def test_rsp_num_rings():
test_smiles = 'O=C(c1c2ccccc2cc2ccccc12)N1CCC(N2CCC[C@@H](C(=O)N3CCOCC3)C2)CC1'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=C(c1c2ccccc2cc2ccccc12)N1CCC(N2CCCCC2)CC1')
max_result = canonize_smiles(
'O=C(c1cccc2ccccc12)N1CCC(N2CCC[C@@H](C(=O)N3CCOCC3)C2)CC1')
_test_rule_min_max(test_mol, RSPNumRings, min_result, max_result)
def test_scp_delta():
test_smiles = 'OC5CC31CC5CC1C4CCc2occc2C4CC3'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles(
'OC1CCC2(CCC3c4ccoc4CCC3C2)C1') # retain spiro
max_result = canonize_smiles('OC1CC23CCC4C=CCCC4C2CC1C3') # retain bridged
_test_rule_min_max(test_mol, SCPDelta, min_result, max_result)
def test_scp_num_linker_bonds():
test_smiles = 'O=C(NCCCCN1CCN(c2ccccc2)CC1)c1ccc2c(c1)Cc1ccccc1-2'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=C(NCCCCN1CCNCC1)c1ccc2c(c1)Cc1ccccc1-2')
max_result = canonize_smiles(
'O=C(NCCCCN1CCN(c2ccccc2)CC1)c1ccc2c(c1)CC=C2')
_test_rule_min_max(test_mol, SCPNumLinkerBonds, min_result, max_result)
def test_rrp_het_atom_linked():
test_smiles = 'O=C(NCc1ccccc1)N1CCN2C(=O)OC(c3ccccc3)(c3ccccc3)[C@@H]2C1'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles(
'O=C(NCc1ccccc1)N1CCN2C(=O)OC(c3ccccc3)[C@@H]2C1')
max_result = canonize_smiles('O=C1OC(c2ccccc2)(c2ccccc2)[C@@H]2CNCCN12')
_test_rule_min_max(test_mol, RRPHetAtomLinked, min_result, max_result)
def get_molecules_for_scaffold(self,
scaffold_smiles,
data=False,
default=None):
"""Return a list of molecule IDs which are represented by a scaffold in the graph.
Note: This is determined by traversing the graph. In the case of a scaffold tree
the results represent the rules used to prioritize the scaffolds.
Parameters
----------
scaffold_smiles : (string) SMILES of query scaffold.
data : (string or bool, optional (default=False)) The molecule node attribute returned in 2-tuple
(n, dict[data]). If True, return entire molecule node attribute dict as (n, dict).
If False, return just the molecule nodes n.
default : (value, optional (default=None)) Value used for molecule nodes that don’t have the
requested attribute. Only relevant if data is not True or False.
"""
molecules = []
if scaffold_smiles not in self:
scaffold_smiles = canonize_smiles(scaffold_smiles, failsafe=True)
if scaffold_smiles not in self:
return molecules
for succ in nx.bfs_tree(self, scaffold_smiles, reverse=False):
if self.nodes[succ].get('type') == 'molecule':
if data is False:
molecules.append(succ)
elif data is True:
molecules.append((succ, self.nodes[succ]))
else:
molecules.append(
(succ, self.nodes[succ].get(data, default)))
return molecules
def _subgraph_from_scf(self, scaffold, traversal):
"""Private: Select a subgraph starting at a scaffold node.
Parameters
----------
scaffold : str
Scaffold node identifier.
traversal : str {'parent', 'child', 'bidirectional'}
The direction of traversal to create the subgraph.
If 'bidirectional' both directions are considered.
Returns
-------
subgraph : ScaffoldGraph
A subgraph starting at `scaffold`.
"""
G = self._graph
query = canonize_smiles(scaffold)
if not G.scaffold_in_graph(query):
raise ValueError(f'scaffold: {query} not in graph {G}')
if traversal == 'parent':
nodes = G.get_parent_scaffolds(query)
elif traversal == 'child':
nodes = list(nx.descendants(G, query))
elif traversal == 'bidirectional':
nodes = G.get_parent_scaffolds(query)
nodes += list(nx.descendants(G, query))
else:
msg = 'traversal must be one of {child, parent, bidirectional}'
raise ValueError(msg)
subgraph = G.subgraph([query] + nodes)
return subgraph
def _get_scaffold_hierarchy(self,
scaffold_smiles,
data=False,
default=None,
max_levels=-1,
traversal='parent'):
"""Private: Return a list of parent/child scaffolds for a query scaffold.
Parameters
----------
scaffold_smiles : str
SMILES of query scaffold.
data : str, bool, optional
The scaffold node attribute returned in 2-tuple (n, ddict[data]).
If True, return entire node attribute dict as (n, ddict).
If False, return just the nodes n. The default is False.
default : value, bool, optional
Value used for nodes that don't have the requested attribute.
Only relevant if data is not True or False.
max_levels : int, optional
If > 0 only return scaffolds with a hierarchy difference to the
query scaffold of `max_levels`.
traversal : {'parent', 'child'}, optional
Direction of traversal.
Returns
-------
list
A list of scaffold parent/child nodes.
"""
assert traversal in {'parent', 'child'}
reverse = traversal == 'parent'
next_hiers = []
if scaffold_smiles not in self:
scaffold_smiles = canonize_smiles(scaffold_smiles, failsafe=True)
if scaffold_smiles not in self:
return next_hiers
level = self.nodes[scaffold_smiles].get('hierarchy', float('inf'))
bfs = iter(nx.bfs_tree(self, scaffold_smiles, reverse=reverse).nodes)
next(bfs) # first entry is the query node
for succ in bfs:
d = self.nodes[succ]
if d.get('type') == 'scaffold' and (
max_levels < 0
or level - d.get('hierarchy', 0) <= max_levels):
if data is False:
next_hiers.append(succ)
elif data is True:
next_hiers.append((succ, self.nodes[succ]))
else:
next_hiers.append(
(succ, self.nodes[succ].get(data, default)))
return next_hiers
def scaffold_in_graph(self, scaffold_smiles):
"""Returns True if specified scaffold SMILES is in the scaffold graph
Parameters
----------
scaffold_smiles : (str) SMILES of query scaffold.
"""
result = scaffold_smiles in self
if result is not True:
scaffold_smiles = canonize_smiles(scaffold_smiles, failsafe=True)
result = scaffold_smiles in self
return result
def get_molecules_for_scaffold(self, scaffold_smiles):
"""Return a list of molecule IDs which are represented by a scaffold in the graph.
Note: This is determined by traversing the graph. In the case of a scaffold tree
the results represent the rules used to prioritize the scaffolds.
Parameters
----------
scaffold_smiles : (str) SMILES of query scaffold.
"""
molecules = []
if scaffold_smiles not in self:
scaffold_smiles = canonize_smiles(scaffold_smiles, failsafe=True)
if scaffold_smiles not in self:
return molecules
for succ in nx.bfs_tree(self, scaffold_smiles, reverse=False):
if self.nodes[succ]['type'] == 'molecule':
molecules.append(succ)
return molecules
def get_child_scaffolds(self,
scaffold_smiles,
data=False,
default=None,
max_levels=-1):
"""Return a list of child scaffolds for a query scaffold.
Parameters
----------
scaffold_smiles : (string) SMILES of query scaffold.
data : (string or bool, optional (default=False)) The scaffold node attribute returned in 2-tuple
(n, dict[data]). If True, return entire scaffold node attribute dict as (n, dict).
If False, return just the scaffold nodes n.
default : (value, optional (default=None)) Value used for scaffold nodes that don’t have the
requested attribute. Only relevant if data is not True or False.
max_levels : (integer, optional (default=-1)) If > 0 only return scaffolds with a hierarchy
difference to the query scaffold of max_levels.
"""
children = []
if scaffold_smiles not in self:
scaffold_smiles = canonize_smiles(scaffold_smiles, failsafe=True)
if scaffold_smiles not in self:
return children
level = self.nodes[scaffold_smiles].get('hierarchy', float('inf'))
bfs = iter(nx.bfs_tree(self, scaffold_smiles, reverse=False).nodes)
next(bfs) # first entry is the query node
for succ in bfs:
d = self.nodes[succ]
if d.get('type') == 'scaffold' and (
max_levels < 0
or d.get('hierarchy', 0) - level <= max_levels):
if data is False:
children.append(succ)
elif data is True:
children.append((succ, self.nodes[succ]))
else:
children.append((succ, self.nodes[succ].get(data,
default)))
return children
def scaffold_in_graph(self, scaffold_smiles):
"""Returns True if the specified scaffold SMILES is in the scaffold graph.
Parameters
----------
scaffold_smiles : str
SMILES of query scaffold.
Returns
-------
bool
True if the scaffold is in the graph.
Notes
-----
If not initially found the SMILES is canonized and the graph is searched
with the canonized SMILES key.
"""
result = scaffold_smiles in self
if result is not True:
scaffold_smiles = canonize_smiles(scaffold_smiles, failsafe=True)
result = scaffold_smiles in self
return result and self.nodes[scaffold_smiles]['type'] == 'scaffold'
def test_scp_num_sulphur_atoms():
test_smiles = 'c1csc2c(NCN3CCN(CCc4ccccc4)CC3)ncnc12'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('c1ccc(CCN2CCN(CNc3ccncn3)CC2)cc1')
max_result = canonize_smiles('c1nc(NCN2CCNCC2)c2sccc2n1')
_test_rule_min_max(test_mol, SCPNumSAtoms, min_result, max_result)
def test_rsp_num_sulphur_atoms():
test_smiles = 'c1nc2ccc3nc(NC(=O)C(c4ccccc4)c4ccccc4)sc3c2s1'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=C(Cc1ccccc1)Nc1nc2ccc3ncsc3c2s1')
max_result = canonize_smiles('O=C(Nc1nc2ccccc2s1)C(c1ccccc1)c1ccccc1')
_test_rule_min_max(test_mol, RSPNumSAtoms, min_result, max_result)
def test_rrp_linker_length():
test_smiles = 'O=C1c2ccccc2-c2c(NCCc3ccccc3)c(=O)[nH]c3cccc1c23'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=C1C=Cc2c(NCCc3ccccc3)c(=O)[nH]c3cccc1c23')
max_result = canonize_smiles('O=C1c2ccccc2-c2cc(=O)[nH]c3cccc1c23')
_test_rule_min_max(test_mol, RRPLinkerLength, min_result, max_result)
def test_rrp_ring_size():
test_smiles = 'n1nc(-c2ccccc2)nc1=S'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('c1ccccc1')
max_result = canonize_smiles('S=C1N=CN=N1')
_test_rule_min_max(test_mol, RRPRingSize, min_result, max_result)
def test_scp_num_oxygen_atoms():
test_smiles = 'C2Oc3ccccc3C(=O)C2=O'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('c1ccccc1')
max_result = canonize_smiles('O=C1C=COCC1=O')
_test_rule_min_max(test_mol, SCPNumOAtoms, min_result, max_result)
def test_scp_num_nitrogen_atoms():
test_smiles = 'N1CCCC(OC(=O)C(O)(c2ccccc2)c2ccccc2)C1'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('OC(c1ccccc1)c1ccccc1')
max_result = canonize_smiles('O=C(OC1CCCNC1)C(O)c1ccccc1')
_test_rule_min_max(test_mol, SCPNumNAtoms, min_result, max_result)
def test_scp_abs_delta():
test_smiles = 'C1CC2CN3C(CC=CC3=O)C4C2N(C1)CCC4'
test_mol = Chem.MolFromSmiles(test_smiles)
max_result = canonize_smiles('C1CC2CNCC3CCCN(C1)C23')
_test_rule(test_mol, SCPAbsDelta('max'), max_result)
def test_scp_num_aromatic_rings():
test_smiles = 'N1CCCC(OC(=O)C(O)(c2ccccc2)c2ccccc2)C1'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=C(OC1CCCNC1)C(O)c1ccccc1')
max_result = canonize_smiles('OC(c1ccccc1)c1ccccc1')
_test_rule_min_max(test_mol, SCPNumAromaticRings, min_result, max_result)
def test_rrp_num_nitrogen_atoms():
test_smiles = 'c1cccc2c(=O)[nH][nH]c(=O)c12'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=c1ccc(=O)[nH][nH]1')
max_result = canonize_smiles('c1ccccc1')
_test_rule_min_max(test_mol, RRPNumNAtoms, min_result, max_result)
def test_rrp_num_oxygen_atoms():
test_smiles = 'C1OC(=O)C2=C1CCC=C2'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=C1C=CCO1')
max_result = canonize_smiles('C1=CCCC=C1')
_test_rule_min_max(test_mol, RRPNumOAtoms, min_result, max_result)
def test_rrp_num_sulphur_atoms():
test_smiles = 'C1CSC(=NNC(=O)C(=O)CC2CCOCC2)N1'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('N=C1NCCS1')
max_result = canonize_smiles('C1CCOCC1')
_test_rule_min_max(test_mol, RRPNumSAtoms, min_result, max_result)
def test_rsp_abs_delta():
test_smiles = 'O=C1N(CCCC3CCNCC3)CCC12CCN1CCCC12'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=C1NCCC12CCN1CCCC12')
max_result = canonize_smiles('O=C1N(CCCC2CCNCC2)CCC12CCNC2')
_test_rule_min_max(test_mol, RSPAbsDelta, min_result, max_result)
def test_rsp_num_oxygen_atoms():
test_smiles = 'c1c2c(c3occ(-c4ccccc4)c(=O)c3c1)C=CCO2'
test_mol = Chem.MolFromSmiles(test_smiles)
min_result = canonize_smiles('O=c1ccoc2c3c(ccc12)OCC=C3')
max_result = canonize_smiles('O=c1c(-c2ccccc2)coc2ccccc12')
_test_rule_min_max(test_mol, RSPNumOAtoms, min_result, max_result)
