def check_esp(conf_dict, qm_dir=".", logfile="check_esp.log"):
check_log = logger(logfile)
check_log.log("# Logfile QM ESP check")
for frag_i, conf_N in conf_dict.items():
check_log.log("Checking frag%s..." % frag_i)
mainpath = qm_dir + "/" + "frag%d" % frag_i
if not os.path.exists(mainpath):
check_log.log(" ! Warning ! %s not found." % mainpath)
continue
for conf_i in range(conf_N):
check_log.log("Checking frag%d-conf%d..." % (frag_i, conf_i))
fragname = 'frag%d-conf%d' % (frag_i, conf_i)
fragpath = mainpath + '/' + 'conf%d' % conf_i
logpath = fragpath + "/" + fragname + "_esp.log"
if not os.path.exists(fragpath):
check_log.log(" ! Warning ! %s not found." % fragpath)
continue
if not os.path.exists(logpath):
check_log.log(" ! Warning ! %s not found" % logpath)
continue
esp_vals_list = list()
esp_crds_list = list()
if get_esp(logpath, esp_vals_list, esp_crds_list):
n_crds = len(esp_crds_list)
n_vals = len(esp_vals_list)
if n_crds == n_vals:
check_log.log(" Found %d esp fit centers." % n_vals)
else:
check_log.log(
"! Warning ! Found %d fit centers but only %d fit values."
% (n_crds, n_vals))
if is_normal_terminate(logpath):
check_log.log(" Normal termination of Gaussian.")
else:
check_log.log(
" ! Warning ! An error occured. Check Gaussian log.")
check_log.close()
def datadump(database, dumpdir):
db = pickle.load(open(database, "rb"))
if os.path.exists(dumpdir):
raise Warning(
"Caution, %s already exists. Already existing data may be overwritten."
)
else:
os.mkdir(dumpdir)
os.mkdir(dumpdir + "/png")
frag2mol = db.get_frag2mol()
frag2lcapconn = db.get_frag2lcapconn()
frag2rcapconn = db.get_frag2rcapconn()
mol2frag = db.get_mol2frag()
mol2conn = db.get_mol2conn()
frag_log = logger(dumpdir + "/frag.dat")
frag_log.log("### datadump of database %s" % database)
frag_log.log("### timestamp %s" %
time.asctime(time.localtime(time.time())))
frag_log.log("### written by run_fragresp.py datadump routine.")
frag_log.log("###")
frag_log.log("### ----------------- ###")
frag_log.log("### FRAGMENT DATA LOG ###")
frag_log.log("### ----------------- ###")
frag_log.log("###")
frag_log.log(
"# id smiles mol_id lcap_id rcap_id Natoms Nbonds Nnonhatoms Chg Nhbd Nhba Nrotbonds Nrings"
)
for frag_i in range(db.get_frag_count()):
frag = db.get_frag(frag_i)
Chem.SanitizeMol(frag)
log_str = list()
### id
log_str.append(str(frag_i) + " ")
### smiles
log_str.append(str(Chem.MolToSmiles(frag, isomericSmiles=True)) + " ")
### mol_id
mol_count = len(frag2mol[frag_i])
if mol_count == 0:
log_str.append("-1 ")
else:
for i in range(mol_count):
mol_i = frag2mol[frag_i][i]
if i < mol_count - 1:
log_str.append(str(mol_i) + ",")
else:
log_str.append(str(mol_i) + " ")
### lcap_id
lcap_count = len(frag2lcapconn[frag_i])
if lcap_count == 0:
log_str.append("-1 ")
else:
for i in range(lcap_count):
cap_i = frag2lcapconn[frag_i][i]
if i < lcap_count - 1:
log_str.append(str(cap_i) + ",")
else:
log_str.append(str(cap_i) + " ")
### rcap_id
rcap_count = len(frag2rcapconn[frag_i])
if rcap_count == 0:
log_str.append("-1 ")
else:
for i in range(rcap_count):
cap_i = frag2rcapconn[frag_i][i]
if i < rcap_count - 1:
log_str.append(str(cap_i) + ",")
else:
log_str.append(str(cap_i) + " ")
### N_atoms
log_str.append(str(frag.GetNumAtoms()) + " ")
### N_bonds
log_str.append(str(frag.GetNumBonds()) + " ")
### Nnonhatoms
log_str.append(str(frag.GetNumHeavyAtoms()) + " ")
### Chg
log_str.append(str(rdmolops.GetFormalCharge(frag)) + " ")
### Nhbd
log_str.append(str(rdMolDescriptors.CalcNumHBD(frag)) + " ")
### Nhba
log_str.append(str(rdMolDescriptors.CalcNumHBA(frag)) + " ")
### Nrotbonds
log_str.append(str(rdMolDescriptors.CalcNumRotatableBonds(frag)) + " ")
### Nrings
log_str.append(str(rdMolDescriptors.CalcNumRings(frag)) + " ")
frag_log.log("".join(log_str))
png_path = dumpdir + "/png/" + "frag_%d.png" % frag_i
try:
Chem.SanitizeMol(frag)
AllChem.Compute2DCoords(frag)
Draw.MolToFile(frag, png_path, size=(500, 500))
except:
#Chem.Kekulize(frag)
print("Could not save frag %d to disk." % frag_i)
frag_log.close()
mol_log = logger(dumpdir + "/mol.dat")
mol_log.log("### datadump of database %s" % database)
mol_log.log("### timestamp %s" % time.asctime(time.localtime(time.time())))
mol_log.log("### written by run_fragresp.py datadump routine.")
mol_log.log("###")
mol_log.log("### ----------------- ###")
mol_log.log("### MOLECULE DATA LOG ###")
mol_log.log("### ----------------- ###")
mol_log.log("###")
mol_log.log(
"# id name smiles frag_id Natoms Nbonds Nnonhatoms Chg Nhbd Nhba Nrotbonds Nrings"
)
for mol_i in range(db.get_mol_count()):
mol = db.get_mol(mol_i)
Chem.SanitizeMol(mol)
name = db.get_name(mol_i)
decomp = db.get_decompose(mol_i)
log_str = list()
log_str.append(str(mol_i) + " ")
log_str.append(name + " ")
log_str.append(str(Chem.MolToSmiles(mol, isomericSmiles=True)) + " ")
frag_count = decomp.get_frag_count()
if frag_count == 0:
log_str.append("-1 ")
else:
for i in range(frag_count):
frag_i = mol2frag[mol_i][i]
if i < frag_count - 1:
log_str.append(str(frag_i) + ",")
else:
log_str.append(str(frag_i) + " ")
log_str.append(str(mol.GetNumAtoms()) + " ")
log_str.append(str(mol.GetNumBonds()) + " ")
log_str.append(str(mol.GetNumHeavyAtoms()) + " ")
log_str.append(str(rdmolops.GetFormalCharge(mol)) + " ")
log_str.append(str(rdMolDescriptors.CalcNumHBD(mol)) + " ")
log_str.append(str(rdMolDescriptors.CalcNumHBA(mol)) + " ")
log_str.append(str(rdMolDescriptors.CalcNumRotatableBonds(mol)) + " ")
log_str.append(str(rdMolDescriptors.CalcNumRings(mol)) + " ")
mol_log.log("".join(log_str))
png_path = dumpdir + "/png/" + "mol_%d.png" % mol_i
AllChem.Compute2DCoords(mol)
Chem.Kekulize(mol)
Draw.MolToFile(mol, png_path, size=(500, 500))
mol_log.close()
surr_log = logger(dumpdir + "/surr.dat")
surr_log.log("### datadump of database %s" % database)
surr_log.log("### timestamp %s" %
time.asctime(time.localtime(time.time())))
surr_log.log("### written by run_fragresp.py datadump routine.")
surr_log.log("###")
surr_log.log("### ----------------- ###")
surr_log.log("### SURROGATE DATA LOG ###")
surr_log.log("### ------------------ ###")
surr_log.log("###")
surr_log.log(
"# id name smiles mol_id Natoms Nbonds Nnonhatoms Chg Nhbd Nhba Nrotbonds Nrings"
)
for conn_i, conn in enumerate(db.get_conn_list()):
if conn.get_terminal():
continue
name = conn.get_name()
conn_cap = conn.get_surrogate_cap()
Chem.SanitizeMol(conn_cap)
log_str = list()
log_str.append(str(conn_i) + " ")
log_str.append(name + " ")
log_str.append(
str(Chem.MolToSmiles(conn_cap, isomericSmiles=True)) + " ")
conn2mol = db.get_conn2mol()[conn_i]
mol_count = len(conn2mol)
if mol_count == 0:
log_str.append("-1 ")
else:
for i in range(mol_count):
mol_i = conn2mol[i]
if i < mol_count - 1:
log_str.append(str(mol_i) + ",")
else:
log_str.append(str(mol_i) + " ")
log_str.append(str(conn_cap.GetNumAtoms()) + " ")
log_str.append(str(conn_cap.GetNumBonds()) + " ")
log_str.append(str(conn_cap.GetNumHeavyAtoms()) + " ")
log_str.append(str(rdmolops.GetFormalCharge(conn_cap)) + " ")
log_str.append(str(rdMolDescriptors.CalcNumHBD(conn_cap)) + " ")
log_str.append(str(rdMolDescriptors.CalcNumHBA(conn_cap)) + " ")
log_str.append(
str(rdMolDescriptors.CalcNumRotatableBonds(conn_cap)) + " ")
log_str.append(str(rdMolDescriptors.CalcNumRings(conn_cap)) + " ")
surr_log.log("".join(log_str))
png_path = dumpdir + "/png/" + "surr_%s.png" % (conn_i)
AllChem.Compute2DCoords(conn_cap)
Chem.Kekulize(conn_cap)
Draw.MolToFile(conn_cap, png_path, size=(500, 500))
surr_log.close()
def prep_qm(frag_list,
overwrite,
limit,
percentage,
write_mol2=True,
nproc=2,
mem=1900,
queue='marc2',
qm_dir=".",
opt_batch="submit_opt_g09.sh",
esp_batch="submit_esp_g09.sh",
logfile="prep_qm.log",
include_list=None):
queue_choices = ['marc2', 'slurm', 'condor', 'none']
if queue not in queue_choices:
print("queue %s not known. Using queue 'none'.")
queue = 'none'
conf_list = list()
if include_list == None:
include_list = range(len(frag_list))
prep_log = logger(logfile)
prep_log.log("### OpenEye conformer pipeline")
if not os.path.exists(qm_dir):
os.mkdir(qm_dir)
if queue == 'marc2':
g09_cmd = "subg09 -p %d -m %d -cc" % (nproc, mem)
elif queue == 'slurm':
g09_cmd = 'subg09_slurm.sh %d' % nproc
elif queue == 'condor':
g09_cmd = 'subg09_condor.sh 1'
else:
g09_cmd = 'g09'
opt_batch_file = logger(qm_dir + "/" + opt_batch)
esp_batch_file = logger(qm_dir + "/" + esp_batch)
opt_batch_file.log("#!/bin/bash")
opt_batch_file.log()
esp_batch_file.log("#!/bin/bash")
esp_batch_file.log()
for frag_i in include_list:
frag = frag_list[frag_i]
mainpath = qm_dir + "/" + "frag%d" % frag_i
if not os.path.exists(mainpath):
os.mkdir(mainpath)
elif not overwrite:
conf_list.append(-1)
continue
### Always isomericSmiles=True, otherwise omega does not
### know what to do.
smi = Chem.MolToSmiles(frag, isomericSmiles=True)
prep_log.log("Fragment %d" % frag_i)
prep_log.log("Smi %s" % smi)
mol = OEGraphMol()
OESmilesToMol(mol, smi)
mol = OEMol(mol)
generate_conformers(mol)
prep_log.log("initial conformers: %d" % mol.NumConfs())
filter_conformers(mol, limit, percentage)
prep_log.log("filtered conformers:%d" % mol.NumConfs())
conf_list.append(mol.NumConfs())
for conf_i, conf in enumerate(mol.GetConfs()):
fragname = 'frag%d-conf%d' % (frag_i, conf_i)
fragpath = mainpath + '/' + 'conf%d' % conf_i
subpath = "frag%d" % frag_i + '/' + 'conf%d' % conf_i
if not os.path.exists(fragpath):
os.mkdir(fragpath)
com_opt, com_esp = make_g09(fragname, conf, nproc, mem, queue)
f = logger(fragpath + "/" + fragname + "_opt.com")
f.log(com_opt)
f.close()
opt_batch_file.log(
"%s %s" % (g09_cmd, subpath + "/" + fragname + "_opt.com"))
del f
f = logger(fragpath + "/" + fragname + "_esp.com")
f.log(com_esp)
f.close()
esp_batch_file.log(
"%s %s" % (g09_cmd, subpath + "/" + fragname + "_esp.com"))
del f
if write_mol2:
ofs = oemolostream()
ofs.open(mainpath + "/" + "f%d.mol2" % frag_i)
OEWriteMolecule(ofs, mol)
del ofs
prep_log.log()
prep_log.close()
opt_batch_file.close()
esp_batch_file.close()
del opt_batch_file
del esp_batch_file
return conf_list
def check_opt(conf_dict,
mol_list,
qm_dir=".",
logfile="check_qm.log",
stdout=None,
stderr=None):
if stdout == None:
stdout = open(os.devnull, 'w')
if stderr == None:
stderr = open(os.devnull, 'w')
check_log = logger(logfile)
check_log.log("# Logfile QM check")
for frag_i, conf_N in conf_dict.items():
check_log.log("Checking frag%s..." % frag_i)
mainpath = qm_dir + "/" + "frag%d" % frag_i
ante_args_general = [
"-fi",
"gout",
"-pf",
"y",
"-at",
"sybyl",
"-dr",
"no",
]
if not os.path.exists(mainpath):
check_log.log(" ! Warning ! %s not found." % mainpath)
continue
for conf_i in range(conf_N):
check_log.log("Checking frag%d-conf%d..." % (frag_i, conf_i))
fragname = 'frag%d-conf%d' % (frag_i, conf_i)
fragpath = mainpath + '/' + 'conf%d' % conf_i
logpath = fragpath + "/" + fragname + "_opt.log"
if not os.path.exists(fragpath):
check_log.log(" ! Warning ! %s not found." % fragpath)
continue
if not os.path.exists(logpath):
check_log.log(" ! Warning ! %s not found" % logpath)
continue
freq_list = list()
if get_freq(logpath, freq_list):
freq_str = 'Frequencies [cm**-1]: '
freq_ok = True
for freq_i, freq in enumerate(freq_list):
if freq < 0.:
check_log.log(
" ! Warning ! Imaginary frequency %f cm**-1 for mode %d"
% (freq, freq_i))
freq_ok = False
freq_str += str(freq)
freq_str += ' '
if freq_ok:
check_log.log(" Frequencies OK!")
#check_log.log(freq_str)
else:
check_log.log(" ! Warning ! No frequencies found in %s" %
logpath)
del freq_list
if is_normal_terminate(logpath):
check_log.log(" Normal termination of Gaussian.")
else:
check_log.log(
" ! Warning ! An error occured. Check Gaussian log.")
mol2path = fragpath + "/" + fragname + "_opt.mol2"
ante_args = [
ante_exe, "-i", logpath, "-o", mol2path, "-fo", "mol2"
] + ante_args_general
call(ante_args, stdout=stdout, stderr=stderr)
mol_opt = Chem.MolFromMol2File(mol2path, removeHs=False)
matches = mol_opt.GetSubstructMatches(mol_list[frag_i])
if len(matches) > 1:
check_log.log(
" ! Warning ! Found more than one substructure match.")
if len(matches) == 0:
check_log.log(
" ! Warning ! Did not find any substructure match.")
check_log.close()
def prep_qm(frag_list,
overwrite,
limit,
percentage,
write_mol2=True,
nproc=2,
mem=1900,
queue='marc2',
qm_dir=".",
opt_batch="submit_opt_g09.sh",
esp_batch="submit_esp_g09.sh",
logfile="prep_qm.log",
include_list=None):
conf_list = list()
if include_list == None:
include_list = range(len(frag_list))
prep_log = logger(logfile)
prep_log.log("### Openbabel conformer pipeline")
if not os.path.exists(qm_dir):
os.mkdir(qm_dir)
if queue == 'marc2':
g09_cmd = "subg09 -p %d -m %d -cc" % (nproc, mem)
elif queue == 'slurm':
g09_cmd = 'subg09_slurm.sh %d' % nproc
elif queue == 'condor':
g09_cmd = 'subg09_condor.sh 1'
else:
g09_cmd = 'g09'
opt_batch_file = logger(qm_dir + "/" + opt_batch)
esp_batch_file = logger(qm_dir + "/" + esp_batch)
opt_batch_file.log("#!/bin/bash")
opt_batch_file.log()
opt_batch_file.log("_pwd=$PWD")
opt_batch_file.log()
esp_batch_file.log("#!/bin/bash")
esp_batch_file.log()
esp_batch_file.log("_pwd=$PWD")
esp_batch_file.log()
obConversion = ob.OBConversion()
obConversion.SetInFormat("smi")
obConversion.SetOutFormat("mol2")
for frag_i in include_list:
frag = frag_list[frag_i]
mainpath = qm_dir + "/" + "frag%d" % frag_i
if not os.path.exists(mainpath):
os.mkdir(mainpath)
elif not overwrite:
conf_list.append(-1)
continue
### Always isomericSmiles=True, otherwise omega does not
### know what to do.
smi = Chem.MolToSmiles(frag, isomericSmiles=True)
prep_log.log("Fragment %d" % frag_i)
prep_log.log("Smi %s" % smi)
pbmol = pb.readstring("smi", smi)
pbmol.make3D()
mol = pbmol.OBMol
generate_conformers(mol)
prep_log.log("initial conformers: %d" % mol.NumConformers())
filter_conformers(mol, limit, percentage)
prep_log.log("filtered conformers:%d" % mol.NumConformers())
conf_list.append(mol.NumConformers())
for conf_i in range(mol.NumConformers()):
mol.SetConformer(conf_i)
fragname = 'frag%d-conf%d' % (frag_i, conf_i)
fragpath = mainpath + '/' + 'conf%d' % conf_i
subpath = "frag%d" % frag_i + '/' + 'conf%d' % conf_i
if not os.path.exists(fragpath):
os.mkdir(fragpath)
com_opt, com_esp = make_g09(fragname, mol, nproc, mem, queue)
f = logger(fragpath + "/" + fragname + "_opt.com")
f.log(com_opt)
f.close()
opt_batch_file.log("cd %s" % subpath)
opt_batch_file.log("%s %s" % (g09_cmd, fragname + "_opt.com"))
opt_batch_file.log("cd $_pwd")
del f
f = logger(fragpath + "/" + fragname + "_esp.com")
f.log(com_esp)
f.close()
esp_batch_file.log("cd %s" % subpath)
esp_batch_file.log("%s %s" % (g09_cmd, fragname + "_esp.com"))
esp_batch_file.log("cd $_pwd")
del f
if write_mol2:
obConversion.WriteFile(mol, mainpath + "/" + "f%d.mol2" % frag_i)
prep_log.log()
prep_log.close()
opt_batch_file.close()
esp_batch_file.close()
del opt_batch_file
del esp_batch_file
return conf_list