def __init__(self,
template_name,
include=[],
nonprot_residues=[],
params={},
working_directory='tleap'):
utils.set_working_directory(working_directory)
enlighten_path = os.path.dirname(__import__(__name__).__file__)
tleap_module_path = os.path.join(enlighten_path, 'tleap')
template_path = os.path.join(tleap_module_path, template_name + '.in')
template_contents = None
if os.path.isfile(template_path):
with open(template_path) as f:
template_contents = f.read()
params['include'] = get_tleap_includes(include, nonprot_residues)
template_module = getattr(
__import__('tleap', fromlist=[template_name]), template_name)
params['pdb'].to_filename('input.pdb')
params['water_pdb'].to_filename('water.pdb')
with open('tleap.in', 'w') as f:
f.write(template_module.run(params, template_contents))
utils.run_in_shell('tleap -f tleap.in', 'tleap.log')
try:
template_module.check(params)
except AttributeError:
pass
def __init__(self,
pdb,
name,
charge=0,
working_directory="antechamber",
create_frcmod=True):
amberhome = get_amberhome()
utils.set_working_directory(working_directory)
pdb.to_filename('ligand.pdb')
antechamber_command = (
amberhome + "/bin/antechamber " +
"-i ligand.pdb -fi pdb -o {name}.prepc "
"-fo prepc -rn {name} -c bcc -nc {charge}".format(name=name,
charge=charge))
utils.run_in_shell(antechamber_command, 'antechamber.out')
utils.check_file(
name + '.prepc',
"Antechamber failed to generate {name}.prepc file".format(
name=name))
self.working_directory = os.getcwd()
if create_frcmod:
parmck_command = (
amberhome + "/bin/parmchk2 " +
"-i {name}.prepc -f prepc -o {name}.frcmod".format(name=name))
utils.run_in_shell(parmck_command, 'parmchk2.out')
# TODO: check for ATTN warnings
os.chdir('..')
def __init__(self, pdb, ph=7.0, ph_offset=0.7, working_directory="propka"):
utils.set_working_directory(working_directory)
pdb.to_filename('input.pdb')
utils.run_in_shell("propka31 input.pdb", "propka31.out")
with open('input.pka') as f:
propka_results = parse_propka_output(f)
self.pdb = pdb.copy()
residues = self.pdb.residues()
self.prot_pka = ph + ph_offset
self.deprot_pka = ph - ph_offset
self.prot_list = []
self.deprot_list = []
for hash, pka_entry in propka_results.items():
if hash not in residues:
continue
if prot_residue(pka_entry, self.prot_pka):
pdb_utils.modify_atoms(residues[hash], 'resName',
PROT_DICT[pka_entry['resName']])
self.prot_list.append(pka_entry)
if deprot_residue(pka_entry, self.deprot_pka):
pdb_utils.modify_atoms(residues[hash], 'resName',
DEPROT_DICT[pka_entry['resName']])
self.deprot_list.append(pka_entry)
# Need to remove hydrogens added by reduce on deprotonated
# residues - else top-file creation will fail.
self.pdb.remove_atom(
pdb_utils.find_atom(residues[hash],
lambda atom: 'new' in atom['extras']))
PRINT_PKA_FORMAT = "{resName:>6}{resSeq:>4}{chainID:>2}{pKa:>9.2f}"
if len(self.prot_list) > 0:
print("The following ASP/GLU residues have predicted pKa's above "
"{:4.2f} and will be protonated (on OD2/OE2):".format(
self.prot_pka))
print(" pKa")
for pka_entry in self.prot_list:
print(PRINT_PKA_FORMAT.format(**pka_entry))
if len(self.deprot_list) > 0:
print("The following CYS/LYS residues have predicted pKa's below "
"{:4.2f} and will be deprotonated:".format(self.deprot_pka))
print(" pKa")
for pka_entry in self.deprot_list:
print(PRINT_PKA_FORMAT.format(**pka_entry))
os.chdir('..')
def __init__(self, pdb, working_directory="pdb4amber_reduce"):
amberhome = get_amberhome()
utils.set_working_directory(working_directory)
pdb.to_filename('input.pdb')
pdb4amber_command = (amberhome + "/bin/pdb4amber "
"-i input.pdb -o pdb4amber.pdb --nohyd --dry")
utils.run_in_shell(pdb4amber_command, 'pdb4amber.out')
reduce_command = (amberhome + "/bin/reduce "
"-build -nuclear pdb4amber.pdb")
utils.run_in_shell(reduce_command, 'reduce.pdb')
with open('reduce.pdb') as f:
self.pdb = pdb_utils.Pdb(f)
renamed_histidines = get_renamed_histidines(self.pdb)
residues = self.pdb.residues()
for res_hash, res_name in renamed_histidines.items():
pdb_utils.modify_atoms(residues.get(res_hash, []), 'resName',
res_name)
# Remove hydrogens on HETATMs
self.pdb.atoms = [
atom for atom in self.pdb.atoms
if (atom['record'] != 'HETATM' or 'new' not in atom['extras'])
]
# Remove hydrogens added by reduce to non-protein residues
with open('pdb4amber_nonprot.pdb') as f:
self.nonprotPdb = pdb_utils.Pdb(f)
self.nonprot_residues = set(atom['resName']
for atom in self.nonprotPdb.atoms)
self.pdb.atoms = [
atom for atom in self.pdb.atoms
if (atom['resName'] not in self.nonprot_residues
or 'new' not in atom['extras'])
]
# store crystalline waters
with open('pdb4amber_water.pdb') as f:
self.waterPdb = pdb_utils.Pdb(f)
os.chdir('..')
# # TEMP (TIRA): For fast debugging and testing
# docs_train = docs_train[:100]
# docs_test = docs_test[:100]
# y_train = y_train[:100]
# author_ids_test = author_ids_test[:100]
X_train, X_test, clf, feature_names = extract_features(
docs_train, docs_test, preset_key)
cross_validate_model(clf, X_train, y_train)
train_model_and_predict(clf, X_train, y_train, X_test, author_ids_test,
preset_key, True,
prediction_xmls_destination_main_directory)
# Log run time
print("@ %.2f seconds: Run finished\n", time.process_time())
'''
The following lines will be executed only if this .py file is run as a script,
and not if it is imported as a module.
• __name__ is one of the import-related module attributes, which holds the name of the module.
• A module's __name__ is set to "__main__" when it is running in
the main scope (the scope in which top-level code executes).
'''
if __name__ == "__main__":
logger = utils.configure_root_logger()
utils.set_working_directory()
main_development()
print('executed successfully LR')
# main_tira_evaluation()
}
}
if args.params is not None:
params = utils.merge_dicts_of_dicts(params, json.load(args.params))
args.params.close()
print("Starting PREP protocol in {}/".format(pdb_name))
if os.path.exists(pdb_name):
print(
"It appears you've already (attempted to) run prep.py with {0}. "
"Delete folder {0} or rename pdb if you want to run it again.".format(
pdb_name))
sys.exit()
utils.set_working_directory(pdb_name)
pdb = pdb_utils.Pdb(args.pdb)
ligands = pdb.get_residues_by_name(ligand_name)
if len(ligands) == 0:
raise ValueError("No ligands found")
ligand_index = params['antechamber']['ligand_index']
if ligand_index > len(ligands):
raise ValueError("ligand_index is larger than the number of ligands")
ligand_atoms = ligands[ligand_index - 1]
if len(ligands) > 1:
print("More than one ligand detected. Using coordinates from the ligand"
" with chainID={} and resSeq={}".format(ligand_atoms[0]['chainID'],