,
procname="gmx mdrun")
# update files after the run
for i in it:
filebase = name if multidir else f"{name}_{i}"
dirname = f"Lambda_{i}" if multidir else "."
with _fileio.Dir(dirname):
self._update_files(i, filebase)
def runSimulation(self, name, multi=False, single_lambda=None, use_mpi=False, use_preset=None, replex=None,
n_cores_per_process=None, n_nodes=1, n_processes=None, dlb=False, **protocol_params):
"""
Runs a simulation in GROMACS.
Parameters
----------
name : str
The name of the simulation.
multi : bool, optional
Whether to run the simulations in parallel, using -multi.
single_lambda : None, int, optional
An integer runs a simulation at a single lambda value and overrides multi and replex.
None runs all lambda values.
use_mpi : bool, optional
Whether to use ProtoCaller.GROMACSEXE or GROMACSMPIEXE.
use_preset : str, None
Which default preset to use. One of: "minimisation", "equilibration_nvt", "equilibration_npt", "production"
"vacuum".
replex : int or None, optional
Attempts replica exchange after replex number of steps using PLUMED. None means no replica exchange.
Overrides use_mpi and multi if not None.
n_cores_per_process : int or None, optional
Number of cores used per process. Default: let GROMACS decide.
n_nodes : int, optional
Number of physical nodes used.
n_processes : int, optional
Number of processes. Default: the same as the number of simulation.
dlb : bool
Whether to enable dynamic load balancing. Default is False due to some possible instabilities with
gmx_mpi mdrun in some cases.
protocol_params
Keyword arguments passed to ProtoCaller.Protocol.Protocol.
"""
if n_processes is None:
n_processes = 1 if single_lambda else self.lambda_size
ppn = n_processes // n_nodes
if n_nodes > 1:
use_mpi = True
if single_lambda is not None:
replex = None
multi = False
if replex is not None:
multi = True
use_mpi = True
dlb = "yes" if dlb else "no"
protocol = _Protocol.Protocol(use_preset=use_preset, **protocol_params, **self.lambda_dict)
self.protocols += [protocol]
_logging.info("Running %s..." % name)
with self._workdir:
with _fileio.Dir(name, overwrite=True):
# run single lambda if needed
if single_lambda is not None:
it = [single_lambda]
else:
it = range(self.lambda_size)
# call grompp for every lambda
for i in it:
filebase = "%s_%d" % (name, i)
protocol.init_lambda_state = i
self.files[i]["mdp"] = _os.path.abspath(protocol.write(engine="GROMACS", filebase=filebase))
grompp_args = {
"-f": "mdp",
"-c": "gro",
"-p": "top",
"-t": "cpt",
}
grompp_command = "%s grompp -maxwarn 10 -o %s.tpr" % (_PC.GROMACSEXE, filebase)
for grompp_arg, filetype in grompp_args.items():
if filetype in self.files[i].keys():
grompp_command += " %s '%s'" % (grompp_arg, self.files[i][filetype])
_runexternal.runExternal(grompp_command, procname="gmx grompp")
# call the simulation consecutively for every lambda if multi is not specified
if not multi:
if not use_mpi:
mdrun_command = "{0} mdrun -s '{1}.tpr' -deffnm {1} -dlb {2}".format(
_PC.GROMACSEXE, filebase, dlb)
else:
mdrun_command = "{0} -np {1} --map-by ppr:{2}:node {3} " \
"mdrun -s '{4}.tpr' -deffnm {4} -dlb {5}".format(
_PC.MPIEXE, n_processes, ppn, _PC.GROMACSMPIEXE, filebase, dlb)
if n_cores_per_process:
mdrun_command += " -ntomp {}".format(n_cores_per_process)
_runexternal.runExternal(mdrun_command, procname="gmx mdrun")
# update files after the run
self.files[i] = {"top" : self.files[i]["top"],}
output_files = _glob.glob("%s.*" % filebase)
for output_file in output_files:
ext = output_file.split(".")[-1].lower()
if ext != "tpr":
self.files[i][ext] = _os.path.abspath(output_file)
# alternatively, run all simulations in parallel
if multi:
filebase = name + "_"
mdrun_command = "{0} -np {1} --map-by ppr:{2}:node {3} mdrun -multi {4} -s {5}.tpr " \
"-deffnm {5} -dlb {6}".format(
_PC.MPIEXE, n_processes, ppn, _PC.GROMACSMPIEXE, self.lambda_size, filebase, dlb)
# run replica exchange with PLUMED if needed
if replex is not None:
open("plumed.dat", "a").close()
mdrun_command += " -plumed plumed.dat -replex {} -hrex".format(replex)
if n_cores_per_process:
mdrun_command += " -ntomp {}".format(n_cores_per_process)
_runexternal.runExternal(mdrun_command, procname="gmx mdrun")
# update files after the run
for i in range(self.lambda_size):
filebase = "%s_%d" % (name, i)
self.files[i] = {"top": self.files[i]["top"], }
output_files = _glob.glob("%s.*" % filebase)
for output_file in output_files:
ext = output_file.split(".")[-1].lower()
if ext != "tpr":
self.files[i][ext] = _os.path.abspath(output_file)
def prepareComplexes(self,
replica_temps=None,
scale_dummy_bonds=1,
dummy_bond_smarts="[*]~[*]",
intermediate_files=False,
store_complexes=False,
output_files=True):
"""
Batch prepares all complexes with an option to output files for REST(2).
Parameters
----------
replica_temps : [float] or None
A list of replica temperatures. Everything is normalised with respect to the lowest temperature. None
means only output the normal files.
scale_dummy_bonds : float
Sets the dummy bond length distance as a fraction of the real bond length distance.
dummy_bond_smarts : str
SMARTS string which indicates which dummy bonds are to be affected by scale_dummy_bonds.
intermediate_files : bool
Whether to store all intermediate files.
store_complexes : bool
Whether to store the final complexes as a dictionary of BioSimSpace System objects.
output_files : bool
Whether to write output files immediately or later via saveSystems.
"""
# make sure the proteins / ligands are parametrised before proceeding
with self.workdir:
self.protein.parametrise(params=self.params, reparametrise=False)
for morph in self.morphs:
morph.ligand1.parametrise(params=self.params,
reparametrise=False)
morph.ligand2.parametrise(params=self.params,
reparametrise=False)
# take care of the replicas if there are any
if replica_temps is None or len(replica_temps) == 1:
scales = [1]
else:
sorted_list = sorted(replica_temps)
if sorted_list != replica_temps:
_warnings.warn(
"Input replica temperatures were not in ascending order. Sorting..."
)
replica_temps = sorted_list
scales = [replica_temps[0] / elem for elem in replica_temps]
for i, morph in enumerate(self.morphs):
if intermediate_files:
curdir = _fileio.Dir(morph.name, overwrite=True)
else:
name = _os.path.basename(
_tempfile.TemporaryDirectory().name)
curdir = _fileio.Dir(name, overwrite=True, temp=True)
with curdir:
_logging.info("Creating morph %s..." % morph.name)
morph_BSS, mcs = morph.alignAndCreateMorph(
self.protein.ligand_ref)
morph_BSS = _BSS._SireWrappers.System(morph_BSS)
box = self.protein.complex_template._sire_object.property(
"space")
morph_BSS._sire_object.setProperty("space", box)
# here we scale the equilibrium bond lengths if needed
if scale_dummy_bonds != 1:
n1 = morph.ligand1.molecule.GetNumAtoms()
n2 = morph.ligand2.molecule.GetNumAtoms()
inv_map = {y: x for x, y in mcs}
du2 = [i for i in range(n2) if i not in inv_map.keys()]
inv_map = {
**{x: n1 + y
for x, y in zip(du2, range(n2 - n1))},
**inv_map
}
mcs_smarts = _Chem.MolFromSmarts(dummy_bond_smarts)
matches_lig1 = morph.ligand1.molecule.\
GetSubstructMatches(mcs_smarts)
matches_lig2 = morph.ligand2.molecule.\
GetSubstructMatches(mcs_smarts)
# here we take care of the index transformation
matches_lig2 = [(inv_map[x[0]], inv_map[x[1]])
for x in matches_lig2
if set(x).issubset(inv_map.keys())]
matches_total = [*matches_lig1, *matches_lig2]
morph_BSS = _BSSwrap.rescaleBondedDummies(
morph_BSS, scale_dummy_bonds,
{"Merged_Molecule": matches_total})
complexes = [self.protein.complex_template + morph_BSS]
# solvate and save the prepared complex and morph with the appropriate box size
_logging.info("Solvating...")
complexes = [
_solvate.solvate(complexes[0],
self.params,
box_length=self.box_length_complex,
shell=self.shell,
neutralise=self.neutralise,
ion_conc=self.ion_conc,
centre=self.centre,
work_dir=curdir.path,
filebase="complex")
]
morph_sol = _solvate.solvate(
morph_BSS,
self.params,
box_length=self.box_length_morph,
shell=self.shell,
neutralise=self.neutralise,
ion_conc=self.ion_conc,
centre=self.centre,
work_dir=curdir.path,
filebase="morph")
# rescale complexes for replica exchange if needed
if len(scales) != 1:
_logging.info("Creating replicas...")
complexes = [
_BSSwrap.rescaleSystemParams(
complexes[0],
scale,
includelist=["Merged_Molecule"])
for scale in scales
]
if store_complexes:
self.systems_prep[morph.name] = (morph_sol, complexes)
if output_files:
self.saveSystems({morph.name: (morph_sol, complexes)})
def solvate(complex,
params=None,
box_length=8,
shell=0,
neutralise=True,
ion_conc=0.154,
centre=True,
work_dir=None,
filebase="complex"):
"""
Uses gmx solvate and gmx genion to solvate the system and (optionally) add NaCl ions. This function preserves the
crystal water molecules.
Parameters
----------
complex : BioSimSpace.System or parmed.structure.Structure
The input unsolvated system.
params : ProtoCaller.Parametrise.Params
The input force field parameters.
box_length : float, iterable
Size of the box in nm.
shell : float
Places a layer of water of the specified thickness in nm around the solute.
neutralise : bool
Whether to add counterions to neutralise the system.
ion_conc : float
Ion concentration of NaCl in mol/L.
centre : bool
Whether to centre the system.
work_dir : str
Work directory. Default: current directory.
filebase : str
Output base name of the file.
Returns
-------
complex : BioSimSpace.System or parmed.structure.Structure
The solvated system.
"""
if params is None:
params = _parametrise.Params()
if isinstance(complex, _pmd.Structure):
centrefunc = _pmdwrap.centre
chargefunc = lambda x: round(sum([atom.charge for atom in x.atoms]))
readfunc = _pmdwrap.openFilesAsParmed
elif _PC.BIOSIMSPACE and isinstance(complex,
(_BSS._SireWrappers._molecule.Molecule,
_BSS._SireWrappers._system.System)):
if isinstance(complex, _BSS._SireWrappers._molecule.Molecule):
complex = complex.toSystem()
centrefunc = _BSSwrap.centre
chargefunc = lambda x: round(x.charge().magnitude())
readfunc = _BSS.IO.readMolecules
else:
raise TypeError("Cannot solvate object of type %s" % type(complex))
if not isinstance(box_length, _Iterable):
box_length = 3 * [box_length]
if work_dir is None:
work_dir = _os.path.basename(_tempfile.TemporaryDirectory().name)
temp = True
else:
temp = False
with _fileio.Dir(dirname=work_dir, temp=temp):
# centre
if centre:
complex, box_length, _ = centrefunc(complex, box_length)
# solvate with gmx solvate and load unparametrised waters into memory
files = _PC.IO.GROMACS.saveAsGromacs(filebase, complex)
# reloading the complex fixes some problems with ParmEd
if isinstance(complex, _pmd.Structure):
complex = _pmdwrap.openFilesAsParmed(files)
new_gro = filebase + "_solvated.gro"
command = "{0} solvate -shell {1} -box {2[0]} {2[1]} {2[2]} -cp \"{3}\" -o \"{4}\"".format(
_PC.GROMACSEXE, shell, box_length, files[1], new_gro)
if params.water_points == 4:
command += " -cs tip4p.gro"
_runexternal.runExternal(command, procname="gmx solvate")
complex_solvated = _pmd.load_file(new_gro, skip_bonds=True)
waters = complex_solvated[":SOL"]
# prepare waters for tleap and parametrise
for residue in waters.residues:
residue.name = "WAT"
for i, atom in enumerate(waters.atoms):
if "H" in atom.name:
atom.name = atom.name[0] + atom.name[2]
elif "O" in atom.name:
atom.name = "O"
else:
atom.name = "EPW"
# here we only parametrise a single water molecule in order to gain performance
waters_prep_filenames = [
filebase + "_waters.top", filebase + "_waters.gro"
]
waters[":1"].save(filebase + "_single_wat.pdb")
water = _parametrise.parametriseAndLoadPmd(
params, filebase + "_single_wat.pdb", "water")
_pmdwrap.saveFilesFromParmed(waters, [waters_prep_filenames[1]],
combine="all")
_pmdwrap.saveFilesFromParmed(water, [waters_prep_filenames[0]])
for line in _fileinput.input(waters_prep_filenames[0], inplace=True):
line_new = line.split()
if len(line_new) == 2 and line_new == ["WAT", "1"]:
line = line.replace(
"1",
"{}".format(len(waters.positions) // params.water_points))
print(line, end="")
waters_prep = _pmdwrap.openFilesAsParmed(waters_prep_filenames)
waters_prep.box = _pmd.load_file(files[1], skip_bonds=True).box
if any([neutralise, ion_conc, shell]):
for residue in waters_prep.residues:
residue.name = "SOL"
_pmdwrap.saveFilesFromParmed(waters_prep, waters_prep_filenames)
# add ions
if any([neutralise, ion_conc, shell]):
# write an MDP file
_protocol.Protocol(use_preset="default").write("GROMACS", "ions")
# neutralise if needed
charge = chargefunc(complex) if neutralise else 0
volume = box_length[0] * box_length[1] * box_length[2] * 10**-24
n_Na, n_Cl = [
int(volume * 6.022 * 10**23 * ion_conc) - abs(charge) // 2
] * 2
if neutralise:
if charge < 0:
n_Na -= charge
else:
n_Cl += charge
# add ions with gmx genion
ions_prep_filenames = [
filebase + "_ions.top", filebase + "_ions.gro"
]
command = "{0} grompp -f ions.mdp -p {1} -c {2} -o \"{3}_solvated.tpr\"".format(
_PC.GROMACSEXE, *waters_prep_filenames, filebase)
_runexternal.runExternal(command, procname="gmx grompp")
command = "{{ echo 2; }} | {0} genion -s \"{1}_solvated.tpr\" -o \"{2}\" -nn {3} -np {4}".format(
_PC.GROMACSEXE, filebase, ions_prep_filenames[1], n_Cl, n_Na)
_runexternal.runExternal(command, procname="gmx genion")
# prepare waters for tleap
ions = _pmd.load_file(ions_prep_filenames[1], skip_bonds=True)
for residue in ions.residues:
if residue.name == "SOL":
residue.name = "WAT"
# here we only parametrise single ions to gain performance
ion = ions[":WAT"][":1"] + ions[":NA"][":1"] + ions[":CL"][":1"]
max_len = len(ion.residues)
ion.save(filebase + "_single_ion.pdb")
ion = _parametrise.parametriseAndLoadPmd(
params, filebase + "_single_ion.pdb", "water")
_pmdwrap.saveFilesFromParmed(ions, [ions_prep_filenames[1]],
combine="all")
_pmdwrap.saveFilesFromParmed(ion, [ions_prep_filenames[0]])
mol_dict = {}
for line in _fileinput.input(ions_prep_filenames[0], inplace=True):
line_new = line.split()
if len(line_new) == 2 and line_new[0] in [
"WAT", "NA", "CL"
] and line_new[1] == "1":
n_mols = len(ions[":{}".format(line_new[0])].positions)
if line_new[0] == "WAT":
n_mols //= params.water_points
line = line.replace("1", "{}".format(n_mols))
mol_dict[line_new[0]] = line
# preserve the order of water, sodium and chloride
if len(mol_dict) == max_len:
for x in ["WAT", "NA", "CL"]:
if x in mol_dict.keys():
print(mol_dict[x], end="")
mol_dict = {}
else:
print(line, end="")
return complex + readfunc(ions_prep_filenames)
else:
complex + readfunc(waters_prep_filenames)