def _solvent_amber_files(self,
lig1_name,
lig1_mol2,
lig1_frcmod,
prmtop_filename,
inpcrd_filename,
leaprc='leaprc.gaff'):
logger = logging.getLogger(__name__)
logger.setLevel(logging.DEBUG)
handler = logging.FileHandler('tleap_solvent_phase.log')
logger.addHandler(handler)
#Get absolute paths for input/output
lig1_mol2 = os.path.abspath(lig1_mol2)
lig1_frcmod = os.path.abspath(lig1_frcmod)
prmtop_filename = os.path.abspath(prmtop_filename)
inpcrd_filename = os.path.abspath(inpcrd_filename)
#Work in a temporary directory, on hard coded filenames, to avoid any issues AMBER may have with spaces and other special characters in filenames
with mdtraj.utils.enter_temp_directory():
shutil.copy(lig1_mol2, 'file.mol2')
shutil.copy(lig1_frcmod, 'file.frcmod')
tleap_input = f"""
source oldff/leaprc.ff99SB
source {leaprc}
loadamberparams file.frcmod
LIG = loadmol2 file.mol2
check LIG
solvateBox LIG TIP3PBOX 10.0
addIons LIG Cl- 5
addIons LIG K+ 5
saveamberparm LIG out.prmtop out.inpcrd
quit
"""
file_handle = open('tleap_commands', 'w')
file_handle.writelines(tleap_input)
file_handle.close()
cmd = f'tleap -f {file_handle.name}'
output = getoutput(cmd)
logger.debug(output)
#Copy back target files
shutil.copy('out.prmtop', prmtop_filename)
shutil.copy('out.inpcrd', inpcrd_filename)
logfile = os.path.abspath('tleap_solvent_phase.log')
destination = os.path.abspath('data')
shutil.move(logfile,
os.path.join(destination, os.path.basename(logfile)))
return prmtop_filename, inpcrd_filename
def build_peptide_tleap(amino_acids):
"""
Use tleap to generate a peptide topology
and positions.
Parameters
----------
amino_acids : list of str
List of amino acids and caps in three-letter names
e.g. ['ACE', 'ALA', 'NME']
Returns
-------
topology : simtk.openmm.app.Topology object
Topology of the amino acid
positions : [n, 3] array
positions of atoms
"""
aa_str = " ".join(amino_acids)
filename = "".join(amino_acids)
tleapstr = """
source oldff/leaprc.ff99SBildn
system = sequence {{ {amino_acid} }}
saveamberparm system {filename}.prmtop {filename}.inpcrd
""".format(amino_acid=aa_str, filename=filename)
cwd = os.getcwd()
temp_dir = tempfile.mkdtemp()
os.chdir(temp_dir)
tleap_file = open('tleap_commands', 'w')
tleap_file.writelines(tleapstr)
tleap_file.close()
tleap_cmd_str = "tleap -f %s " % tleap_file.name
#call tleap, log output to logger
output = getoutput(tleap_cmd_str)
logging.debug(output)
prmtop = app.AmberPrmtopFile("{filename}.prmtop".format(filename=filename))
inpcrd = app.AmberInpcrdFile("{filename}.inpcrd".format(filename=filename))
topology = prmtop.topology
positions = inpcrd.positions
os.chdir(cwd)
shutil.rmtree(temp_dir)
return topology, positions
def build_peptide_tleap(amino_acids):
"""
Use tleap to generate a peptide topology
and positions.
Parameters
----------
amino_acids : list of str
List of amino acids and caps in three-letter names
e.g. ['ACE', 'ALA', 'NME']
Returns
-------
topology : simtk.openmm.app.Topology object
Topology of the amino acid
positions : [n, 3] array
positions of atoms
"""
aa_str = " ".join(amino_acids)
filename = "".join(amino_acids)
tleapstr = """
source oldff/leaprc.ff99SBildn
system = sequence {{ {amino_acid} }}
saveamberparm system {filename}.prmtop {filename}.inpcrd
""".format(amino_acid=aa_str, filename=filename)
cwd = os.getcwd()
temp_dir = tempfile.mkdtemp()
os.chdir(temp_dir)
tleap_file = open('tleap_commands', 'w')
tleap_file.writelines(tleapstr)
tleap_file.close()
tleap_cmd_str = "tleap -f %s " % tleap_file.name
#call tleap, log output to logger
output = getoutput(tleap_cmd_str)
logging.debug(output)
prmtop = app.AmberPrmtopFile("{filename}.prmtop".format(filename=filename))
inpcrd = app.AmberInpcrdFile("{filename}.inpcrd".format(filename=filename))
topology = prmtop.topology
positions = inpcrd.positions
os.chdir(cwd)
shutil.rmtree(temp_dir)
return topology, positions
def build_mixture_prmtop(mol2_filenames, frcmod_filenames, box_filename, prmtop_filename, inpcrd_filename, water_model = 'TIP3P'):
"""Create a prmtop and inpcrd from a collection of mol2 and frcmod files
as well as a single box PDB. We have used this for setting up
simulations of neat liquids or binary mixtures.
Parameters
----------
mol2_filenames : list(str)
Filenames of GAFF flavored mol2 files. Each must contain exactly
ONE ligand.
frcmod_filenames : str
Filename of input GAFF frcmod filenames.
box_filename : str
Filename of PDB containing an arbitrary box of the mol2 molecules.
prmtop_filename : str
output prmtop filename. Should have suffix .prmtop
inpcrd_filename : str
output inpcrd filename. Should have suffix .inpcrd
water_model : str, optional. Default: "TIP3P"
String specifying water model to be used IF water is present as a component of the mixture. Valid options are currently "TIP3P", "SPC", or None. If None is specified, flexible GAFF-water will be used as for any other solute (old behavior).
Returns
-------
tleap_commands : str
The string of commands piped to tleap for building the prmtop
and inpcrd files. This will *already* have been run, but the
output can be useful for debugging or archival purposes. However,
this will reflect temporary file names for both input and output
file as these are used to avoid tleap filename restrictions.
Notes
-----
This can be easily broken if there are missing, duplicated, or
inconsistent ligand residue names in your box, mol2, and frcmod files.
You can use mdtraj to edit the residue names with something like
this: trj.top.residue(0).name = "L1"
"""
# Check for one residue name per mol2 file and uniqueness between all mol2 files
all_names = set()
for filename in mol2_filenames:
t = md.load(filename)
names = set([r.name for r in t.top.residues])
if len(names) != 1:
raise(ValueError("Must have a SINGLE residue name in each mol2 file."))
all_names = all_names.union(list(names))
if len(all_names) != len(mol2_filenames):
raise(ValueError("Must have UNIQUE residue names in each mol2 file."))
if len(mol2_filenames) != len(frcmod_filenames):
raise(ValueError("Must provide an equal number of frcmod and mol2 file names."))
#Get number of files
nfiles = len(mol2_filenames)
#Check validity of water model options
valid_water = ['TIP3P', 'SPC', None]
if not water_model in valid_water:
raise(ValueError("Must provide a valid water model."))
#If we are requesting a different water model, check if there is water present
if not water_model==None:
parmed = import_("parmed")
solventIsWater = []
waterPresent = False
for i in range(nfiles):
mol = parmed.load_file( mol2_filenames[i] )
#Check if it is water by checking GAFF atom names
types = [ atom.type for atom in mol.atoms ]
if 'oh' in types and types.count('ho')==2 and len(types)==3:
solventIsWater.append(True)
waterPresent = True
else:
solventIsWater.append(False)
#In this case, if we have any water, we will now work on fewer .mol2 and .frcmod files and instead use the force field files for those. So, reduce nfiles and remove the files we don't need from the .mol2 and .frcmod filename lists
#After doing so, go on to interpret the specified water model and compose the water model string needed for tleap
if waterPresent:
new_mol2_filenames = []
new_frcmod_filenames = []
water_mol2_filenames = []
for i in range( nfiles ):
if not solventIsWater[i]:
new_mol2_filenames.append( mol2_filenames[i] )
new_frcmod_filenames.append( frcmod_filenames[i] )
else:
water_mol2_filenames.append( mol2_filenames[i] )
nfiles = len(new_mol2_filenames)
mol2_filenames = new_mol2_filenames
frcmod_filenames = new_frcmod_filenames
#Now interpret the specified water model and translate into AMBER nomenclature
if water_model=='TIP3P':
water_model = 'TP3'
elif water_model =='SPC':
water_model = 'SPC'
else:
raise(ValueError("Cannot translate specified water model into one of the available models."))
#Compose string for loading specified water molecule
water_string = '\n'
water_names = [md.load(filename).top.residue(0).name for filename in water_mol2_filenames]
for name in water_names:
water_string += '%s = %s\n' % (name, water_model )
#Also if not TIP3P, update to source correct frcmod file
if water_model == 'SPC':
water_string += 'loadamberparams frcmod.spce\n'
elif water_model =='TP3':
continue
else:
raise(ValueError("Cannot identify water frcmod file to be loaded."))
#Rename water atoms in box file to match what is expected by AMBER
packmol = import_("openmoltools.packmol")
packmol.rename_water_atoms(box_filename)
else:
waterPresent = False
#Make temporary, hardcoded filenames for mol2 and frcmod input to avoid tleap filename restrictions
tmp_mol2_filenames = [ 'in%d.mol2' % n for n in range(nfiles) ]
tmp_frcmod_filenames = [ 'in%d.frcmod' % n for n in range(nfiles) ]
#Make temporary, hardcoded filenames for output files to avoid tleap filename restrictions
tmp_prmtop_filename = 'out.prmtop'
tmp_inpcrd_filename = 'out.inpcrd'
tmp_box_filename = 'tbox.pdb'
#Build absolute paths of input files so we can use context and temporary directory
infiles = mol2_filenames + frcmod_filenames + [box_filename]
infiles = [ os.path.abspath(filenm) for filenm in infiles ]
#Build absolute paths of output files so we can copy them back
prmtop_filename = os.path.abspath( prmtop_filename )
inpcrd_filename = os.path.abspath( inpcrd_filename )
#Use temporary directory and do the setup
with mdtraj.utils.enter_temp_directory():
#Copy input files to temporary file names in target directory
for (infile, outfile) in zip( infiles, tmp_mol2_filenames+tmp_frcmod_filenames+[tmp_box_filename] ):
shutil.copy( infile, outfile)
logger.debug('Copying input file %s to %s...\n' % (infile, outfile))
all_names = [md.load(filename).top.residue(0).name for filename in tmp_mol2_filenames]
mol2_section = "\n".join("%s = loadmol2 %s" % (all_names[k], filename) for k, filename in enumerate(tmp_mol2_filenames))
#If non-GAFF water is present, load desired parameters for that water as well.
if waterPresent:
mol2_section += water_string
amberparams_section = "\n".join("loadamberparams %s" % (filename) for k, filename in enumerate(tmp_frcmod_filenames))
tleap_commands = TLEAP_TEMPLATE % dict(mol2_section=mol2_section, amberparams_section=amberparams_section, box_filename=tmp_box_filename, prmtop_filename=tmp_prmtop_filename, inpcrd_filename=tmp_inpcrd_filename)
print(tleap_commands)
file_handle = open('tleap_commands', 'w')
file_handle.writelines(tleap_commands)
file_handle.close()
logger.debug('Running tleap in temporary directory.')
cmd = "tleap -f %s " % file_handle.name
logger.debug(cmd)
output = getoutput(cmd)
logger.debug(output)
check_for_errors( output, other_errors = ['Improper number of arguments'], ignore_errors = ['unperturbed charge of the unit', 'ignoring the error'] )
#Copy stuff back to right filenames
for (tfile, finalfile) in zip( [tmp_prmtop_filename, tmp_inpcrd_filename], [prmtop_filename, inpcrd_filename] ):
shutil.copy( tfile, finalfile)
return tleap_commands
def run_tleap(molecule_name, gaff_mol2_filename, frcmod_filename, prmtop_filename=None, inpcrd_filename=None, log_debug_output=False):
"""Run AmberTools tleap to create simulation files for AMBER
Parameters
----------
molecule_name : str
The name of the molecule
gaff_mol2_filename : str
GAFF format mol2 filename produced by antechamber
frcmod_filename : str
Amber frcmod file produced by prmchk
prmtop_filename : str, optional, default=None
Amber prmtop file produced by tleap, defaults to molecule_name
inpcrd_filename : str, optional, default=None
Amber inpcrd file produced by tleap, defaults to molecule_name
log_debug_output : bool, optional, default=False
If true, will send output of tleap to logger.
Returns
-------
prmtop_filename : str
Amber prmtop file produced by tleap
inpcrd_filename : str
Amber inpcrd file produced by tleap
"""
if prmtop_filename is None:
prmtop_filename = "%s.prmtop" % molecule_name
if inpcrd_filename is None:
inpcrd_filename = "%s.inpcrd" % molecule_name
#Get absolute paths for input/output
gaff_mol2_filename = os.path.abspath( gaff_mol2_filename )
frcmod_filename = os.path.abspath( frcmod_filename )
prmtop_filename = os.path.abspath( prmtop_filename )
inpcrd_filename = os.path.abspath( inpcrd_filename )
#Work in a temporary directory, on hard coded filenames, to avoid any issues AMBER may have with spaces and other special characters in filenames
with mdtraj.utils.enter_temp_directory():
shutil.copy( gaff_mol2_filename, 'file.mol2' )
shutil.copy( frcmod_filename, 'file.frcmod' )
tleap_input = """
source oldff/leaprc.ff99SB
source leaprc.gaff
LIG = loadmol2 file.mol2
check LIG
loadamberparams file.frcmod
saveamberparm LIG out.prmtop out.inpcrd
quit
"""
file_handle = open('tleap_commands', 'w')
file_handle.writelines(tleap_input)
file_handle.close()
cmd = "tleap -f %s " % file_handle.name
if log_debug_output: logger.debug(cmd)
output = getoutput(cmd)
if log_debug_output: logger.debug(output)
check_for_errors( output, other_errors = ['Improper number of arguments'] )
#Copy back target files
shutil.copy( 'out.prmtop', prmtop_filename )
shutil.copy( 'out.inpcrd', inpcrd_filename )
return prmtop_filename, inpcrd_filename
def run_antechamber(molecule_name, input_filename, charge_method="bcc", net_charge=None, gaff_mol2_filename=None, frcmod_filename=None,
input_format='mol2', resname=False, log_debug_output=False):
"""Run AmberTools antechamber and parmchk2 to create GAFF mol2 and frcmod files.
Parameters
----------
molecule_name : str
Name of the molecule to be parameterized, will be used in output filenames.
ligand_filename : str
The molecule to be parameterized. Must be tripos mol2 format.
charge_method : str, optional
If not None, the charge method string will be passed to Antechamber.
net_charge : int, optional
If not None, net charge of the molecule to be parameterized.
If None, Antechamber sums up partial charges from the input file.
gaff_mol2_filename : str, optional, default=None
Name of GAFF mol2 filename to output. If None, uses local directory
and molecule_name
frcmod_filename : str, optional, default=None
Name of GAFF frcmod filename to output. If None, uses local directory
and molecule_name
input_format : str, optional, default='mol2'
Format specifier for input file to pass to antechamber.
resname : bool, optional, default=False
Set the residue name used within output files to molecule_name
log_debug_output : bool, optional, default=False
If true, will send output of tleap to logger.
Returns
-------
gaff_mol2_filename : str
GAFF format mol2 filename produced by antechamber
frcmod_filename : str
Amber frcmod file produced by prmchk
"""
utils = import_("openmoltools.utils")
ext = utils.parse_ligand_filename(input_filename)[1]
if gaff_mol2_filename is None:
gaff_mol2_filename = molecule_name + '.gaff.mol2'
if frcmod_filename is None:
frcmod_filename = molecule_name + '.frcmod'
#Build absolute paths for input and output files
gaff_mol2_filename = os.path.abspath( gaff_mol2_filename )
frcmod_filename = os.path.abspath( frcmod_filename )
input_filename = os.path.abspath( input_filename )
def read_file_contents(filename):
infile = open(filename, 'r')
contents = infile.read()
infile.close()
return contents
#Use temporary directory context to do this to avoid issues with spaces in filenames, etc.
with mdtraj.utils.enter_temp_directory():
local_input_filename = 'in.' + input_format
shutil.copy( input_filename, local_input_filename )
# Run antechamber.
cmd = "antechamber -i %(local_input_filename)s -fi %(input_format)s -o out.mol2 -fo mol2 -s 2" % vars()
if charge_method is not None:
cmd += ' -c %s' % charge_method
if net_charge is not None:
cmd += ' -nc %d' % net_charge
if resname:
cmd += ' -rn %s' % molecule_name
if log_debug_output: logger.debug(cmd)
output = getoutput(cmd)
if not os.path.exists('out.mol2'):
msg = "antechamber failed to produce output mol2 file\n"
msg += "command: %s\n" % cmd
msg += "output:\n"
msg += 8 * "----------" + '\n'
msg += output
msg += 8 * "----------" + '\n'
msg += "input mol2:\n"
msg += 8 * "----------" + '\n'
msg += read_file_contents(local_input_filename)
msg += 8 * "----------" + '\n'
raise Exception(msg)
if log_debug_output: logger.debug(output)
# Run parmchk.
cmd = "parmchk2 -i out.mol2 -f mol2 -o out.frcmod"
if log_debug_output: logger.debug(cmd)
output = getoutput(cmd)
if not os.path.exists('out.frcmod'):
msg = "parmchk2 failed to produce output frcmod file\n"
msg += "command: %s\n" % cmd
msg += "output:\n"
msg += 8 * "----------" + '\n'
msg += output
msg += 8 * "----------" + '\n'
msg += "input mol2:\n"
msg += 8 * "----------" + '\n'
msg += read_file_contents('out.mol2')
msg += 8 * "----------" + '\n'
raise Exception(msg)
if log_debug_output: logger.debug(output)
check_for_errors(output)
#Copy back
shutil.copy( 'out.mol2', gaff_mol2_filename )
shutil.copy( 'out.frcmod', frcmod_filename )
return gaff_mol2_filename, frcmod_filename
def do_solvate( top_filename, gro_filename, top_solv_filename, gro_solv_filename, box_dim, box_type, water_model, water_top, FF = 'amber99sb-ildn.ff' ):
""" This function creates water solvated molecule coordinate files and its corresponding topology
PARAMETERS:
top_filename: str
Topology path/filename
gro_filename: str
Coordinates path/filename
top_solv_filename: str
Topology path/filename (solvated system)
gro_solv_filename: str
Coordinates path/filename (solvated system)
box_dim: float
cubic box dimension (nm); will be passed to GROMACS wiin .2f format
box_type: str
box type (string passed to gmx solvate)
water_model: str
Water model string to tell gmx solvate to use when solvating, i.e. "spc216"
water_top: str
Water include file to ensure is present in topology file, i.e. "tip3p.itp"
FF : str, optional, default = 'amber99sb-ildn.ff'
String specifying base force field directory for include files (i.e. 'amber99sb-ildn.ff').
NOTES:
-----
Primarily tested on 3 point water models. May need adjustment for other models.
"""
#Setting up proper environment variables
os.environ['GMX_MAXBACKUP'] = '-1' # Avoids unnecessary GROMACS backup files
os.environ['GMX_NO_QUOTES'] = '1' # Supresses end-of-file quotes (gcq)
#Get absolute paths for input/output
top_filename = os.path.abspath( top_filename )
gro_filename = os.path.abspath( gro_filename )
top_solv_filename = os.path.abspath( top_solv_filename )
gro_solv_filename = os.path.abspath( gro_solv_filename )
with mdtraj.utils.enter_temp_directory(): #Work on hard coded filenames in temporary directory
shutil.copy( gro_filename, 'in.gro' )
shutil.copy( top_filename, 'out.top' )
#string with the Gromacs 5.0.4 box generating commands
cmdbox = 'gmx editconf -f in.gro -o out.gro -c -d %.2f -bt %s' % ( box_dim, box_type)
output = getoutput(cmdbox)
logger.debug(output)
check_for_errors(output)
#string with the Gromacs 5.0.4 solvation tool (it is not genbox anymore)
cmdsolv = 'gmx solvate -cp out.gro -cs %s -o out.gro -p out.top' % (water_model)
output = getoutput(cmdsolv)
logger.debug(output)
check_for_errors(output)
#Insert Force Field specifications
ensure_forcefield( 'out.top', 'out.top', FF = FF)
#Insert line for water topology portion of the code
try:
file = open('out.top','r')
text = file.readlines()
file.close()
except:
raise NameError('The file out.top is missing' )
#Insert water model
wateritp = os.path.join(FF, water_top ) # e.g water_top = 'tip3p.itp'
index = 0
while '[ system ]' not in text[index]:
index += 1
text[index] = '#include "%s"\n\n' % wateritp + text[index]
#Write the topology file
try:
file = open('out.top','w+')
file.writelines( text )
file.close()
except:
raise NameError('The file %s is missing' % 'out.top')
#Check if file exist and is not empty;
if os.stat( 'out.gro' ) == 0 or os.stat( 'out.top' ).st_size == 0:
raise(ValueError("Solvent insertion failed"))
#Copy back files
shutil.copy( 'out.gro', gro_solv_filename )
shutil.copy( 'out.top', top_solv_filename )
return
def build_mixture_prmtop(mol2_filenames, frcmod_filenames, box_filename, prmtop_filename, inpcrd_filename, water_model = 'TIP3P'):
"""Create a prmtop and inpcrd from a collection of mol2 and frcmod files
as well as a single box PDB. We have used this for setting up
simulations of neat liquids or binary mixtures.
Parameters
----------
mol2_filenames : list(str)
Filenames of GAFF flavored mol2 files. Each must contain exactly
ONE ligand.
frcmod_filenames : str
Filename of input GAFF frcmod filenames.
box_filename : str
Filename of PDB containing an arbitrary box of the mol2 molecules.
prmtop_filename : str
output prmtop filename. Should have suffix .prmtop
inpcrd_filename : str
output inpcrd filename. Should have suffix .inpcrd
water_model : str, optional. Default: "TIP3P"
String specifying water model to be used IF water is present as a component of the mixture. Valid options are currently "TIP3P", "SPC", or None. If None is specified, flexible GAFF-water will be used as for any other solute (old behavior).
Returns
-------
tleap_commands : str
The string of commands piped to tleap for building the prmtop
and inpcrd files. This will *already* have been run, but the
output can be useful for debugging or archival purposes. However,
this will reflect temporary file names for both input and output
file as these are used to avoid tleap filename restrictions.
Notes
-----
This can be easily broken if there are missing, duplicated, or
inconsistent ligand residue names in your box, mol2, and frcmod files.
You can use mdtraj to edit the residue names with something like
this: trj.top.residue(0).name = "L1"
"""
# Check for one residue name per mol2 file and uniqueness between all mol2 files
all_names = set()
for filename in mol2_filenames:
t = md.load(filename)
names = set([r.name for r in t.top.residues])
if len(names) != 1:
raise(ValueError("Must have a SINGLE residue name in each mol2 file."))
all_names = all_names.union(list(names))
if len(all_names) != len(mol2_filenames):
raise(ValueError("Must have UNIQUE residue names in each mol2 file."))
if len(mol2_filenames) != len(frcmod_filenames):
raise(ValueError("Must provide an equal number of frcmod and mol2 file names."))
#Get number of files
nfiles = len(mol2_filenames)
#Check validity of water model options
valid_water = ['TIP3P', 'SPC', None]
if not water_model in valid_water:
raise(ValueError("Must provide a valid water model."))
#If we are requesting a different water model, check if there is water present
if not water_model==None:
parmed = import_("parmed")
solventIsWater = []
waterPresent = False
for i in range(nfiles):
mol = parmed.load_file( mol2_filenames[i] )
#Check if it is water by checking GAFF atom names
types = [ atom.type for atom in mol.atoms ]
if 'oh' in types and types.count('ho')==2 and len(types)==3:
solventIsWater.append(True)
waterPresent = True
else:
solventIsWater.append(False)
#In this case, if we have any water, we will now work on fewer .mol2 and .frcmod files and instead use the force field files for those. So, reduce nfiles and remove the files we don't need from the .mol2 and .frcmod filename lists
#After doing so, go on to interpret the specified water model and compose the water model string needed for tleap
if waterPresent:
new_mol2_filenames = []
new_frcmod_filenames = []
water_mol2_filenames = []
for i in range( nfiles ):
if not solventIsWater[i]:
new_mol2_filenames.append( mol2_filenames[i] )
new_frcmod_filenames.append( frcmod_filenames[i] )
else:
water_mol2_filenames.append( mol2_filenames[i] )
nfiles = len(new_mol2_filenames)
mol2_filenames = new_mol2_filenames
frcmod_filenames = new_frcmod_filenames
#Now interpret the specified water model and translate into AMBER nomenclature
if water_model=='TIP3P':
water_model = 'TP3'
elif water_model =='SPC':
water_model = 'SPC'
else:
raise(ValueError("Cannot translate specified water model into one of the available models."))
#Compose string for loading specified water molecule
water_string = '\n'
water_names = [md.load(filename).top.residue(0).name for filename in water_mol2_filenames]
for name in water_names:
water_string += '%s = %s\n' % (name, water_model )
#Also if not TIP3P, update to source correct frcmod file
if water_model == 'SPC':
water_string += 'loadamberparams frcmod.spce\n'
elif water_model =='TP3':
continue
else:
raise(ValueError("Cannot identify water frcmod file to be loaded."))
#Rename water atoms in box file to match what is expected by AMBER
packmol = import_("openmoltools.packmol")
packmol.rename_water_atoms(box_filename)
else:
waterPresent = False
#Make temporary, hardcoded filenames for mol2 and frcmod input to avoid tleap filename restrictions
tmp_mol2_filenames = [ 'in%d.mol2' % n for n in range(nfiles) ]
tmp_frcmod_filenames = [ 'in%d.frcmod' % n for n in range(nfiles) ]
#Make temporary, hardcoded filenames for output files to avoid tleap filename restrictions
tmp_prmtop_filename = 'out.prmtop'
tmp_inpcrd_filename = 'out.inpcrd'
tmp_box_filename = 'tbox.pdb'
#Build absolute paths of input files so we can use context and temporary directory
infiles = mol2_filenames + frcmod_filenames + [box_filename]
infiles = [ os.path.abspath(filenm) for filenm in infiles ]
#Build absolute paths of output files so we can copy them back
prmtop_filename = os.path.abspath( prmtop_filename )
inpcrd_filename = os.path.abspath( inpcrd_filename )
#Use temporary directory and do the setup
with mdtraj.utils.enter_temp_directory():
#Copy input files to temporary file names in target directory
for (infile, outfile) in zip( infiles, tmp_mol2_filenames+tmp_frcmod_filenames+[tmp_box_filename] ):
shutil.copy( infile, outfile)
logger.debug('Copying input file %s to %s...\n' % (infile, outfile))
all_names = [md.load(filename).top.residue(0).name for filename in tmp_mol2_filenames]
mol2_section = "\n".join("%s = loadmol2 %s" % (all_names[k], filename) for k, filename in enumerate(tmp_mol2_filenames))
#If non-GAFF water is present, load desired parameters for that water as well.
if waterPresent:
mol2_section += water_string
amberparams_section = "\n".join("loadamberparams %s" % (filename) for k, filename in enumerate(tmp_frcmod_filenames))
tleap_commands = TLEAP_TEMPLATE % dict(mol2_section=mol2_section, amberparams_section=amberparams_section, box_filename=tmp_box_filename, prmtop_filename=tmp_prmtop_filename, inpcrd_filename=tmp_inpcrd_filename)
print(tleap_commands)
file_handle = open('tleap_commands', 'w')
file_handle.writelines(tleap_commands)
file_handle.close()
logger.debug('Running tleap in temporary directory.')
cmd = "tleap -f %s " % file_handle.name
logger.debug(cmd)
output = getoutput(cmd)
logger.debug(output)
check_for_errors( output, other_errors = ['Improper number of arguments'], ignore_errors = ['unperturbed charge of the unit', 'ignoring the error'] )
#Copy stuff back to right filenames
for (tfile, finalfile) in zip( [tmp_prmtop_filename, tmp_inpcrd_filename], [prmtop_filename, inpcrd_filename] ):
shutil.copy( tfile, finalfile)
return tleap_commands
def run_tleap(molecule_name, gaff_mol2_filename, frcmod_filename, prmtop_filename=None, inpcrd_filename=None, log_debug_output=False):
"""Run AmberTools tleap to create simulation files for AMBER
Parameters
----------
molecule_name : str
The name of the molecule
gaff_mol2_filename : str
GAFF format mol2 filename produced by antechamber
frcmod_filename : str
Amber frcmod file produced by prmchk
prmtop_filename : str, optional, default=None
Amber prmtop file produced by tleap, defaults to molecule_name
inpcrd_filename : str, optional, default=None
Amber inpcrd file produced by tleap, defaults to molecule_name
log_debug_output : bool, optional, default=False
If true, will send output of tleap to logger.
Returns
-------
prmtop_filename : str
Amber prmtop file produced by tleap
inpcrd_filename : str
Amber inpcrd file produced by tleap
"""
if prmtop_filename is None:
prmtop_filename = "%s.prmtop" % molecule_name
if inpcrd_filename is None:
inpcrd_filename = "%s.inpcrd" % molecule_name
#Get absolute paths for input/output
gaff_mol2_filename = os.path.abspath( gaff_mol2_filename )
frcmod_filename = os.path.abspath( frcmod_filename )
prmtop_filename = os.path.abspath( prmtop_filename )
inpcrd_filename = os.path.abspath( inpcrd_filename )
#Work in a temporary directory, on hard coded filenames, to avoid any issues AMBER may have with spaces and other special characters in filenames
with mdtraj.utils.enter_temp_directory():
shutil.copy( gaff_mol2_filename, 'file.mol2' )
shutil.copy( frcmod_filename, 'file.frcmod' )
tleap_input = """
source oldff/leaprc.ff99SB
source leaprc.gaff
LIG = loadmol2 file.mol2
check LIG
loadamberparams file.frcmod
saveamberparm LIG out.prmtop out.inpcrd
quit
"""
file_handle = open('tleap_commands', 'w')
file_handle.writelines(tleap_input)
file_handle.close()
cmd = "tleap -f %s " % file_handle.name
if log_debug_output: logger.debug(cmd)
output = getoutput(cmd)
if log_debug_output: logger.debug(output)
check_for_errors( output, other_errors = ['Improper number of arguments'] )
#Copy back target files
shutil.copy( 'out.prmtop', prmtop_filename )
shutil.copy( 'out.inpcrd', inpcrd_filename )
return prmtop_filename, inpcrd_filename
def run_antechamber(molecule_name, input_filename, charge_method="bcc", net_charge=None, gaff_mol2_filename=None, frcmod_filename=None,
input_format='mol2', resname=False, log_debug_output=False):
"""Run AmberTools antechamber and parmchk2 to create GAFF mol2 and frcmod files.
Parameters
----------
molecule_name : str
Name of the molecule to be parameterized, will be used in output filenames.
ligand_filename : str
The molecule to be parameterized. Must be tripos mol2 format.
charge_method : str, optional
If not None, the charge method string will be passed to Antechamber.
net_charge : int, optional
If not None, net charge of the molecule to be parameterized.
If None, Antechamber sums up partial charges from the input file.
gaff_mol2_filename : str, optional, default=None
Name of GAFF mol2 filename to output. If None, uses local directory
and molecule_name
frcmod_filename : str, optional, default=None
Name of GAFF frcmod filename to output. If None, uses local directory
and molecule_name
input_format : str, optional, default='mol2'
Format specifier for input file to pass to antechamber.
resname : bool, optional, default=False
Set the residue name used within output files to molecule_name
log_debug_output : bool, optional, default=False
If true, will send output of tleap to logger.
Returns
-------
gaff_mol2_filename : str
GAFF format mol2 filename produced by antechamber
frcmod_filename : str
Amber frcmod file produced by prmchk
"""
utils = import_("openmoltools.utils")
ext = utils.parse_ligand_filename(input_filename)[1]
if gaff_mol2_filename is None:
gaff_mol2_filename = molecule_name + '.gaff.mol2'
if frcmod_filename is None:
frcmod_filename = molecule_name + '.frcmod'
#Build absolute paths for input and output files
gaff_mol2_filename = os.path.abspath( gaff_mol2_filename )
frcmod_filename = os.path.abspath( frcmod_filename )
input_filename = os.path.abspath( input_filename )
def read_file_contents(filename):
infile = open(filename, 'r')
contents = infile.read()
infile.close()
return contents
#Use temporary directory context to do this to avoid issues with spaces in filenames, etc.
with mdtraj.utils.enter_temp_directory():
local_input_filename = 'in.' + input_format
shutil.copy( input_filename, local_input_filename )
# Run antechamber.
cmd = "antechamber -i %(local_input_filename)s -fi %(input_format)s -o out.mol2 -fo mol2 -s 2" % vars()
if charge_method is not None:
cmd += ' -c %s' % charge_method
if net_charge is not None:
cmd += ' -nc %d' % net_charge
if resname:
cmd += ' -rn %s' % molecule_name
if log_debug_output: logger.debug(cmd)
output = getoutput(cmd)
if not os.path.exists('out.mol2'):
msg = "antechamber failed to produce output mol2 file\n"
msg += "command: %s\n" % cmd
msg += "output:\n"
msg += 8 * "----------" + '\n'
msg += output
msg += 8 * "----------" + '\n'
msg += "input mol2:\n"
msg += 8 * "----------" + '\n'
msg += read_file_contents(local_input_filename)
msg += 8 * "----------" + '\n'
raise Exception(msg)
if log_debug_output: logger.debug(output)
# Run parmchk.
cmd = "parmchk2 -i out.mol2 -f mol2 -o out.frcmod"
if log_debug_output: logger.debug(cmd)
output = getoutput(cmd)
if not os.path.exists('out.frcmod'):
msg = "parmchk2 failed to produce output frcmod file\n"
msg += "command: %s\n" % cmd
msg += "output:\n"
msg += 8 * "----------" + '\n'
msg += output
msg += 8 * "----------" + '\n'
msg += "input mol2:\n"
msg += 8 * "----------" + '\n'
msg += read_file_contents('out.mol2')
msg += 8 * "----------" + '\n'
raise Exception(msg)
if log_debug_output: logger.debug(output)
check_for_errors(output)
#Copy back
shutil.copy( 'out.mol2', gaff_mol2_filename )
shutil.copy( 'out.frcmod', frcmod_filename )
return gaff_mol2_filename, frcmod_filename
def build_mixture_prmtop(mol2_filenames, frcmod_filenames, box_filename,
prmtop_filename, inpcrd_filename):
"""Create a prmtop and inpcrd from a collection of mol2 and frcmod files as well as a single box PDB.
We have used this for setting up simulations of binary mixtures.
- Original code by : Chodera Lab / Openmoltools project (https://github.com/choderalab/openmoltools)
Parameters
----------
mol2_filenames : list(str)
Filenames of GAFF flavored mol2 files. Each must contain exactly
ONE ligand.
frcmod_filenames : str
Filename of input GAFF frcmod filenames.
box_filename : str
Filename of PDB containing an arbitrary box of the mol2 molecules.
prmtop_filename : str
output prmtop filename. Should have suffix .prmtop
inpcrd_filename : str
output inpcrd filename. Should have suffix .inpcrd
water_model : str, optional. Default: "TIP3P"
String specifying water model to be used IF water is present as a component of the mixture. Valid options are currently "TIP3P", "SPC", or None. If None is specified, flexible GAFF-water will be used as for any other solute (old behavior).
Returns
-------
tleap_commands : str
The string of commands piped to tleap for building the prmtop
and inpcrd files. This will *already* have been run, but the
output can be useful for debugging or archival purposes. However,
this will reflect temporary file names for both input and output
file as these are used to avoid tleap filename restrictions.
Notes
-----
This can be easily broken if there are missing, duplicated, or
inconsistent ligand residue names in your box, mol2, and frcmod files.
You can use mdtraj to edit the residue names with something like
this: trj.top.residue(0).name = "L1"
"""
# Check for one residue name per mol2 file and uniqueness between all mol2 files
all_names = set()
for filename in mol2_filenames:
t = md.load(filename)
names = set([r.name for r in t.top.residues])
if len(names) != 1:
raise (ValueError(
"Must have a SINGLE residue name in each mol2 file."))
all_names = all_names.union(list(names))
if len(all_names) != len(mol2_filenames):
raise (ValueError("Must have UNIQUE residue names in each mol2 file."))
if len(mol2_filenames) != len(frcmod_filenames):
raise (ValueError(
"Must provide an equal number of frcmod and mol2 file names."))
#Get number of files
nfiles = len(mol2_filenames)
#Build absolute paths of input files so we can use context and temporary directory
infiles = mol2_filenames + frcmod_filenames + [box_filename]
infiles = [os.path.abspath(filenm) for filenm in infiles]
#Build absolute paths of output files so we can copy them back
prmtop_filename = os.path.abspath(prmtop_filename)
inpcrd_filename = os.path.abspath(inpcrd_filename)
#Use temporary directory and do the setup
with md.utils.enter_temp_directory():
all_names = [
md.load(filename).top.residue(0).name
for filename in mol2_filenames
]
mol2_section = "\n".join("%s = loadmol2 %s" % (all_names[k], filename)
for k, filename in enumerate(mol2_filenames))
amberparams_section = "\n".join(
"loadamberparams %s" % (filename)
for k, filename in enumerate(frcmod_filenames))
tleap_commands = TLEAP_TEMPLATE % dict(
mol2_section=mol2_section,
amberparams_section=amberparams_section,
box_filename=box_filename,
prmtop_filename=prmtop_filename,
inpcrd_filename=inpcrd_filename)
print(tleap_commands)
file_handle = open('tleap_commands', 'w')
file_handle.writelines(tleap_commands)
file_handle.close()
logger.debug('Running tleap in temporary directory.')
cmd = "tleap -f %s " % file_handle.name
logger.debug(cmd)
output = getoutput(cmd)
logger.debug(output)
return tleap_commands