def randomize_atoms(mol: Chem.rdchem.Mol) -> Optional[Chem.rdchem.Mol]:
"""Randomize the position of the atoms in a mol.
Args:
mol: a molecule.
Returns:
mol: a molecule.
"""
if mol.GetNumAtoms() == 0:
return mol
atom_indices = list(range(mol.GetNumAtoms()))
random.shuffle(atom_indices)
return Chem.RenumberAtoms(mol, atom_indices)
def reorder_atoms(
mol: Chem.rdchem.Mol,
break_ties: bool = True,
include_chirality: bool = True,
include_isotopes: bool = True,
) -> Optional[Chem.rdchem.Mol]:
"""Reorder the atoms in a mol. It ensures a single atom order for the same molecule,
regardless of its original representation.
Args:
mol: a molecule.
break_ties: Force breaking of ranked ties.
include_chirality: Use chiral information when computing rank.
include_isotopes: Use isotope information when computing rank.
Returns:
mol: a molecule.
"""
if mol.GetNumAtoms() == 0:
return mol
new_order = Chem.CanonicalRankAtoms(
mol,
breakTies=break_ties,
includeChirality=include_chirality,
includeIsotopes=include_isotopes,
)
new_order = sorted([(y, x) for x, y in enumerate(new_order)])
return Chem.RenumberAtoms(mol, [y for (x, y) in new_order])
def complete_labels(mol: Chem.rdchem.Mol,
mollabels_dict: Dict,
mark_upmatched: bool = True) -> List:
"""
Complete the gaps in the atom labels dictionary (normally a list), by given names like CX1.
:param mol: the molecule to be labelled _in place_.
:type mol: Chem.rdchem.Mol
:param mollabels_dict: key is index (int) and value is name like for a normal atomlabels (but with gaps)
:type mollabels_dict: Dict
:param mark_upmatched: Add an X between the symbol and the number
:type mark_upmatched: bool
:return: atom labels
:rtype: List[str]
"""
mollabels = []
counters = {}
for i in range(mol.GetNumAtoms()):
if i in mollabels_dict:
mollabels.append(mollabels_dict[i])
else:
el = mol.GetAtomWithIdx(i).GetSymbol().upper()
if el in counters:
counters[el] += 1
else:
counters[el] = 1
if mark_upmatched:
mollabels.append(f'{el}X{counters[el]}')
else:
mollabels.append(el + str(counters[el]))
return mollabels
def label(mol: Chem.rdchem.Mol,
atomlabels: List) -> None: # -> mol inplace.
"""
Assign the prop ``AtomLabel``... https://www.rdkit.org/docs/RDKit_Book.html
:param mol: the molecule to be labelled _in place_.
:type mol: Chem.rdchem.Mol
:param atomlabels: atom labels
:type atomlabels: List[str]
:return: None
"""
assert len(atomlabels) == mol.GetNumAtoms(
), 'the number of atoms in mol has to be the same as atomlabels. Hydrogens? dehydrogenate!'
for idx in range(mol.GetNumAtoms()):
mol.GetAtomWithIdx(idx).SetProp('AtomLabel', atomlabels[idx])
return None
def with_message_passing(mol: Chem.rdchem.Mol):
"""
Molecule with heavy atom(s)=1 or 2 is not processed message passing action. (Only for D-MPNN)
"""
m_passing_vector = [0, 0, 0] # m_passing_vector = [hv=1, hv=2, hv>2]
num_atoms = mol.GetNumAtoms()
if num_atoms == 1:
m_passing_vector[0] = 1
elif num_atoms == 2:
m_passing_vector[1] = 1
else:
m_passing_vector[2] = 1
return m_passing_vector
def num_bond_in_ring(mol: Chem.rdchem.Mol):
"""
Check the number of the bonds that are in the ring, count vector.
"""
count = 0
n_atoms = mol.GetNumAtoms()
for a1 in range(n_atoms):
for a2 in range(a1 + 1, n_atoms):
bond = mol.GetBondBetweenAtoms(a1, a2)
if bond is None:
continue
elif bond.IsInRing():
count += 1
return [count]
def display(mol: Chem.rdchem.Mol, show='name'):
# show = 'index' | 'name'
if show:
atoms = mol.GetNumAtoms()
mol = copy.deepcopy(mol)
for idx in range(atoms):
if show == 'index':
mol.GetAtomWithIdx(idx).SetProp('molAtomMapNumber',
str(idx))
elif show == 'name':
raise NotImplementedError(
'I need to figure out what property is needed as molAtomMapNumber is an str(int)'
)
mol.GetAtomWithIdx(idx).SetProp(
'molAtomMapNumber',
str(mol.GetAtomWithIdx(idx).GetProp('AtomLabel')))
else:
raise ValueError
display(Draw.MolToImage(mol))
return None
def createRDKITconf(self, mol: Chem.rdchem.Mol, conversionFactor: float = 0.1):
"""creates a PyGromosTools CNF type from a rdkit molecule. If a conformation exists the first one will be used.
Parameters
----------
mol : Chem.rdchem.Mol
Molecule, possibly with a conformation
conversionFactor : float
the factor used to convert length from rdkit to Gromos
(default: angstrom -> nano meter = 0.1)
"""
inchi = Chem.MolToInchi(mol).split("/")
if len(inchi) >= 2:
name = inchi[1]
else:
name = "XXX"
self.__setattr__("TITLE", TITLE("\t" + name + " created from RDKit"))
# check if conformations exist else create a new one
if mol.GetNumConformers() < 1:
mol = Chem.AddHs(mol)
AllChem.EmbedMolecule(mol)
AllChem.UFFOptimizeMolecule(mol)
conf = mol.GetConformer(0)
# fill a list with atomP types from RDKit data
atomList = []
for i in range(mol.GetNumAtoms()):
x = conversionFactor * conf.GetAtomPosition(i).x
y = conversionFactor * conf.GetAtomPosition(i).y
z = conversionFactor * conf.GetAtomPosition(i).z
atomType = mol.GetAtomWithIdx(i).GetSymbol()
atomList.append(blocks.atomP(resID=1, resName=name, atomType=atomType, atomID=i + 1, xp=x, yp=y, zp=z))
# set POSITION attribute
self.__setattr__("POSITION", blocks.POSITION(atomList))
# Defaults set for GENBOX - for liquid sim adjust manually
self.__setattr__("GENBOX", blocks.GENBOX(pbc=1, length=[4, 4, 4], angles=[90, 90, 90]))
def tree_decomp(
mol: Chem.rdchem.Mol) -> Tuple[List[List[int]], List[Tuple[int, int]]]:
n_atoms = mol.GetNumAtoms()
cliques = []
for atom in mol.GetAtoms():
if atom.GetDegree() == 0:
cliques.append([atom.GetIdx()])
for bond in mol.GetBonds():
a1 = bond.GetBeginAtom().GetIdx()
a2 = bond.GetEndAtom().GetIdx()
if not bond.IsInRing():
cliques.append([a1, a2])
ssr = [list(x) for x in Chem.GetSymmSSSR(mol)]
cliques.extend(ssr)
nei_list = [[] for i in range(n_atoms)]
for i in range(len(cliques)):
for atom in cliques[i]:
nei_list[atom].append(i)
# Merge Rings with intersection > 2 atoms
for i in range(len(cliques)):
if len(cliques[i]) <= 2: continue
for atom in cliques[i]:
for j in nei_list[atom]:
if i >= j or len(cliques[j]) <= 2: continue
inter = set(cliques[i]) & set(cliques[j])
if len(inter) > 2:
cliques[i].extend(cliques[j])
cliques[i] = list(set(cliques[i]))
cliques[j] = []
cliques = [c for c in cliques if len(c) > 0]
nei_list = [[] for i in range(n_atoms)]
for i in range(len(cliques)):
for atom in cliques[i]:
nei_list[atom].append(i)
# Build edges and add singleton cliques
edges = defaultdict(int)
for atom in range(n_atoms):
if len(nei_list[atom]) <= 1:
continue
cnei = nei_list[atom]
bonds = [c for c in cnei if len(cliques[c]) == 2]
rings = [c for c in cnei if len(cliques[c]) > 4]
if len(bonds) > 2 or (
len(bonds) == 2 and len(cnei) > 2
): # In general, if len(cnei) >= 3, a singleton should be added, but 1 bond + 2 ring is currently not dealt with.
cliques.append([atom])
c2 = len(cliques) - 1
for c1 in cnei:
edges[(c1, c2)] = 1
elif len(rings) > 2: # Multiple (n>2) complex rings
cliques.append([atom])
c2 = len(cliques) - 1
for c1 in cnei:
edges[(c1, c2)] = MST_MAX_WEIGHT - 1
else:
for i in range(len(cnei)):
for j in range(i + 1, len(cnei)):
c1, c2 = cnei[i], cnei[j]
inter = set(cliques[c1]) & set(cliques[c2])
if edges[(c1, c2)] < len(inter):
edges[(c1, c2)] = len(
inter) # cnei[i] < cnei[j] by construction
edges = [u + (MST_MAX_WEIGHT - v, ) for u, v in edges.items()]
if len(edges) == 0:
return cliques, edges
# Compute Maximum Spanning Tree
row, col, data = zip(*edges)
n_clique = len(cliques)
clique_graph = csr_matrix((data, (row, col)), shape=(n_clique, n_clique))
junc_tree = minimum_spanning_tree(clique_graph)
row, col = junc_tree.nonzero()
edges = [(row[i], col[i]) for i in range(len(row))]
return cliques, edges
def featurization(r_mol: Chem.rdchem.Mol,
p_mol: Chem.rdchem.Mol,
):
"""
Generates features of the reactant and product for one reaction as input for the network.
Args:
r_mol: RDKit molecule object for the reactant.
p_mol: RDKit molecule object for the product.
Returns:
data: Torch Geometric Data object, storing the atom and bond features
"""
# compute properties with rdkit (only works if dataset is clean)
r_mol.UpdatePropertyCache()
p_mol.UpdatePropertyCache()
# fake the number of "atoms" if we are collapsing substructures
n_atoms = r_mol.GetNumAtoms()
# topological and 3d distance matrices
tD_r = Chem.GetDistanceMatrix(r_mol)
tD_p = Chem.GetDistanceMatrix(p_mol)
D_r = Chem.Get3DDistanceMatrix(r_mol)
D_p = Chem.Get3DDistanceMatrix(p_mol)
f_atoms = list() # atom (node) features
edge_index = list() # list of tuples indicating presence of bonds
f_bonds = list() # bond (edge) features
for a1 in range(n_atoms):
# Node features
f_atoms.append(atom_features(r_mol.GetAtomWithIdx(a1)))
# Edge features
for a2 in range(a1 + 1, n_atoms):
# fully connected graph
edge_index.extend([(a1, a2), (a2, a1)])
# for now, naively include both reac and prod
b1_feats = [D_r[a1][a2], D_p[a1][a2]]
b2_feats = [D_r[a2][a1], D_p[a2][a1]]
# r_bond = r_mol.GetBondBetweenAtoms(a1, a2)
# b1_feats.extend(bond_features(r_bond))
# b2_feats.extend(bond_features(r_bond))
#
# p_bond = p_mol.GetBondBetweenAtoms(a1, a2)
# b1_feats.extend(bond_features(p_bond))
# b2_feats.extend(bond_features(p_bond))
f_bonds.append(b1_feats)
f_bonds.append(b2_feats)
data = tg.data.Data()
data.x = torch.tensor(f_atoms, dtype=torch.float)
data.edge_index = torch.tensor(edge_index, dtype=torch.long).t().contiguous()
data.edge_attr = torch.tensor(f_bonds, dtype=torch.float)
return data