def cbhone(ich, balance=True):
"""
Fragments molecule in a way that conserves each heavy-atom/heavy-atom bond
INPUT:
ich -- STR inchi name for molecule
OUTPUT
frags -- DIC dictionary with keys as STR inchi name for fragments
and value as INT their coefficient
"""
# Graphical info about molecule
gra = automol.inchi.graph(ich)
atms = automol.graph.atoms(gra)
bnd_ords = automol.graph.one_resonance_dominant_bond_orders(gra)
rad_atms = list(automol.graph.sing_res_dom_radical_atom_keys(gra))
atm_vals = automol.graph.atom_element_valences(gra)
adj_atms = automol.graph.atoms_neighbor_atom_keys(gra)
# Determine CBHone fragments
frags = {}
for atm in atm_vals:
for adj in list(adj_atms[atm]):
if atm > adj:
vali = atm_vals[atm]
valj = atm_vals[adj]
if atm in rad_atms:
vali -= 1
if adj in rad_atms:
valj -= 1
key = frozenset({atm, adj})
bnd_ord = list(bnd_ords[key])[0]
vali -= bnd_ord
valj -= bnd_ord
atm_dic = {
0: (atms[atm][0], int(vali), None),
1: (atms[adj][0], int(valj), None)
}
bnd_dic = {frozenset({0, 1}): (1, None)}
gra = (atm_dic, bnd_dic)
frag = automol.graph.inchi(gra)
util.add2dic(frags, frag)
frags = {k: v for k, v in frags.items() if v}
if not frags:
frags = cbhzed(ich)
# Balance
if balance:
balance_ = util.balance(ich, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
zedfrags = cbhzed(ich, balance=False)
newfrags = frags.copy()
for frag in zedfrags:
util.add2dic(newfrags, frag, -zedfrags[frag])
frags = {k: v for k, v in newfrags.items() if v}
balance_ = util.balance(ich, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
frags = util.balance_frags(ich, frags)
return frags
def _add_frag_to_frags(key, coeff, grai, frags):
newname = None
repeat = False
for name, frags_dct in frags.items():
if key in frags_dct:
if key == 'exp_gra':
if automol.graph.full_isomorphism(frags_dct[key], grai):
newname = name
repeat = True
else:
if frags_dct[key] == grai:
newname = name
repeat = True
if not repeat:
newname = len(frags.keys())
frags[newname] = {}
frags[newname][key] = grai
util.add2dic(frags[newname], 'coeff', coeff)
return frags
def cbhzed(ich, balance=True):
"""
Fragments molecule so that each heavy-atom is a seperate fragment
INPUT:
ich -- STR inchi name for molecule
OUTPUT
frags -- DIC dictionary with keys as STR inchi name for fragments
and value as INT their coefficient
"""
# Graphical info about molecule
gra = automol.inchi.graph(ich)
rad_atms = list(automol.graph.sing_res_dom_radical_atom_keys(gra))
atm_vals = automol.graph.atom_element_valences(gra)
atms = automol.graph.atoms(gra)
adj_atms = automol.graph.atoms_neighbor_atom_keys(gra)
# Determine CBHzed fragments
frags = {}
for atm in atm_vals:
coeff = 1
if not balance:
coeff = (util.branch_point(adj_atms[atm]) *
util.terminal_moiety(adj_atms[atm]))
if atm in rad_atms:
atm_vals[atm] -= 1
atm_dic = {0: (atms[atm][0], int(atm_vals[atm]), None)}
gra = (atm_dic, {})
frag = automol.graph.inchi(gra)
util.add2dic(frags, frag, coeff)
if balance:
balance_ = util.balance(ich, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
frags = util.balance_frags(ich, frags)
return frags
def cbhthree(ich, balance=True):
"""
Fragments molecule to retain each heavy-atom -- heavy-atom bond,
and keep the bonds of each atm1 b1 atm2 b2 atm3 b3 atm4 b4 atm5
would give atm1 [b1] atm2 b2 atm3, atm1 b1 atm2 [b2] atm3 b3 atm4,
atm2 b2 atm3 [b3] atm4 b4 atm5, and atm3 b3 atm4 [b4] atm5
INPUT:
ich -- STR inchi name for molecule
OUTPUT
frags -- DIC dictionary with keys as STR inchi name
for fragments and value as INT their coefficient
"""
# Graphical info about molecule
gra = automol.inchi.graph(ich)
atms = automol.graph.atoms(gra)
bnd_ords = automol.graph.one_resonance_dominant_bond_orders(gra)
rad_atms = list(automol.graph.sing_res_dom_radical_atom_keys(gra))
atm_vals = automol.graph.atom_element_valences(gra)
adj_atms = automol.graph.atoms_neighbor_atom_keys(gra)
# Determine CBHfour fragments
frags = {}
for bnd in list(bnd_ords):
atm_dic = {}
bnd_dic = {}
bnd_dic[frozenset({0, 1})] = (list(bnd_ords[bnd])[0], None)
for i, atm in enumerate(list(bnd)):
vali = atm_vals[atm]
if atm in rad_atms:
vali -= 1
for adj in list(adj_atms[atm]):
key = frozenset({atm, adj})
bnd_ord = list(bnd_ords[key])[0]
vali -= bnd_ord
atm_dic[i] = (atms[atm][0], int(vali), None)
for j, adj in enumerate(list(adj_atms[atm]), start=1):
if adj not in list(bnd):
valj = atm_vals[adj]
if adj in rad_atms:
valj -= 1
key = frozenset({atm, adj})
bnd_ord = list(bnd_ords[key])[0]
valj -= bnd_ord
atm_dic[i * 4 + j + 1] = (atms[adj][0], int(valj), None)
bnd_dic[frozenset({i, i * 4 + j + 1})] = (bnd_ord, None)
gra = (atm_dic, bnd_dic)
frag = automol.graph.inchi(gra)
util.add2dic(frags, frag)
if not frags:
frags = cbhtwo(frags)
if balance:
balance_ = util.balance(ich, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
newfrags = frags.copy()
twofrags = cbhtwo(ich, balance=False)
for frag in twofrags:
util.add2dic(newfrags, frag, -twofrags[frag])
frags = {k: v for k, v in newfrags.items() if v}
# balance_ = util.balancec(ich, frags)
# balance_ = {k: v for k, v in balance_.items() if v}
# if balance_:
# newfrags = frags.copy()
# newerfrags = {}
# onefrags = cbhone(ich, balance=False)
# for frag in onefrags:
# util.add2dic(newfrags, frag, - onefrags[frag])
# frags = {k: v for k, v in newfrags.items() if v}
# balance_ = util.balancec(ich, frags)
# balance_ = {k: v for k, v in balance_.items() if v}
# if balance_:
# newfrags = frags.copy()
# zedfrags = cbhzed(ich, balance=False)
# for frag in zedfrags:
# util.add2dic(newfrags, frag, zedfrags[frag])
# frags = {k: v for k, v in newfrags.items() if v}
# balance_ = util.balancec(ich, frags)
# balance_ = {k: v for k, v in balance_.items() if v}
# if balance_:
# frags = util.balancec_frags(ich, frags)
# balance = util.balancec(ich, frags)
# balance_ = {k: v for k, v in balance_.items() if v}
return frags
def cbhtwo(ich, balance=True):
"""
Fragments molecule for each heavy-atom to stay bonded to its adjacent atoms
INPUT:
ich -- STR inchi name for molecule
OUTPUT
frags -- DIC dictionary with keys as STR inchi name for fragments and
value as INT their coefficient
"""
# Graphical info about molecule
gra = automol.inchi.graph(ich)
atms = automol.graph.atoms(gra)
bnd_ords = automol.graph.one_resonance_dominant_bond_orders(gra)
rad_atms = list(automol.graph.sing_res_dom_radical_atom_keys(gra))
atm_vals = automol.graph.atom_element_valences(gra)
adj_atms = automol.graph.atoms_neighbor_atom_keys(gra)
# Determine CBHtwo fragments
frags = {}
for atm in atms:
vali = atm_vals[atm]
if atm in rad_atms:
vali -= 1
# First loop over all atoms of this frag to get saturation of atomi
for adj in list(adj_atms[atm]):
key = frozenset({atm, adj})
bnd_ord = list(bnd_ords[key])[0]
vali -= bnd_ord
atm_dic = {0: (atms[atm][0], int(vali), None)}
bnd_dic = {}
# Then start adding bonds to the bnddic and atomdic
j = 0
coeff = 1
if not balance:
coeff = (util.branch_point(adj_atms[atm]) *
util.terminal_moiety(adj_atms[atm]))
for adj in list(adj_atms[atm]):
j += 1
valj = atm_vals[adj]
if adj in rad_atms:
valj -= 1
key = frozenset({atm, adj})
bnd_ord = list(bnd_ords[key])[0]
valj -= bnd_ord
atm_dic[j] = (atms[adj][0], int(valj), None)
bnd_dic[frozenset({0, j})] = (1, None)
gra = (atm_dic, bnd_dic)
frag = automol.graph.inchi(gra)
util.add2dic(frags, frag, coeff)
frags = {k: v for k, v in frags.items() if v}
if not frags:
frags = cbhone(frags)
# Balance
if balance:
balance_ = util.balance(ich, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
newfrags = frags.copy()
onefrags = cbhone(ich, balance=False)
for frag in onefrags:
util.add2dic(newfrags, frag, -onefrags[frag])
frags = {k: v for k, v in newfrags.items() if v}
balance_ = util.balance(ich, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
newfrags = frags.copy()
zedfrags = cbhzed(ich, balance=False)
for frag in zedfrags:
util.add2dic(newfrags, frag, zedfrags[frag])
frags = {k: v for k, v in newfrags.items() if v}
balance_ = util.balance(ich, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
frags = util.balance_frags(ich, frags)
return frags
def cbhone_habs(gra, site, atms, bnd_ords, atm_vals, adj_atms, bal=True):
"""
Fragments molecule so that each heavy-atom is a seperate fragment
INPUT:
ich -- STR inchi name for molecule
OUTPUT
frags -- DIC dictionary with keys as STR inchi name for fragments
and value as INT their coefficient
"""
frags = {}
for bnd in bnd_ords:
atma, atmb = bnd
grai = (atms.copy(), bnd_ords.copy())
extended_site = None
if ((atms[atma][0] != 'H' or atma in site)
and (atms[atmb][0] != 'H' or atmb in site)):
if atma in site and atmb in site:
continue
coeff = 1.0
if atmb in site:
atmb, atma = atma, atmb
if atma in site:
key = 'ts_gra'
extended_site = [*site, atmb]
elif atms[atma][0] != 'H' and atms[atmb][0] != 'H':
key = 'exp_gra'
extended_site = [atma, atmb]
if extended_site is not None:
for site_atm in extended_site:
for atm_x in adj_atms[site_atm]:
if atm_x not in extended_site and atms[atm_x][0] != 'H':
grai = cleave_group_and_saturate(
grai, bnd_ords, site_atm, atm_x)
grai = automol.graph.explicit(grai)
frags = _add_frag_to_frags(key, coeff, grai, frags)
frags = tsutil.simplify_gra_frags(frags)
if bal:
balance_ = util.balance_ts(gra, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
if not frags:
zedfrags = cbhzed_habs(gra,
site,
atms,
bnd_ords,
atm_vals,
adj_atms,
bal=True)
else:
zedfrags = cbhzed_habs(gra,
site,
atms,
bnd_ords,
atm_vals,
adj_atms,
bal=False)
newfrags = frags.copy()
for zedfrags_dct in zedfrags.values():
newname = None
repeat = False
if 'exp_gra' in zedfrags_dct:
key = 'exp_gra'
for onename in newfrags:
if 'exp_gra' in newfrags[onename]:
if automol.graph.full_isomorphism(
newfrags[onename][key], zedfrags_dct[key]):
newname = onename
repeat = True
else:
key = 'ts_gra'
if not repeat:
newname = len(newfrags.keys())
newfrags[newname] = {}
newfrags[newname][key] = zedfrags_dct[key]
if not frags:
util.add2dic(newfrags[newname], 'coeff',
zedfrags_dct['coeff'])
else:
util.add2dic(newfrags[newname], 'coeff',
-zedfrags_dct['coeff'])
frags = newfrags
balance_ = util.balance_ts(gra, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
frags = util.balance_frags_ts(gra, frags)
frags = tsutil.simplify_gra_frags(frags)
return frags
def cbhone_habs(gra, site, bal=True):
"""
Fragments molecule so that each heavy-atom is a seperate fragment
INPUT:
ich -- STR inchi name for molecule
OUTPUT
frags -- DIC dictionary with keys as STR inchi name for fragments
and value as INT their coefficient
"""
# Graphical info about molecule
_, atms, bnd_ords, atm_vals, _ = tsutil.ts_graph(gra, site)
# Determine CBHone fragments
frags = {}
for bnd in bnd_ords:
atma, atmb = bnd
if ((atms[atma][0] != 'H' or atma in site) and
(atms[atmb][0] != 'H' or atmb in site)):
if atma in site and atmb in site:
continue
coeff = 1.0
if atmb in site:
atmb, atma = atma, atmb
if atma in site:
key1 = [site[0], site[1]]
key1.sort()
key = frozenset({*key1})
bnd_ord1 = list(bnd_ords[key])[0]
atm_dic = {
0: (atms[site[0]][0], atm_vals[site[0]]-bnd_ord1, None)}
bnd_dct = {frozenset({0, 1}): (bnd_ord1, None)}
key1 = [site[2], site[1]]
key1.sort()
key = frozenset({*key1})
bnd_ord2 = list(bnd_ords[key])[0]
atm_dic[2] = (
atms[site[2]][0], atm_vals[site[2]]-bnd_ord2, None)
atm_dic[1] = (
atms[site[1]][0],
atm_vals[site[1]]-bnd_ord1-bnd_ord2, None)
bnd_dct[frozenset({1, 2})] = (bnd_ord2, None)
key1 = [atma, atmb]
key1.sort()
key = frozenset({*key1})
bnd_ord1 = list(bnd_ords[key])[0]
atm_dic[3] = (atms[atmb][0], atm_vals[atmb]-bnd_ord1, None)
if atma == site[0]:
atm_dic[0] = (
atm_dic[0][0], atm_dic[0][1] - bnd_ord1, None)
bnd_dct[frozenset({0, 3})] = (bnd_ord1, None)
elif atma == site[2]:
atm_dic[2] = (
atm_dic[2][0], atm_dic[2][1] - bnd_ord1, None)
bnd_dct[frozenset({2, 3})] = (bnd_ord1, None)
elif atma == site[1]:
atm_dic[1] = (
atm_dic[1][0], atm_dic[1][1] - bnd_ord1, None)
bnd_dct[frozenset({1, 3})] = (bnd_ord1, None)
else:
bnd_ord = list(bnd_ords[bnd])[0]
atm_dic = {
0: (atms[atma][0], int(atm_vals[atma])-bnd_ord, None)}
atm_dic[1] = (atms[atmb][0], int(atm_vals[atmb])-bnd_ord, None)
bnd_dct = {frozenset({0, 1}): (bnd_ord, None)}
grai = (atm_dic, bnd_dct)
try:
grai = automol.graph.explicit(grai)
key = 'exp_gra'
except:
key = 'ts_gra'
newname = None
repeat = False
for name in frags:
if key in frags[name]:
if key == 'exp_gra':
if automol.graph.full_isomorphism(
frags[name][key], grai):
newname = name
repeat = True
else:
if frags[name][key] == grai:
newname = name
repeat = True
if not repeat:
newname = len(frags.keys())
frags[newname] = {}
frags[newname][key] = grai
util.add2dic(frags[newname], 'coeff', coeff)
frags = tsutil.simplify_gra_frags(frags)
if bal:
balance_ = util.balance_ts(gra, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
zedfrags = cbhzed_habs(gra, site, bal=False)
newfrags = frags.copy()
for zedname in zedfrags:
newname = None
repeat = False
if 'exp_gra' in zedfrags[zedname]:
key = 'exp_gra'
for onename in newfrags:
if 'exp_gra' in newfrags[onename]:
if automol.graph.full_isomorphism(
newfrags[onename][key],
zedfrags[zedname][key]):
newname = onename
repeat = True
else:
key = 'ts_gra'
if not repeat:
newname = len(newfrags.keys())
newfrags[newname] = {}
newfrags[newname][key] = zedfrags[zedname][key]
util.add2dic(
newfrags[newname], 'coeff', -zedfrags[zedname]['coeff'])
frags = newfrags
balance_ = util.balance_ts(gra, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
frags = util.balance_frags_ts(gra, frags)
frags = tsutil.simplify_gra_frags(frags)
return frags
def cbhzed_habs(gra, site, bal=True):
"""
Fragments molecule so that each heavy-atom is a seperate fragment
INPUT:
ich -- STR inchi name for molecule
OUTPUT
frags -- DIC dictionary with keys as STR inchi name for fragments
and value as INT their coefficient
"""
# Graphical info about molecule
_, atms, bnd_ords, atm_vals, adj_atms = tsutil.ts_graph(gra, site)
# Determine CBHzed fragments
frags = {}
for atm in atm_vals:
if (atms[atm][0] != 'H' or atm in site):
if atm in (site[1], site[2]):
continue
coeff = 1.0
if not bal:
if atm in site:
nonhyd_adj_atms1 = tsutil.remove_hyd_from_adj_atms(
atms, adj_atms[site[0]], site,
other_adj=adj_atms[site[2]])
nonhyd_adj_atms2 = tsutil.remove_hyd_from_adj_atms(
atms, adj_atms[site[2]],
site, other_adj=adj_atms[site[0]])
nonhyd_adj_atms3 = []
for adj in adj_atms[site[0]]:
if adj in adj_atms[site[2]]:
nonhyd_adj_atms3 = tsutil.remove_hyd_from_adj_atms(
atms, adj_atms[adj], othersite=site)
coeff = (
util.branch_point(
nonhyd_adj_atms1, nonhyd_adj_atms2,
nonhyd_adj_atms3) *
util.terminal_moiety(
nonhyd_adj_atms1, nonhyd_adj_atms2,
nonhyd_adj_atms3, endisterm=False)
)
else:
nonhyd_adj_atms = tsutil.remove_hyd_from_adj_atms(
atms, adj_atms[atm])
coeff = (
util.branch_point(nonhyd_adj_atms) *
util.terminal_moiety(nonhyd_adj_atms)
)
if atm == site[0]:
key1 = [site[0], site[1]]
key1.sort()
key = frozenset({*key1})
bnd_ord1 = list(bnd_ords[key])[0]
atm_dic = {0: (atms[atm][0], atm_vals[atm]-bnd_ord1, None)}
bnd_dct = {frozenset({0, 1}): (bnd_ord1, None)}
key1 = [site[2], site[1]]
key1.sort()
key = frozenset({*key1})
bnd_ord2 = list(bnd_ords[key])[0]
atm_dic[2] = (
atms[site[2]][0], atm_vals[site[2]]-bnd_ord2, None)
atm_dic[1] = (
atms[site[1]][0],
atm_vals[site[1]]-bnd_ord1-bnd_ord2, None)
bnd_dct[frozenset({1, 2})] = (bnd_ord2, None)
else:
bnd_dct = {}
atm_dic = {0: (atms[atm][0], int(atm_vals[atm]), None)}
grai = (atm_dic, bnd_dct)
try:
grai = automol.graph.explicit(grai)
key = 'exp_gra'
except:
key = 'ts_gra'
newname = None
repeat = False
for name in frags:
if key in frags[name]:
if automol.graph.full_isomorphism(frags[name][key], grai):
newname = name
repeat = True
if not repeat:
newname = len(frags.keys())
frags[newname] = {}
frags[newname][key] = grai
util.add2dic(frags[newname], 'coeff', coeff)
frags = tsutil.simplify_gra_frags(frags)
if bal:
balance_ = util.balance_ts(gra, frags)
balance_ = {k: v for k, v in balance_.items() if v}
if balance_:
frags = util.balance_frags_ts(gra, frags)
frags = tsutil.simplify_gra_frags(frags)
return frags