def two_bond_idxs(gra, asymb1='H', cent='C', asymb2='H'):
""" Determine triplet of idxs for describing bond
idxs = (asymb1_idx, cent_idx, asymb2_idx)
"""
grps = tuple()
neigh_dct = atom_neighbor_keys(gra)
idx_symb_dct = atom_symbols(gra)
symb_idx_dct = atom_symbol_idxs(gra)
cent_idxs = symb_idx_dct.get(cent, tuple())
for cent_idx in cent_idxs:
neighs = tuple(neigh_dct[cent_idx])
neigh_symbs = atom_idx_to_symb(neighs, idx_symb_dct)
if neigh_symbs == (asymb1, asymb2):
grp_idxs = (neighs[0], cent_idx, neighs[1])
elif neigh_symbs == (asymb2, asymb1):
grp_idxs = (neighs[1], cent_idx, neighs[0])
else:
grp_idxs = ()
if grp_idxs:
grps += ((grp_idxs),)
return grps
def chem_unique_atoms_of_type(gra, asymb):
""" For the given atom type, determine the idxs of all the
chemically unique atoms.
"""
# Get the indices for the atom type
symb_idx_dct = atom_symbol_idxs(gra)
atom_idxs = symb_idx_dct[asymb]
# Loop over each idx
uni_idxs = tuple()
uni_del_gras = []
for idx in atom_idxs:
# Remove the atom from the graph
del_gra = remove_atoms(gra, [idx])
# Test if the del_gra is isomorphic to any of the uni_del_gras
new_uni = True
for uni_del_gra in uni_del_gras:
iso_dct = full_isomorphism(del_gra, uni_del_gra)
if iso_dct:
new_uni = False
break
# Add graph and idx to lst if del gra is unique
if new_uni:
uni_del_gras.append(del_gra)
uni_idxs += (idx,)
return uni_idxs
def _unique_atoms(gra):
""" Determine the symbols of unique atom types.
:param gra: molecular graph
:type gra: molecular graph data structure
:rtype: tuple(str)
"""
symb_idx_dct = atom_symbol_idxs(gra)
return tuple(symb_idx_dct.keys())
def _prod_group_loss(gra, grp='H'):
""" products of hydrogen loss. Need to generalize to group loss
"""
prod_gras = tuple()
symb_idx_dct = atom_symbol_idxs(gra)
h_idxs = symb_idx_dct[grp]
for idx in h_idxs:
prod_gras += ((remove_atoms(gra, [idx]), ), )
return _unique_gras(prod_gras)
def halide_groups(gra):
""" Determine the location of halide groups
"""
hal_grps = tuple()
symb_idx_dct = atom_symbol_idxs(gra)
for symb in ('F', 'Cl', 'Br', 'I'):
hal_idxs = symb_idx_dct.get(symb, ())
for hal_idx in hal_idxs:
hal_neighs = neighbors_of_type(gra, hal_idx, asymb='C')
hal_grps += ((hal_neighs[0], hal_idx),)
return hal_grps
def halide_groups(gra):
""" Determine the location of halide groups. The locations are
specified as tuple-of-tuple of idxs indicating the C-X atoms
of the group: (C-idx, X-idx).
:param gra: molecular graph
:type gra: molecular graph data structure
:rtype: tuple(int)
"""
hal_grps = tuple()
symb_idx_dct = atom_symbol_idxs(gra)
for symb in ('F', 'Cl', 'Br', 'I'):
hal_idxs = symb_idx_dct.get(symb, ())
for hal_idx in hal_idxs:
hal_neighs = neighbors_of_type(gra, hal_idx, symb='C')
hal_grps += ((hal_neighs[0], hal_idx), )
return hal_grps
def two_bond_idxs(gra, symb1, cent, symb2):
""" Determine the triplet of indices of atoms of specified
types that are connected in a chain by two bonds:
(symb1_idx, cent_idx, symb2_idx).
:param gra: molecular graph
:type gra: molecular graph data structure
:param symb1: symbol of atom at one end of chain
:type symb1: str
:param cent: symbol of atom in the middle of a chain
:type cent: str
:param symb2: symbol of atom at other end of chain
:type symb2: str
"""
grps = tuple()
neigh_dct = atoms_neighbor_atom_keys(gra)
idx_symb_dct = atom_symbols(gra)
symb_idx_dct = atom_symbol_idxs(gra)
cent_idxs = symb_idx_dct.get(cent, tuple())
for cent_idx in cent_idxs:
neighs = tuple(neigh_dct[cent_idx])
neigh_symbs = atom_idx_to_symb(neighs, idx_symb_dct)
if neigh_symbs == (symb1, symb2):
grp_idxs = (neighs[0], cent_idx, neighs[1])
elif neigh_symbs == (symb2, symb1):
grp_idxs = (neighs[1], cent_idx, neighs[0])
else:
grp_idxs = ()
if grp_idxs:
grps += ((grp_idxs), )
return grps
def _unique_atoms(gra):
""" Determine the symbols of unique atom types
"""
symb_idx_dct = atom_symbol_idxs(gra)
return tuple(symb_idx_dct.keys())