def test_init_from_and_to_stream(self):
mols = makeStereoExamples() * 10
holder = rdSubstructLibrary.CachedSmilesMolHolder()
# one day I'll fix this, but we need to write text but read binary
# grrr.... something about the python_streambuf handler.
slib = rdSubstructLibrary.SubstructLibrary(holder, None)
for mol in mols:
holder.AddSmiles(Chem.MolToSmiles(mol, isomericSmiles=True))
if rdSubstructLibrary.SubstructLibraryCanSerialize():
fd, path = tempfile.mkstemp()
with open(path, 'w') as file:
slib.ToStream(file)
with open(path, 'rb') as file:
slib2 = rdSubstructLibrary.SubstructLibrary()
slib2.InitFromStream(file)
self.assertEqual(len(slib), len(slib2))
from io import StringIO, BytesIO
s = StringIO()
slib.ToStream(s)
sb = BytesIO(s.getvalue().encode("ascii"))
self.assertTrue(len(sb.getvalue()) > 0)
slib3 = rdSubstructLibrary.SubstructLibrary()
slib3.InitFromStream(sb)
self.assertEqual(len(slib), len(slib2))
def test0SubstructLibrary(self):
for fpholderCls in [None, rdSubstructLibrary.PatternHolder]:
for holder in [
rdSubstructLibrary.MolHolder(),
rdSubstructLibrary.CachedMolHolder(),
rdSubstructLibrary.CachedSmilesMolHolder()
]:
if fpholderCls: fpholder = fpholderCls()
else: fpholder = None
slib = rdSubstructLibrary.SubstructLibrary(holder, fpholder)
for i in range(100):
m = Chem.MolFromSmiles("c1ccccc1")
self.assertEqual(slib.AddMol(m), i)
res = slib.GetMatches(m)
t2 = time.time()
self.assertTrue(len(res) == 100)
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
self.assertTrue(set(res) == set(list(range(100))))
res = slib.GetMatches(m, maxResults=100)
self.assertEqual(len(res), 100)
self.assertEqual(
len(slib.GetMatches(m, startIdx=0, endIdx=100)), 100)
self.assertTrue(slib.HasMatch(m))
self.assertEqual(slib.CountMatches(m), 100)
def test0SubstructLibrary(self):
for fpholderCls in [None, rdSubstructLibrary.PatternHolder]:
for holder in [rdSubstructLibrary.MolHolder(), rdSubstructLibrary.CachedMolHolder(),
rdSubstructLibrary.CachedSmilesMolHolder()]:
if fpholderCls: fpholder = fpholderCls()
else: fpholder = None
slib_ = rdSubstructLibrary.SubstructLibrary(holder, fpholder)
for i in range(100):
m = Chem.MolFromSmiles("c1ccccc1")
self.assertEqual(slib_.AddMol(m), i)
libs = [slib_]
if rdSubstructLibrary.SubstructLibraryCanSerialize():
serialized1 = pickle.loads(pickle.dumps(slib_))
serialized2 = rdSubstructLibrary.SubstructLibrary(slib_.Serialize())
libs.append(serialized1)
libs.append(serialized2)
for slib in libs:
res = slib.GetMatches(m)
t2 = time.time()
self.assertTrue(len(res) == 100)
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
self.assertTrue(set(res) == set(list(range(100))))
res = slib.GetMatches(m, maxResults=100);
self.assertEqual(len(res), 100)
self.assertEqual(len(slib.GetMatches(m, startIdx=0, endIdx=100)), 100)
self.assertTrue(slib.HasMatch(m))
self.assertEqual(slib.CountMatches(m), 100)
def testSmilesCache(self):
mols = makeStereoExamples() * 10
holder = rdSubstructLibrary.CachedSmilesMolHolder()
slib_ = rdSubstructLibrary.SubstructLibrary(holder, None)
for mol in mols:
holder.AddSmiles(Chem.MolToSmiles(mol, isomericSmiles=True))
libs = [slib_]
if rdSubstructLibrary.SubstructLibraryCanSerialize():
serialized1 = pickle.loads(pickle.dumps(slib_))
serialized2 = rdSubstructLibrary.SubstructLibrary(
slib_.Serialize())
libs.append(serialized1)
libs.append(serialized2)
for slib in libs:
core = Chem.MolFromSmarts("C-1-C-C-O-C(-*)(-*)1")
res = slib.GetMatches(core)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=True)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-C(-[O])(-[N])1")
core.SetProp("core", "core")
res = slib.GetMatches(core, useChirality=False)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=False)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-[C@@](-[O])(-[N])1")
res = slib.GetMatches(core, useChirality=False)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=False)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-[C@@](-[O])(-[N])1")
res = slib.GetMatches(core)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=True)
]))
def test_basic_addpatterns(self):
# add mols
pdb_ligands = [
"CCS(=O)(=O)c1ccc(OC)c(Nc2ncc(-c3cccc(-c4ccccn4)c3)o2)c1",
"COc1ccc(S(=O)(=O)NCC2CC2)cc1Nc1ncc(-c2cccc(-c3cccnc3)c2)o1",
"COc1ccc(-c2oc3ncnc(N)c3c2-c2ccc(NC(=O)Nc3cc(C(F)(F)F)ccc3F)cc2)cc1",
"COC(=O)Nc1nc2ccc(Oc3ccc(NC(=O)Nc4cc(C(F)(F)F)ccc4F)cc3)cc2[nH]1",
"COc1cc(Nc2ncnc(-c3cccnc3Nc3ccccc3)n2)cc(OC)c1OC",
"O=C(Nc1ccc(Oc2ccccc2)cc1)c1cccnc1NCc1ccncc1",
"O=C(Nc1ccc(Oc2ccccc2)cc1)c1cccnc1NCc1ccncc1",
"CNC(=O)c1cc(Oc2ccc3[nH]c(Nc4ccc(Cl)c(C(F)(F)F)c4)nc3c2)ccn1",
"CNC(=O)c1cc(Oc2ccc3oc(Nc4ccc(Cl)c(OCC5CCC[NH+]5C)c4)nc3c2)ccn1",
"CNC(=O)c1cc(Oc2ccc3oc(Nc4ccc(Cl)c(OCC5CCC[NH+]5C)c4)nc3c2)ccn1",
"COc1cc2nccc(Oc3ccc4c(c3)OCCN4C(=O)Nc3ccc(Cl)cc3)c2cc1OC",
"CNC(=O)c1c(C)oc2cc(Oc3cc[nH+]c4cc(OCCN5CCOCC5)ccc34)ccc12",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)Nc5ccc(Cl)cc5)cccc4c3)c2cc1OC",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)Nc5ccc(Cl)cc5)cccc4c3)c2cc1OC",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)NC5CC5)cccc4c3)c2cc1OC",
"COc1cc2[nH+]ccc(Oc3ccc4c(C(=O)NC5CC5)cccc4c3)c2cc1OC",
"Cc1ccc(C(=O)Nc2cc(CCC[NH+](C)C)cc(C(F)(F)F)c2)cc1Nc1ncccc1-c1ccncn1",
"COc1cc(Nc2nccc(Nc3ccc4c(C)n[nH]c4c3)n2)cc(OC)c1OC",
"COc1cc(Nc2nccc(N(C)c3ccc4c(C)n[nH]c4c3)n2)cc(OC)c1OC",
"Cc1ccn(-c2ccc3c(c2)NCC3(C)C)c(=O)c1-c1ccc2nc(N)ncc2c1",
"Cc1ccn(-c2ccc3c(c2)NCC3(C)C)c(=O)c1-c1ccc2nc(N)ncc2c1",
"Cc1ccc(C(=O)NCCC2CCCC2)cc1C(=O)Nc1ccc(N)nc1",
"Cc1ccc(C(=O)NCCC2CCCC2)cc1C(=O)Nc1ccc(N)nc1",
"Cc1ccn(-c2cccc(C(F)(F)F)c2)c(=O)c1-c1ccc2nc(N)ncc2c1",
"Cc1ccn(-c2cccc(C(F)(F)F)c2)c(=O)c1-c1ccc2nc(N)ncc2c1",
"O=C(Nc1cncnc1)c1c(Cl)ccc2c(Nc3cccc(C(F)(F)F)c3)noc12",
"O=C(Nc1cncnc1)c1c(Cl)ccc2c(Nc3cccc(C(F)(F)F)c3)noc12",
"CC1(C)CNc2cc(NC(=O)c3cccnc3NCc3ccncc3)ccc21",
"CC1(C)CNc2cc(NC(=O)c3cccnc3NCc3ccncc3)ccc21"
]
for holder in [
rdSubstructLibrary.CachedSmilesMolHolder(),
rdSubstructLibrary.CachedTrustedSmilesMolHolder()
]:
for smi in pdb_ligands:
holder.AddSmiles(smi)
for patttern in [
None,
rdSubstructLibrary.PatternHolder(),
rdSubstructLibrary.TautomerPatternHolder()
]:
lib = rdSubstructLibrary.SubstructLibrary(holder)
rdSubstructLibrary.AddPatterns(lib, numThreads=-1)
self.assertEqual(len(lib.GetMolHolder()),
len(lib.GetFpHolder()))
for smi in pdb_ligands:
self.assertTrue(lib.CountMatches(Chem.MolFromSmiles(smi)))
def testSmilesCache(self):
mols = makeStereoExamples() * 10
holder = rdSubstructLibrary.CachedSmilesMolHolder()
slib = rdSubstructLibrary.SubstructLibrary(holder, None)
for mol in mols:
holder.AddSmiles(Chem.MolToSmiles(mol, isomericSmiles=True))
core = Chem.MolFromSmarts("C-1-C-C-O-C(-*)(-*)1")
res = slib.GetMatches(core)
self.assertEqual(
len(res),
len([
x for x in mols if x.HasSubstructMatch(core, useChirality=True)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-C(-[O])(-[N])1")
core.SetProp("core", "core")
res = slib.GetMatches(core, useChirality=False)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=False)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-[C@@](-[O])(-[N])1")
res = slib.GetMatches(core, useChirality=False)
self.assertEqual(
len(res),
len([
x for x in mols
if x.HasSubstructMatch(core, useChirality=False)
]))
core = Chem.MolFromSmarts("C-1-C-C-O-[C@@](-[O])(-[N])1")
res = slib.GetMatches(core)
self.assertEqual(
len(res),
len([
x for x in mols if x.HasSubstructMatch(core, useChirality=True)
]))
def test1SubstructLibrary(self):
for fpholderCls in [None, rdSubstructLibrary.PatternHolder]:
for holder in [
rdSubstructLibrary.MolHolder(),
rdSubstructLibrary.CachedMolHolder(),
rdSubstructLibrary.CachedSmilesMolHolder()
]:
if fpholderCls: fpholder = fpholderCls()
else: fpholder = None
slib = rdSubstructLibrary.SubstructLibrary(holder, fpholder)
mols = []
for i in range(100):
m = Chem.MolFromSmiles("c1ccccc1")
self.assertEqual(slib.AddMol(m), i * 2)
mols.append(m)
m2 = Chem.MolFromSmiles("CCCC")
self.assertEqual(slib.AddMol(m2), i * 2 + 1)
mols.append(m2)
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
self.assertEqual(set(res), set(list(range(0, 200, 2))))
res = slib.GetMatches(m2)
self.assertEqual(len(res), 100)
self.assertTrue(set(res) == set(list(range(1, 200, 2))))
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
res = slib.GetMatches(m, maxResults=100)
self.assertEqual(len(res), 100)
self.assertEqual(
len(slib.GetMatches(m, startIdx=0, endIdx=50 * 2)), 50)
self.assertEqual(
len(slib.GetMatches(m2, startIdx=1, endIdx=50 * 2 + 1)),
50)
self.assertTrue(slib.HasMatch(m))
self.assertTrue(slib.HasMatch(m2))
self.assertEqual(slib.CountMatches(m), 100)
self.assertEqual(slib.CountMatches(m2), 100)
def test1SubstructLibrary(self):
for keyholderCls in [None, rdSubstructLibrary.KeyFromPropHolder]:
for fpholderCls in [None, rdSubstructLibrary.PatternHolder]:
for holder in [
rdSubstructLibrary.MolHolder(),
rdSubstructLibrary.CachedMolHolder(),
rdSubstructLibrary.CachedSmilesMolHolder()
]:
if fpholderCls: fpholder = fpholderCls()
else: fpholder = None
if keyholderCls:
keyholder = keyholderCls()
self.assertEqual(keyholder.GetPropName(), "_Name")
else:
keyholder = None
slib_ = rdSubstructLibrary.SubstructLibrary(
holder, fpholder, keyholder)
mols = []
for i in range(100):
m = Chem.MolFromSmiles("c1ccccc1")
m.SetProp("_Name", str(i * 2))
self.assertEqual(slib_.AddMol(m), i * 2)
mols.append(m)
m2 = Chem.MolFromSmiles("CCCC")
m2.SetProp("_Name", str(i * 2 + 1))
self.assertEqual(slib_.AddMol(m2), i * 2 + 1)
mols.append(m2)
libs = [slib_]
if rdSubstructLibrary.SubstructLibraryCanSerialize():
serialized1 = pickle.loads(pickle.dumps(slib_))
serialized2 = rdSubstructLibrary.SubstructLibrary(
slib_.Serialize())
libs.append(serialized1)
libs.append(serialized2)
for slib in libs:
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
self.assertEqual(set(res), set(list(range(0, 200, 2))))
if keyholderCls:
self.assertEqual(
[str(idx) for idx in res],
[str(idx) for idx in range(0, 200, 2)])
res = slib.GetMatches(m2)
self.assertEqual(len(res), 100)
self.assertTrue(
set(res) == set(list(range(1, 200, 2))))
if keyholderCls:
self.assertEqual(
[str(idx) for idx in res],
[str(idx) for idx in range(1, 200, 2)])
res = slib.GetMatches(m)
self.assertEqual(len(res), 100)
res = slib.GetMatches(m, maxResults=100)
self.assertEqual(len(res), 100)
self.assertEqual(
len(slib.GetMatches(m, startIdx=0, endIdx=50 * 2)),
50)
self.assertEqual(
len(
slib.GetMatches(m2,
startIdx=1,
endIdx=50 * 2 + 1)), 50)
self.assertTrue(slib.HasMatch(m))
self.assertTrue(slib.HasMatch(m2))
self.assertEqual(slib.CountMatches(m), 100)
self.assertEqual(slib.CountMatches(m2), 100)
