def get_formula(self):
""" Get the chemical formula
Returns:
:obj:`EmpiricalFormula`: chemical formula
"""
if self.structure:
el_table = openbabel.OBElementTable()
formula = {}
mass = 0
for i_atom in range(self.structure.NumAtoms()):
atom = self.structure.GetAtom(i_atom + 1)
el = el_table.GetSymbol(atom.GetAtomicNum())
if el in formula:
formula[el] += 1
else:
formula[el] = 1
mass += el_table.GetMass(atom.GetAtomicNum())
formula = EmpiricalFormula(formula)
# calc hydrogens because OpenBabel doesn't output this
formula['H'] = round((self.structure.GetMolWt() - mass) / el_table.GetMass(1))
return formula
else:
return None
def save_structfile(poltype,molstruct, structfname):
"""
Intent: Output the data in the OBMol structure to a file (such as *.xyz)
Input:
molstruct: OBMol structure
structfname: output file name
Output:
file is output to structfname
Referenced By: tor_opt_sp, compute_mm_tor_energy
Description: -
"""
strctext = os.path.splitext(structfname)[1]
tmpconv = openbabel.OBConversion()
if strctext in '.xyz':
tmpfh = open(structfname, "w")
maxidx = max(poltype.symmetryclass)
iteratom = openbabel.OBMolAtomIter(molstruct)
etab = openbabel.OBElementTable()
tmpfh.write('%6d %s\n' % (molstruct.NumAtoms(), molstruct.GetTitle()))
for ia in iteratom:
tmpfh.write( '%6d %2s %13.6f %11.6f %11.6f %5d' % (ia.GetIdx(), etab.GetSymbol(ia.GetAtomicNum()), ia.x(), ia.y(), ia.z(), poltype.prmstartidx + (maxidx - poltype.symmetryclass[ia.GetIdx() - 1])))
iteratomatom = openbabel.OBAtomAtomIter(ia)
neighbors = []
for iaa in iteratomatom:
neighbors.append(iaa.GetIdx())
neighbors = sorted(neighbors)
for iaa in neighbors:
tmpfh.write('%5d' % iaa)
tmpfh.write('\n')
else:
inFormat = openbabel.OBConversion.FormatFromExt(structfname)
tmpconv.SetOutFormat(inFormat)
return tmpconv.WriteFile(molstruct, structfname)
def _parse_atom_site_openbabel(parsed):
"""Parse _atom_site record to OBMolecule
Args:
parsed (dict of str): Parsed mmcif file.
Returns:
[OBMol]: openbabel representation of the protein structure.
"""
perceived_atom_site = parsed["_atom_site"]
atom_site = _trim_models(perceived_atom_site)
table = ob.OBElementTable()
last_res_id = None
last_res_name = None
last_chain_id = None
chain_num = 0
res = None
mol = ob.OBMol()
mol.SetChainsPerceived()
mol.BeginModify()
for i in range(len(atom_site["id"])):
current_res_id = _get_res_id(atom_site, i)
ins_code = _get_ins_code(atom_site, i)
if last_chain_id != atom_site["auth_asym_id"][i]:
chain_num += 1
last_chain_id = atom_site["auth_asym_id"][i]
if (current_res_id != last_res_id
or atom_site["auth_asym_id"][i] != last_chain_id
or atom_site["label_comp_id"][i] != last_res_name):
last_res_id = current_res_id
last_res_name = atom_site["label_comp_id"][i]
res = mol.NewResidue()
res.SetChainNum(chain_num)
res.SetNum(str(last_res_id))
res.SetName(atom_site["label_comp_id"][i])
res.SetInsertionCode(ins_code)
_init_openbabel_atom(table, mol, res, atom_site, i)
resdat = ob.OBResidueData()
resdat.AssignBonds(mol, ob.OBBitVec())
mol.ConnectTheDots()
mol.PerceiveBondOrders()
if "_struct_conn" in parsed:
parse_struct_conn_bonds(mol, parsed)
mol.EndModify()
return mol
def get_mass(types):
m = []
etab = ob.OBElementTable()
for value in types:
a = etab.GetAtomicNum(value)
c = etab.GetMass(a)
m.append(c)
return m
def get_mol_species(mol):
species = []
num_atoms = mol.NumAtoms()
element_table = ob.OBElementTable()
for i in range(1, num_atoms + 1):
a = mol.GetAtom(i)
atomic_num = a.GetAtomicNum()
symbol = element_table.GetSymbol(atomic_num)
species.append(symbol)
return species
def read_txt_to_mol(txtfile):
# lists to store the charges and LJ parameters
q = []
eps = []
sig = []
# convert the .txt to a .xyz to read in pybel mol to perceive all the bonds, angles, dihedrals..
fd, temp_path = tempfile.mkstemp(suffix=".xyz")
fxyz = os.fdopen(fd, 'w')
# table to convert atomic number to symbols
if not ob3:
etab = openbabel.OBElementTable()
with open(txtfile, 'r') as f:
line = f.readline()
combrule = line.strip()
line = f.readline()
if int(line.strip()) != 1:
print(
"Your .txt should have only one molecule, the one present in the .dfr"
)
sys.exit(0)
line = f.readline()
natoms = int(line.split()[0])
# write .xyz header
fxyz.write("%d\nGenerated from %s\n" % (natoms, txtfile))
for i in range(natoms):
# write the atomic symbol and coordinates
line = f.readline()
atnum = int(line.split()[1])
x, y, z = [float(x) for x in line.split()[2:5]]
if ob3:
fxyz.write("%s\t%f\t%f\t%f\n" %
(openbabel.GetSymbol(atnum), x, y, z))
else:
fxyz.write("%s\t%f\t%f\t%f\n" %
(etab.GetSymbol(atnum), x, y, z))
# store the parameters
qv, epsv, sigv = [float(x) for x in line.split()[5:]]
q.append(qv)
eps.append(epsv)
sig.append(sigv)
fxyz.close()
# read filel to a mol, remove temp and return
mol = pybel.readfile("xyz", temp_path).__next__()
os.remove(temp_path)
return mol, q, eps, sig
def get_mol_info(mol):
# table to convert atomic number to symbols
etab = openbabel.OBElementTable()
q_atoms = []
q_all = []
for atom in mol:
q_atoms.append(etab.GetSymbol(atom.atomicnum))
q_all.append(atom.coords)
return np.asarray(q_atoms), np.asarray(q_all)
def to_openbabel_Mol(mol, CalcBondmap = False):
obmol = ob.OBMol()
if not mol.coords == None:
for atom, coord in zip(mol.atoms, mol.coords):
obatom = ob.OBAtom()
obatom.SetAtomicNum(ob.OBElementTable().GetAtomicNum(atom))
coord_vec = ob.vector3(coord[0], coord[1], coord[2])
obatom.SetVector(coord_vec)
obmol.InsertAtom(obatom)
else:
for atom in mol.atoms:
obatom = ob.OBAtom()
obatom.SetAtomicNum(ob.OBElementTable().GetAtomicNum(atom))
obmol.InsertAtom(obatom)
if CalcBondmap == True:
obmol.ConnectTheDots()
obmol.PerceiveBondOrders()
else:
for bond in mol.bondmap:
obmol.AddBond(bond[0] + 1, bond[1] + 1, bond[2])
return obmol
def CreatePsi4TorESPInputFile(poltype,finalstruct,torxyzfname,optmol,molecprefix,a,b,c,d,torang,phaseangle,makecube=None):
inputname= '%s-sp-%d-%d-%d-%d-%03d_psi4.dat' % (molecprefix,a,b,c,d,round((torang+phaseangle)%360))
temp=open(inputname,'w')
temp.write('molecule { '+'\n')
temp.write('%d %d\n' % (optmol.GetTotalCharge(), 1))
iteratom = openbabel.OBMolAtomIter(optmol)
etab = openbabel.OBElementTable()
if os.path.isfile(torxyzfname):
xyzstr = open(torxyzfname,'r')
xyzstrl = xyzstr.readlines()
i = 0
for atm in iteratom:
i = i + 1
ln = xyzstrl[i]
temp.write('%2s %11.6f %11.6f %11.6f\n' % (etab.GetSymbol(atm.GetAtomicNum()), float(ln.split()[2]),float(ln.split()[3]),float(ln.split()[4])))
xyzstr.close()
temp.write('}'+'\n')
if poltype.torsppcm==True:
temp.write('set {'+'\n')
temp.write(' basis '+poltype.torspbasisset+'\n')
temp.write(' e_convergence 10 '+'\n')
temp.write(' d_convergence 10 '+'\n')
temp.write(' scf_type pk'+'\n')
temp.write(' pcm true'+'\n')
temp.write(' pcm_scf_type total '+'\n')
temp.write('}'+'\n')
temp.write('pcm = {'+'\n')
temp.write(' Units = Angstrom'+'\n')
temp.write(' Medium {'+'\n')
temp.write(' SolverType = IEFPCM'+'\n')
temp.write(' Solvent = Water'+'\n')
temp.write(' }'+'\n')
temp.write(' Cavity {'+'\n')
temp.write(' RadiiSet = UFF'+'\n')
temp.write(' Type = GePol'+'\n')
temp.write(' Scaling = False'+'\n')
temp.write(' Area = 0.3'+'\n')
temp.write(' Mode = Implicit'+'\n')
temp.write(' }'+'\n')
temp.write('}'+'\n')
temp.write('memory '+poltype.torsmem+'\n')
temp.write('set_num_threads(%s)'%(poltype.numproc)+'\n')
temp.write('psi4_io.set_default_path("%s")'%(poltype.scratchdir)+'\n')
temp.write('set freeze_core True'+'\n')
temp.write("E, wfn = energy('%s/%s',return_wfn=True)" % (poltype.torspmethod.lower(),poltype.torspbasisset)+'\n')
temp.write('oeprop("WIBERG_LOWDIN_INDICES")'+'\n')
temp.write('clean()'+'\n')
temp.close()
outputname=os.path.splitext(inputname)[0] + '.log'
return inputname,outputname
def _add_atom_collection(self, zorder=100, **kwargs):
"""Draw atoms as colored circles on the axes."""
col = []
colors = []
etab = ob.OBElementTable()
for atom in ob.OBMolAtomIter(self.molecule):
colors.append(etab.GetRGB(atom.GetAtomicNum()))
radius = etab.GetCovalentRad(atom.GetAtomicNum())
circle = Circle(self._2Dcoords(atom), radius)
col.append(circle)
kw = {'facecolors': colors, 'edgecolors': colors}
kwargs.update(kw)
self._mol_atoms = PatchCollection(col, zorder=zorder, **kwargs)
self.axes.add_collection(self._mol_atoms)
def gen_optcomfile(poltype,comfname,numproc,maxmem,maxdisk,chkname,mol):
"""
Intent: Create *.com file for qm opt
Input:
comfname: com file name
numproc: number of processors
maxmem: max memory size
chkname: chk file name
mol: OBMol object
Output:
*opt-*.com is written
Referenced By: run_gaussian
Description: -
"""
restraintlist = []
write_com_header(poltype,comfname,chkname,maxdisk,maxmem,numproc)
tmpfh = open(comfname, "a")
optimizeoptlist = ["maxcycle=%s"%(poltype.optmaxcycle)]
if restraintlist:
optimizeoptlist.insert(0,poltype.gausoptcoords)
optstr=gen_opt_str(poltype,optimizeoptlist)
if ('I ' in mol.GetSpacedFormula()):
optstring="%s HF/Gen freq Guess=INDO MaxDisk=%s\n" % (optstr,maxdisk)
else:
if poltype.freq==True:
if poltype.optpcm==True:
optstring= "%s %s/%s freq Guess=INDO MaxDisk=%s SCRF=(PCM)\n" % (optstr,poltype.optmethod,poltype.optbasisset,maxdisk)
else:
optstring= "%s %s/%s freq Guess=INDO MaxDisk=%s\n" % (optstr,poltype.optmethod,poltype.optbasisset,maxdisk)
else:
if poltype.optpcm==True:
optstring= "%s %s/%s Guess=INDO MaxDisk=%s SCRF=(PCM)\n" % (optstr,poltype.optmethod,poltype.optbasisset,maxdisk)
else:
optstring= "%s %s/%s Guess=INDO MaxDisk=%s\n" % (optstr,poltype.optmethod,poltype.optbasisset,maxdisk)
tmpfh.write(optstring)
commentstr = poltype.molecprefix + " Gaussian SP Calculation on " + gethostname()
tmpfh.write('\n%s\n\n' % commentstr)
tmpfh.write('%d %d\n' % (mol.GetTotalCharge(), mol.GetTotalSpinMultiplicity()))
tmpfh.close()
iteratombab = openbabel.OBMolAtomIter(mol)
print('atomnumber',mol.NumAtoms())
tmpfh = open(comfname, "a")
etab = openbabel.OBElementTable()
for atm in iteratombab:
tmpfh.write('%2s %11.6f %11.6f %11.6f\n' % (etab.GetSymbol(atm.GetAtomicNum()), atm.x(), atm.y(), atm.z()))
tmpfh.write('\n')
tmpfh.close()
def get_radius(self, mol):
radius = []
ref_radius = CovalentRadius.radius
num_atoms = mol.NumAtoms()
element_table = ob.OBElementTable()
for i in range(1, num_atoms + 1):
a = mol.GetAtom(i)
atomic_num = a.GetAtomicNum()
symbol = element_table.GetSymbol(atomic_num)
if symbol in self.metals:
scale = self.metal_radius_scale
else:
scale = self.covalent_radius_scale
rad = ref_radius[symbol] * self.angstrom2au * scale
radius.append(rad)
return radius
def __sortElements(elts):
"""
Sorts elements in list according to atomic number
"""
table = openbabel.OBElementTable()
res = list()
for e in elts:
for l in [3, 2, 1]:
num = table.GetAtomicNum(e[:l])
if num > 0:
res.append("%03d%s" % (num, e))
break
res.sort()
#print "__sortElements", res
del elts[:]
for e in res:
elts.append(e[3:])
def _add_atom_labels(self, zorder=100, **kwargs):
"""Draw atom labels on the axes."""
box_props = dict(boxstyle='round', facecolor='white', edgecolor='none')
etab = ob.OBElementTable()
for atom in ob.OBMolAtomIter(self.molecule):
x, y = self._2Dcoords(atom)
kw = dict(horizontalalignment="center",
verticalalignment="center",
bbox=box_props)
kwargs.update(kw)
label = AtomText(x,
y,
etab.GetSymbol(atom.GetAtomicNum()),
atom.GetIdx(),
etab.GetRGB(atom.GetAtomicNum()),
zorder=zorder,
**kwargs)
self._mol_labels.append(label)
self.axes.add_artist(label)
def atom_groups(self):
group_dict = {}
pse = openbabel.OBElementTable()
for at in openbabel.OBMolAtomIter(self.obmol):
atname = pse.GetSymbol(at.GetAtomicNum())
if atname in group_dict:
group_dict[atname].append(at.GetIdx() - 1)
else:
group_dict[atname] = [at.GetIdx() - 1]
groupnames = []
groups = []
for g in group_dict:
groupnames.append(g)
groups.append(group_dict[g])
return groups, groupnames
def iter_bonds( self ) :
"""bonds iterator"""
if self._verbose : sys.stdout.write( "%s.iter_bonds()\n" % (self.__class__.__name__,) )
et = openbabel.OBElementTable()
for i in xrange( self._mol.OBMol.NumBonds() ) :
bond = self._mol.OBMol.GetBond( i )
rc = { "id" : (bond.GetIdx() + 1) }
rc["type"] = "covalent"
if bond.IsAmide() : rc["type"] = "amide"
elif bond.IsEster() : rc["type"] = "ester"
elif bond.IsCarbonyl() : rc["type"] = "carbonyl"
elt1 = et.GetSymbol( bond.GetBeginAtom().GetAtomicNum() )
if str( elt1 ).lower() == "xx" : elt1 = "x"
rc["atom1"] = "%s%d" % (elt1,bond.GetBeginAtom().GetIdx())
elt2 = et.GetSymbol( bond.GetEndAtom().GetAtomicNum() )
if str( elt2 ).lower() == "xx" : elt2 = "x"
rc["atom2"] = "%s%d" % (elt2,bond.GetEndAtom().GetIdx())
yield rc
def parse_mol_info(fname, fcharges, axis, buffa, buffo, pbcbonds, printdih,
ignorebonds, ignoreimproper):
iaxis = {"x": 0, "y": 1, "z": 2}
if axis in iaxis:
repaxis = iaxis[axis]
else:
print("Error: invalid axis")
sys.exit(0)
if fcharges:
chargesLabel = {}
with open(fcharges, "r") as f:
for line in f:
chargesLabel[line.split()[0]] = float(line.split()[1])
# set openbabel file format
base, ext = os.path.splitext(fname)
obConversion = openbabel.OBConversion()
obConversion.SetInAndOutFormats(ext[1:], "xyz")
# trick to disable ring perception and make the ReadFile waaaay faster
# Source: https://sourceforge.net/p/openbabel/mailman/openbabel-discuss/thread/56e1812d-396a-db7c-096d-d378a077853f%40ipcms.unistra.fr/#msg36225392
obConversion.AddOption("b", openbabel.OBConversion.INOPTIONS)
# read molecule to OBMol object
mol = openbabel.OBMol()
obConversion.ReadFile(mol, fname)
mol.ConnectTheDots() # necessary because of the 'b' INOPTION
# split the molecules
molecules = mol.Separate()
# detect the molecules types
mTypes = {}
mapmTypes = {}
atomIdToMol = {}
nty = 0
for i, submol in enumerate(molecules, start=1):
atomiter = openbabel.OBMolAtomIter(submol)
atlist = []
for at in atomiter:
atlist.append(at.GetAtomicNum())
atomIdToMol[at.GetId()] = i
foundType = None
for ty in mTypes:
# check if there's already a molecule of this type
if atlist == mTypes[ty]:
foundType = ty
# if not, create a new type
if not foundType:
nty += 1
foundType = nty
mTypes[nty] = atlist
mapmTypes[i] = foundType
# get atomic labels from pdb
idToAtomicLabel = {}
if ext[1:] == "pdb":
for res in openbabel.OBResidueIter(mol):
for atom in openbabel.OBResidueAtomIter(res):
if (atomIdToMol[atom.GetId()] > 1) and (len(mTypes) > 1):
idToAtomicLabel[
atom.GetId()] = res.GetAtomID(atom).strip() + str(
mapmTypes[atomIdToMol[atom.GetId()]])
else:
idToAtomicLabel[atom.GetId()] = res.GetAtomID(atom).strip()
else:
if not ob3:
etab = openbabel.OBElementTable()
for atom in openbabel.OBMolAtomIter(mol):
if (atomIdToMol[atom.GetId()] > 1) and (len(mTypes) > 1):
if ob3:
idToAtomicLabel[atom.GetId()] = openbabel.GetSymbol(
atom.GetAtomicNum()) + str(
mapmTypes[atomIdToMol[atom.GetId()]])
else:
idToAtomicLabel[atom.GetId()] = etab.GetSymbol(
atom.GetAtomicNum()) + str(
mapmTypes[atomIdToMol[atom.GetId()]])
else:
if ob3:
idToAtomicLabel[atom.GetId()] = openbabel.GetSymbol(
atom.GetAtomicNum())
else:
idToAtomicLabel[atom.GetId()] = etab.GetSymbol(
atom.GetAtomicNum())
# print(idToAtomicLabel)
# identify atom types and get masses
outMasses = "Masses\n\n"
massTypes = {}
mapTypes = {}
nmassTypes = 0
atomIterator = openbabel.OBMolAtomIter(mol)
for atom in atomIterator:
i = atom.GetId()
if idToAtomicLabel[i] not in massTypes:
nmassTypes += 1
mapTypes[nmassTypes] = idToAtomicLabel[i]
massTypes[idToAtomicLabel[i]] = nmassTypes
outMasses += "\t%d\t%.3f\t# %s\n" % (
nmassTypes, atom.GetAtomicMass(), idToAtomicLabel[i])
# create atoms list
outAtoms = "Atoms # full\n\n"
xmin = float("inf")
xmax = float("-inf")
ymin = float("inf")
ymax = float("-inf")
zmin = float("inf")
zmax = float("-inf")
natoms = 0
acoords = []
for mnum, imol in enumerate(molecules, start=1):
atomIterator = openbabel.OBMolAtomIter(imol)
for atom in sorted(atomIterator, key=lambda x: x.GetId()):
natoms += 1
i = atom.GetId()
apos = (atom.GetX(), atom.GetY(), atom.GetZ())
acoords.append(Atom(atom.GetAtomicNum(), apos))
# look for the maximum and minimum x for the box (improve later with numpy and all coordinates)
if apos[0] > xmax:
xmax = apos[0]
if apos[0] < xmin:
xmin = apos[0]
if apos[1] > ymax:
ymax = apos[1]
if apos[1] < ymin:
ymin = apos[1]
if apos[2] > zmax:
zmax = apos[2]
if apos[2] < zmin:
zmin = apos[2]
if fcharges:
outAtoms += "\t%d\t%d\t%d\t%.6f\t%.4f\t%.4f\t%.4f\t# %s\n" % (
i + 1, mnum, massTypes[idToAtomicLabel[i]],
chargesLabel[idToAtomicLabel[i]], atom.GetX(), atom.GetY(),
atom.GetZ(), idToAtomicLabel[i])
else:
outAtoms += "\t%d\t%d\t%d\tX.XXXXXX\t%.4f\t%.4f\t%.4f\t# %s\n" % (
i + 1, mnum, massTypes[idToAtomicLabel[i]], atom.GetX(),
atom.GetY(), atom.GetZ(), idToAtomicLabel[i])
# define box shape and size
try:
fromBounds = False
rcell = mol.GetData(12)
cell = openbabel.toUnitCell(rcell)
v1 = [
cell.GetCellVectors()[0].GetX(),
cell.GetCellVectors()[0].GetY(),
cell.GetCellVectors()[0].GetZ()
]
v2 = [
cell.GetCellVectors()[1].GetX(),
cell.GetCellVectors()[1].GetY(),
cell.GetCellVectors()[1].GetZ()
]
v3 = [
cell.GetCellVectors()[2].GetX(),
cell.GetCellVectors()[2].GetY(),
cell.GetCellVectors()[2].GetZ()
]
boxinfo = [v1, v2, v3]
orthogonal = True
for i, array in enumerate(boxinfo):
for j in range(3):
if i == j:
continue
if not math.isclose(0., array[j], abs_tol=1e-6):
orthogonal = False
except:
fromBounds = True
v1 = [xmax - xmin, 0., 0.]
v2 = [0., ymax - ymin, 0.]
v3 = [0., 0., zmax - zmin]
orthogonal = True
# add buffer
if orthogonal:
buf = []
boxinfo = [v1, v2, v3]
for i, val in enumerate(boxinfo[repaxis]):
if i == repaxis:
buf.append(val + buffa)
else:
buf.append(val)
boxinfo[repaxis] = buf
for i in range(3):
if i == repaxis:
continue
buf = []
for j, val in enumerate(boxinfo[i]):
if j == i:
buf.append(val + buffo)
else:
buf.append(val)
boxinfo[i] = buf
# print(boxinfo)
# Duplicate to get the bonds in the PBC. Taken from (method _crd2bond):
# https://github.com/tongzhugroup/mddatasetbuilder/blob/66eb0f15e972be0f5534dcda27af253cd8891ff2/mddatasetbuilder/detect.py#L213
if pbcbonds:
acoords = Atoms(acoords, cell=boxinfo, pbc=True)
repatoms = acoords.repeat(
2
)[natoms:] # repeat the unit cell in each direction (len(repatoms) = 7*natoms)
tree = cKDTree(acoords.get_positions())
d = tree.query(repatoms.get_positions(), k=1)[0]
nearest = d < 8.
ghost_atoms = repatoms[nearest]
realnumber = np.where(nearest)[0] % natoms
acoords += ghost_atoms
write("replicated.xyz",
acoords) # write the structure with the replicated atoms
# write new mol with new bonds
nmol = openbabel.OBMol()
nmol.BeginModify()
for idx, (num, position) in enumerate(
zip(acoords.get_atomic_numbers(), acoords.positions)):
a = nmol.NewAtom(idx)
a.SetAtomicNum(int(num))
a.SetVector(*position)
nmol.ConnectTheDots()
# nmol.PerceiveBondOrders() # super slow becauses it looks for rings
nmol.EndModify()
else:
acoords = Atoms(acoords, cell=boxinfo, pbc=False)
nmol = openbabel.OBMol()
nmol.BeginModify()
for idx, (num, position) in enumerate(
zip(acoords.get_atomic_numbers(), acoords.positions)):
a = nmol.NewAtom(idx)
a.SetAtomicNum(int(num))
a.SetVector(*position)
nmol.ConnectTheDots()
# nmol.PerceiveBondOrders() # super slow becauses it looks for rings
nmol.EndModify()
# identify bond types and create bond list
outBonds = "Bonds # harmonic\n\n"
bondTypes = {}
mapbTypes = {}
nbondTypes = 0
nbonds = 0
bondsToDelete = []
bondIterators = []
if ignorebonds:
sepmols = nmol.Separate()
for smol in sepmols[1:]:
bondIterators.append(openbabel.OBMolBondIter(smol))
else:
bondIterators.append(openbabel.OBMolBondIter(nmol))
lastidx = 1
for iterator in bondIterators:
for i, bond in enumerate(iterator, lastidx):
b1 = bond.GetBeginAtom().GetId()
b2 = bond.GetEndAtom().GetId()
# check if its a bond of the replica only
if (b1 >= natoms) and (b2 >= natoms):
bondsToDelete.append(bond)
continue
# remap to a real atom if needed
if b1 >= natoms:
b1 = realnumber[b1 - natoms]
if b2 >= natoms:
b2 = realnumber[b2 - natoms]
# identify bond type
btype1 = "%s - %s" % (idToAtomicLabel[b1], idToAtomicLabel[b2])
btype2 = "%s - %s" % (idToAtomicLabel[b2], idToAtomicLabel[b1])
if btype1 in bondTypes:
bondid = bondTypes[btype1]
bstring = btype1
elif btype2 in bondTypes:
bondid = bondTypes[btype2]
bstring = btype2
else:
nbondTypes += 1
mapbTypes[nbondTypes] = btype1
bondid = nbondTypes
bondTypes[btype1] = nbondTypes
bstring = btype1
nbonds += 1
outBonds += "\t%d\t%d\t%d\t%d\t# %s\n" % (nbonds, bondid, b1 + 1,
b2 + 1, bstring)
lastidx = i
# delete the bonds of atoms from other replicas
for bond in bondsToDelete:
nmol.DeleteBond(bond)
# identify angle types and create angle list
angleTypes = {}
mapaTypes = {}
nangleTypes = 0
nangles = 0
angleIterators = []
if ignorebonds:
sepmols = nmol.Separate()
for smol in sepmols[1:]:
smol.FindAngles()
angleIterators.append(openbabel.OBMolAngleIter(smol))
prevnumatoms = sepmols[0].NumAtoms()
else:
nmol.FindAngles()
angleIterators.append(openbabel.OBMolAngleIter(nmol))
outAngles = "Angles # harmonic\n\n"
lastidx = 1
for j, iterator in enumerate(angleIterators, 1):
for i, angle in enumerate(iterator, lastidx):
if ignorebonds:
a1 = angle[1] + prevnumatoms
a2 = angle[0] + prevnumatoms
a3 = angle[2] + prevnumatoms
else:
a1 = angle[1]
a2 = angle[0]
a3 = angle[2]
# remap to a real atom if needed
if a1 >= natoms:
a1 = realnumber[a1 - natoms]
if a2 >= natoms:
a2 = realnumber[a2 - natoms]
if a3 >= natoms:
a3 = realnumber[a3 - natoms]
atype1 = "%s - %s - %s" % (
idToAtomicLabel[a1], idToAtomicLabel[a2], idToAtomicLabel[a3])
atype2 = "%s - %s - %s" % (
idToAtomicLabel[a3], idToAtomicLabel[a2], idToAtomicLabel[a1])
if atype1 in angleTypes:
angleid = angleTypes[atype1]
astring = atype1
elif atype2 in angleTypes:
angleid = angleTypes[atype2]
astring = atype2
else:
nangleTypes += 1
mapaTypes[nangleTypes] = atype1
angleid = nangleTypes
angleTypes[atype1] = nangleTypes
astring = atype1
nangles += 1
outAngles += "\t%d\t%d\t%d\t%d\t%d\t# %s\n" % (
nangles, angleid, a1 + 1, a2 + 1, a3 + 1, astring)
lastidx = i
if ignorebonds:
prevnumatoms += sepmols[j].NumAtoms()
# identify dihedral types and create dihedral list
if printdih:
dihedralTypes = {}
mapdTypes = {}
ndihedralTypes = 0
ndihedrals = 0
dihedralIterators = []
if ignorebonds:
sepmols = nmol.Separate()
for smol in sepmols[1:]:
smol.FindTorsions()
dihedralIterators.append(openbabel.OBMolTorsionIter(smol))
else:
nmol.FindTorsions()
dihedralIterators.append(openbabel.OBMolTorsionIter(nmol))
outDihedrals = "Dihedrals # charmmfsw\n\n"
lastidx = 1
for iterator in dihedralIterators:
for i, dihedral in enumerate(iterator, lastidx):
a1 = dihedral[0]
a2 = dihedral[1]
a3 = dihedral[2]
a4 = dihedral[3]
# remap to a real atom if needed
if a1 >= natoms:
a1 = realnumber[a1 - natoms]
if a2 >= natoms:
a2 = realnumber[a2 - natoms]
if a3 >= natoms:
a3 = realnumber[a3 - natoms]
if a4 >= natoms:
a4 = realnumber[a4 - natoms]
dtype1 = "%s - %s - %s - %s" % (
idToAtomicLabel[a1], idToAtomicLabel[a2],
idToAtomicLabel[a3], idToAtomicLabel[a4])
dtype2 = "%s - %s - %s - %s" % (
idToAtomicLabel[a4], idToAtomicLabel[a3],
idToAtomicLabel[a2], idToAtomicLabel[a1])
if dtype1 in dihedralTypes:
dihedralid = dihedralTypes[dtype1]
dstring = dtype1
elif dtype2 in dihedralTypes:
dihedralid = dihedralTypes[dtype2]
dstring = dtype2
else:
ndihedralTypes += 1
mapdTypes[ndihedralTypes] = dtype1
dihedralid = ndihedralTypes
dihedralTypes[dtype1] = ndihedralTypes
dstring = dtype1
ndihedrals += 1
outDihedrals += "\t%d\t%d\t%d\t%d\t%d\t%d\t# %s\n" % (
ndihedrals, dihedralid, a1 + 1, a2 + 1, a3 + 1, a4 + 1,
dstring)
lastidx = i
if not ignoreimproper:
# look for the improper dihedrals
improperDihedralTypes = {}
mapiDTypes = {}
niDihedralTypes = 0
niDihedrals = 0
mollist = []
if ignorebonds:
sepmols = nmol.Separate()
for smol in sepmols[1:]:
smol.PerceiveBondOrders()
mollist.append(smol)
else:
nmol.PerceiveBondOrders()
mollist.append(nmol)
outImpropers = "Impropers # harmonic\n\n"
for imol in mollist:
atomIterator = openbabel.OBMolAtomIter(imol)
for atom in atomIterator:
try:
# print(atom.GetHyb(), atom.GetAtomicNum(), atom.GetValence())
expDegree = atom.GetValence()
except:
# print(atom.GetHyb(), atom.GetAtomicNum(), atom.GetExplicitDegree())
expDegree = atom.GetExplicitDegree()
# returns impropers for atoms with connected to other 3 atoms and SP2 hybridization
if atom.GetHyb() == 2 and expDegree == 3:
connectedAtoms = []
for atom2, depth in openbabel.OBMolAtomBFSIter(
imol,
atom.GetId() + 1):
if depth == 2:
connectedAtoms.append(atom2)
torsional = [
atom.GetId() + 1, connectedAtoms[0].GetId() + 1,
connectedAtoms[1].GetId() + 1,
connectedAtoms[2].GetId() + 1
]
a1 = torsional[0] - 1
a2 = torsional[1] - 1
a3 = torsional[2] - 1
a4 = torsional[3] - 1
# remap to a real atom if needed
if a1 >= natoms:
a1 = realnumber[a1 - natoms]
if a2 >= natoms:
a2 = realnumber[a2 - natoms]
if a3 >= natoms:
a3 = realnumber[a3 - natoms]
if a4 >= natoms:
a4 = realnumber[a4 - natoms]
dtype1 = "%s - %s - %s - %s" % (
idToAtomicLabel[a1], idToAtomicLabel[a2],
idToAtomicLabel[a3], idToAtomicLabel[a4])
dtype2 = "%s - %s - %s - %s" % (
idToAtomicLabel[a4], idToAtomicLabel[a3],
idToAtomicLabel[a2], idToAtomicLabel[a1])
if dtype1 in improperDihedralTypes:
idihedralid = improperDihedralTypes[dtype1]
dstring = dtype1
elif dtype2 in improperDihedralTypes:
idihedralid = improperDihedralTypes[dtype2]
dstring = dtype2
else:
niDihedralTypes += 1
mapiDTypes[niDihedralTypes] = dtype1
idihedralid = niDihedralTypes
improperDihedralTypes[dtype1] = niDihedralTypes
dstring = dtype1
niDihedrals += 1
outImpropers += "\t%d\t%d\t%d\t%d\t%d\t%d\t# %s\n" % (
niDihedrals, idihedralid, a1 + 1, a2 + 1, a3 + 1,
a4 + 1, dstring)
# print header
if printdih and (ndihedrals > 0):
if ignoreimproper or (niDihedrals == 0):
header = "LAMMPS topology created from %s using pdb2lmp.py - By Henrique Musseli Cezar, 2020\n\n\t%d atoms\n\t%d bonds\n\t%d angles\n\t%d dihedrals\n\n\t%d atom types\n\t%d bond types\n\t%d angle types\n\t%d dihedral types\n\n" % (
fname, natoms, nbonds, nangles, ndihedrals, nmassTypes,
nbondTypes, nangleTypes, ndihedralTypes)
else:
header = "LAMMPS topology created from %s using pdb2lmp.py - By Henrique Musseli Cezar, 2020\n\n\t%d atoms\n\t%d bonds\n\t%d angles\n\t%d dihedrals\n\t%d impropers\n\n\t%d atom types\n\t%d bond types\n\t%d angle types\n\t%d dihedral types\n\t%d improper types\n\n" % (
fname, natoms, nbonds, nangles, ndihedrals, niDihedrals,
nmassTypes, nbondTypes, nangleTypes, ndihedralTypes,
niDihedralTypes)
else:
header = "LAMMPS topology created from %s using pdb2lmp.py - By Henrique Musseli Cezar, 2020\n\n\t%d atoms\n\t%d bonds\n\t%d angles\n\n\t%d atom types\n\t%d bond types\n\t%d angle types\n\n" % (
fname, natoms, nbonds, nangles, nmassTypes, nbondTypes,
nangleTypes)
# add box info
if fromBounds:
boxsize = [(xmin, xmax), (ymin, ymax), (zmin, zmax)]
boxsize[repaxis] = (boxsize[repaxis][0] - buffa / 2.,
boxsize[repaxis][1] + buffa / 2.)
for i in range(3):
if i == repaxis:
continue
boxsize[i] = (boxsize[i][0] - buffo / 2.,
boxsize[i][1] + buffo / 2.)
header += "\t%.8f\t%.8f\t xlo xhi\n\t%.8f\t%.8f\t ylo yhi\n\t%.8f\t%.8f\t zlo zhi\n" % (
boxsize[0][0], boxsize[0][1], boxsize[1][0], boxsize[1][1],
boxsize[2][0], boxsize[2][1])
else:
if orthogonal:
header += "\t%.8f\t%.8f\t xlo xhi\n\t%.8f\t%.8f\t ylo yhi\n\t%.8f\t%.8f\t zlo zhi\n" % (
0., boxinfo[0][0], 0., boxinfo[1][1], 0., boxinfo[2][2])
else:
header += "\t%.8f\t%.8f\t xlo xhi\n\t%.8f\t%.8f\t ylo yhi\n\t%.8f\t%.8f\t zlo zhi\n\t%.8f\t%.8f\t%.8f\t xy xz yz\n" % (
0., boxinfo[0][0], 0., boxinfo[1][1], 0., boxinfo[2][2],
boxinfo[1][0], boxinfo[2][0], boxinfo[2][1])
# print Coeffs
outCoeffs = "Pair Coeffs\n\n"
for i in range(1, nmassTypes + 1):
outCoeffs += "\t%d\teps\tsig\t# %s\n" % (i, mapTypes[i])
outCoeffs += "\nBond Coeffs\n\n"
for i in range(1, nbondTypes + 1):
outCoeffs += "\t%d\tK\tr_0\t# %s\n" % (i, mapbTypes[i])
outCoeffs += "\nAngle Coeffs\n\n"
for i in range(1, nangleTypes + 1):
outCoeffs += "\t%d\tK\ttetha_0 (deg)\t# %s\n" % (i, mapaTypes[i])
if printdih and (ndihedrals > 0):
outCoeffs += "\nDihedral Coeffs\n\n"
for i in range(1, ndihedralTypes + 1):
outCoeffs += "\t%d\tK\tn\tphi_0 (deg)\tw\t# %s\n" % (i,
mapdTypes[i])
if not ignoreimproper and (niDihedralTypes > 0):
outCoeffs += "\nImproper Coeffs\n\n"
for i in range(1, niDihedralTypes + 1):
outCoeffs += "\t%d\tK\txi_0 (deg)\t# %s\n" % (i, mapiDTypes[i])
if printdih and (ndihedrals > 0):
if ignoreimproper or (niDihedralTypes == 0):
return header + "\n" + outMasses + "\n" + outCoeffs + "\n" + outAtoms + "\n" + outBonds + "\n" + outAngles + "\n" + outDihedrals
else:
return header + "\n" + outMasses + "\n" + outCoeffs + "\n" + outAtoms + "\n" + outBonds + "\n" + outAngles + "\n" + outDihedrals + "\n" + outImpropers
else:
return header + "\n" + outMasses + "\n" + outCoeffs + "\n" + outAtoms + "\n" + outBonds + "\n" + outAngles
import openbabel
import sys
import pickle
#import plpy
def notice(msg):
print msg
plpy.notice(msg)
# functions used in atoms, bonds and fragments functions
tbl = openbabel.OBElementTable()
def nbr_smarts(a1, a2=None):
# construct smarts of each neighbor, possible excluding a2
smarts = [] # each neighbor as array element for later sort
for nbr in openbabel.OBAtomAtomIter(a1):
if nbr.GetAtomicNum() == 1: continue
#if a2 and nbr.GetIdx() == a2.GetIdx():
# continue
# include bond order if necessary (not single or aromatic)
b = a1.GetBond(nbr)
if b.IsSingle() and a1.IsAromatic():
bnd = "-"
elif b.IsDouble():
bnd = "="
elif b.IsTriple():
bnd = "#"
else:
class Molecule(object):
# for more rendering options visit:
# http://www.ggasoftware.com/opensource/indigo/api/options#rendering
_indigo = indigo.Indigo()
_renderer = indigo_renderer.IndigoRenderer(_indigo)
_indigo.setOption('render-output-format', 'svg')
_indigo.setOption('render-margins', 10, 10)
_indigo.setOption('render-stereo-style', 'none')
_indigo.setOption('render-implicit-hydrogens-visible', False)
_indigo.setOption('render-coloring', True)
_indigo.setOption('render-bond-length', 20.0)
_indigo.setOption('render-label-mode', 'hetero')
_obElements = openbabel.OBElementTable()
@staticmethod
def GetNumberOfElements():
return Molecule._obElements.GetNumberOfElements()
@staticmethod
def GetAllElements():
return [
Molecule._obElements.GetSymbol(i)
for i in xrange(Molecule.GetNumberOfElements())
]
@staticmethod
def GetSymbol(atomic_num):
return Molecule._obElements.GetSymbol(atomic_num)
@staticmethod
def GetAtomicNum(elem):
if type(elem) == types.UnicodeType:
elem = str(elem)
return Molecule._obElements.GetAtomicNum(elem)
@staticmethod
def SetBondLength(l):
Molecule._indigo.setOption('render-bond-length', l)
@staticmethod
def VerifySmarts(smarts):
try:
pybel.Smarts(smarts)
return True
except IOError:
return False
def __init__(self):
self.title = None
self.obmol = openbabel.OBMol()
self.pybel_mol = None
self.smiles = None
self.inchi = None
def __str__(self):
return self.title or self.smiles or self.inchi or ""
def __len__(self):
return self.GetNumAtoms()
def Clone(self):
tmp = Molecule()
tmp.title = self.title
tmp.obmol = openbabel.OBMol(self.obmol)
tmp.pybel_mol = pybel.Molecule(tmp.obmol)
tmp.smiles = self.smiles
tmp.inchi = self.inchi
return tmp
def SetTitle(self, title):
self.title = title
@staticmethod
def FromSmiles(smiles):
m = Molecule()
m.smiles = smiles
obConversion = openbabel.OBConversion()
obConversion.SetInFormat("smiles")
if not obConversion.ReadString(m.obmol, m.smiles):
raise OpenBabelError("Cannot read the SMILES string: " + smiles)
try:
m.UpdateInChI()
m.UpdatePybelMol()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from SMILES: " +
smiles)
m.SetTitle(smiles)
return m
@staticmethod
def FromInChI(inchi):
m = Molecule()
m.inchi = inchi
obConversion = openbabel.OBConversion()
obConversion.SetInFormat("inchi")
obConversion.ReadString(m.obmol, m.inchi)
try:
m.UpdateSmiles()
m.UpdatePybelMol()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from InChI: " +
inchi)
m.SetTitle(inchi)
return m
@staticmethod
def FromMol(mol):
m = Molecule()
obConversion = openbabel.OBConversion()
obConversion.SetInFormat("mol")
obConversion.ReadString(m.obmol, mol)
try:
m.UpdateInChI()
m.UpdateSmiles()
m.UpdatePybelMol()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from MOL file:\n" +
mol)
m.SetTitle("")
return m
@staticmethod
def FromOBMol(obmol):
m = Molecule()
m.obmol = obmol
try:
m.UpdateInChI()
m.UpdateSmiles()
m.UpdatePybelMol()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from OBMol")
m.SetTitle("")
return m
@staticmethod
def _FromFormat(s, fmt='inchi'):
if fmt == 'smiles' or fmt == 'smi':
return Molecule.FromSmiles(s)
if fmt == 'inchi':
return Molecule.FromInChI(s)
if fmt == 'mol':
return Molecule.FromMol(s)
if fmt == 'obmol':
return Molecule.FromOBMol(s)
@staticmethod
def _ToFormat(obmol, fmt='inchi'):
obConversion = openbabel.OBConversion()
obConversion.SetOutFormat(fmt)
res = obConversion.WriteString(obmol)
if not res:
raise OpenBabelError("Cannot convert OBMol to %s" % fmt)
if fmt == 'smiles' or fmt == 'smi':
res = res.split()
if res == []:
raise OpenBabelError("Cannot convert OBMol to %s" % fmt)
else:
return res[0]
elif fmt == 'inchi':
return res.strip()
else:
return res
@staticmethod
def Smiles2InChI(smiles):
obConversion = openbabel.OBConversion()
obConversion.SetInAndOutFormats("smiles", "inchi")
obmol = openbabel.OBMol()
if not obConversion.ReadString(obmol, smiles):
raise OpenBabelError("Cannot read the SMILES string: " + smiles)
return obConversion.WriteString(obmol).strip()
@staticmethod
def InChI2Smiles(inchi):
obConversion = openbabel.OBConversion()
obConversion.SetInAndOutFormats("inchi", "smiles")
obmol = openbabel.OBMol()
if not obConversion.ReadString(obmol, inchi):
raise OpenBabelError("Cannot read the InChI string: " + inchi)
return obConversion.WriteString(obmol).split()[0]
def RemoveHydrogens(self):
self.pybel_mol.removeh()
def RemoveAtoms(self, indices):
self.obmol.BeginModify()
for i in sorted(indices, reverse=True):
self.obmol.DeleteAtom(self.obmol.GetAtom(i + 1))
self.obmol.EndModify()
self.smiles = None
self.inchi = None
def SetAtomicNum(self, index, new_atomic_num):
self.obmol.GetAtom(index + 1).SetAtomicNum(new_atomic_num)
self.smiles = None
self.inchi = None
def ToOBMol(self):
return self.obmol
def ToPybelMol(self):
return self.pybel_mol
def ToFormat(self, fmt='inchi'):
return Molecule._ToFormat(self.obmol, fmt=fmt)
def ToMolfile(self):
return self.ToFormat('mol')
def UpdateInChI(self):
self.inchi = Molecule._ToFormat(self.obmol, 'inchi')
def ToInChI(self):
"""
Lazy storage of the InChI identifier (calculate once only when
asked for and store for later use).
"""
if not self.inchi:
self.UpdateInChI()
return self.inchi
def UpdateSmiles(self):
self.smiles = Molecule._ToFormat(self.obmol, 'smiles')
def ToSmiles(self):
"""
Lazy storage of the SMILES identifier (calculate once only when
asked for and store for later use).
"""
if not self.smiles:
self.UpdateSmiles()
return self.smiles
def UpdatePybelMol(self):
self.pybel_mol = pybel.Molecule(self.obmol)
@staticmethod
def _GetFormulaFromInChI(inchi):
tokens = re.findall('/f([0-9A-Za-z\.]+/)', inchi)
if len(tokens) == 0:
tokens = re.findall('InChI=1S?/([0-9A-Za-z\.]+)', inchi)
if len(tokens) == 1:
return tokens[0]
elif len(tokens) > 1:
raise ValueError('Bad InChI: ' + inchi)
else:
return ''
@staticmethod
def _GetAtomBagAndChargeFromInChI(inchi):
fixed_charge = 0
for q in re.findall('/q([0-9\+\-\;]+)', inchi):
for s in q.split(';'):
if s:
fixed_charge += int(s)
fixed_protons = 0
for p in re.findall('/p([0-9\+\-\;]+)', inchi):
for s in p.split(';'):
if s:
fixed_protons += int(s)
formula = Molecule._GetFormulaFromInChI(inchi)
atom_bag = {}
for mol_formula_times in formula.split('.'):
for times, mol_formula in re.findall('^(\d+)?(\w+)',
mol_formula_times):
if not times:
times = 1
else:
times = int(times)
for atom, count in re.findall("([A-Z][a-z]*)([0-9]*)",
mol_formula):
if count == '':
count = 1
else:
count = int(count)
atom_bag[atom] = atom_bag.get(atom, 0) + count * times
if fixed_protons:
atom_bag['H'] = atom_bag.get('H', 0) + fixed_protons
fixed_charge += fixed_protons
return atom_bag, fixed_charge
@staticmethod
def _GetNumElectronsFromInChI(inchi):
"""Calculates the number of electrons in a given molecule."""
atom_bag, fixed_charge = Molecule._GetAtomBagAndChargeFromInChI(inchi)
n_protons = 0
for elem, count in atom_bag.iteritems():
n_protons += count * Molecule._obElements.GetAtomicNum(elem)
return n_protons - fixed_charge
def GetFormula(self):
return Molecule._GetFormulaFromInChI(self.ToInChI())
def GetExactMass(self):
return self.obmol.GetExactMass()
def GetAtomBagAndCharge(self):
return Molecule._GetAtomBagAndChargeFromInChI(self.ToInChI())
def GetHydrogensAndCharge(self):
atom_bag, charge = self.GetAtomBagAndCharge()
return atom_bag.get('H', 0), charge
def GetNumElectrons(self):
return Molecule._GetNumElectronsFromInChI(self.ToInChI())
def GetNumAtoms(self):
return self.obmol.NumAtoms()
def GetAtoms(self):
return self.pybel_mol.atoms
def FindSmarts(self, smarts):
"""
Corrects the pyBel version of Smarts.findall() which returns results as tuples,
with 1-based indices even though Molecule.atoms is 0-based.
Args:
mol: the molecule to search in.
smarts_str: the SMARTS query to search for.
Returns:
The re-mapped list of SMARTS matches.
"""
if type(smarts) == types.StringType:
smarts = pybel.Smarts(smarts)
shift_left = lambda m: [(n - 1) for n in m]
return map(shift_left, smarts.findall(self.pybel_mol))
def ToSVG(self, comment=None):
if comment:
Molecule._indigo.setOption('render-comment', comment)
else:
Molecule._indigo.setOption('render-comment', '')
try:
indigo_mol = Molecule._indigo.loadMolecule(self.ToSmiles())
indigo_mol.aromatize()
indigo_mol.layout()
svg_str = Molecule._renderer.renderToBuffer(indigo_mol).tostring()
id = str(uuid.uuid4())
i = 0
while True:
symbol = 'glyph0-%d' % i
if svg_str.find('id="' + symbol + '"') != -1:
svg_str = svg_str.replace('id="' + symbol + '"',
'id="' + id + "_" + symbol + '"')
svg_str = svg_str.replace(
'href="#' + symbol + '"',
'href="#' + id + "_" + symbol + '"')
else:
break
i += 1
return svg_str
except indigo.IndigoException as e:
return "<b>Indigo error</b>: %s</br>\n" % str(e)
def Draw(self, show_title=False):
def expose_cairo(win, event, svg):
cr = win.window.cairo_create()
svg.render_cairo(cr)
return True
try:
if show_title:
svg = rsvg.Handle(data=self.ToSVG(self.title))
else:
svg = rsvg.Handle(data=self.ToSVG())
except glib.GError: #@UndefinedVariable
return
_x, _y, w, h = svg.get_dimension_data()
win = gtk.Window()
win.resize(int(w), int(h))
win.connect("delete-event", lambda w, e: gtk.main_quit())
win.connect("expose-event", expose_cairo, svg)
win.show_all()
win.connect("destroy", lambda w: gtk.main_quit())
gtk.main()
def GetAtomCharges(self):
"""
Returns:
A list of charges, according to the number of atoms
in the molecule
"""
return [atom.formalcharge for atom in self.pybel_mol.atoms]
@staticmethod
def _GetDissociationTable(molstring,
fmt='inchi',
mid_pH=default_pH,
min_pKa=0,
max_pKa=14,
T=default_T,
transform_multiples=False):
"""
Returns the relative potentials of pseudoisomers,
relative to the most abundant one at pH 7.
"""
from pygibbs.dissociation_constants import DissociationTable
from toolbox import chemaxon
diss_table = DissociationTable()
try:
pKa_table, major_ms = chemaxon.GetDissociationConstants(
molstring,
mid_pH=mid_pH,
transform_multiples=transform_multiples)
mol = Molecule.FromSmiles(major_ms)
nH, z = mol.GetHydrogensAndCharge()
diss_table.SetMolString(nH, nMg=0, s=major_ms)
diss_table.SetCharge(nH, z, nMg=0)
pKa_higher = [x for x in pKa_table if mid_pH < x[0] < max_pKa]
pKa_lower = [x for x in pKa_table if mid_pH > x[0] > min_pKa]
for i, (pKa, _, smiles_above) in enumerate(sorted(pKa_higher)):
diss_table.AddpKa(pKa,
nH_below=(nH - i),
nH_above=(nH - i - 1),
nMg=0,
ref='ChemAxon',
T=T)
diss_table.SetMolString((nH - i - 1), nMg=0, s=smiles_above)
for i, (pKa, smiles_below,
_) in enumerate(sorted(pKa_lower, reverse=True)):
diss_table.AddpKa(pKa,
nH_below=(nH + i + 1),
nH_above=(nH + i),
nMg=0,
ref='ChemAxon',
T=T)
diss_table.SetMolString((nH + i + 1), nMg=0, s=smiles_below)
except chemaxon.ChemAxonError:
mol = Molecule._FromFormat(molstring, fmt)
diss_table.SetOnlyPseudoisomerMolecule(mol)
return diss_table
def GetDissociationTable(self,
fmt='inchi',
mid_pH=default_pH,
min_pKa=0,
max_pKa=14,
T=default_T):
"""
Returns the relative potentials of pseudoisomers,
relative to the most abundant one at pH 7.
"""
return Molecule._GetDissociationTable(self.ToInChI(), 'inchi', mid_pH,
min_pKa, max_pKa, T)
def __init__(self):
self.etab = openbabel.OBElementTable()
self.non_ad_metal_names = [
"Cu",
"Fe",
"Na",
"K",
"Hg",
"Co",
"U",
"Cd",
"Ni",
"Si",
]
self.atom_equivalence_data = [("Se", "S")]
self.atom_type_data = [
Info(
"Hydrogen",
"Hydrogen",
"H",
1,
1.000000,
0.020000,
0.000510,
0.000000,
0.370000,
0.000000,
False,
False,
False,
False,
),
Info(
"PolarHydrogen",
"PolarHydrogen",
"HD",
1,
1.000000,
0.020000,
0.000510,
0.000000,
0.370000,
0.000000,
False,
False,
False,
False,
),
Info(
"AliphaticCarbonXSHydrophobe",
"AliphaticCarbonXSHydrophobe",
"C",
6,
2.000000,
0.150000,
-0.001430,
33.510300,
0.770000,
1.900000,
True,
False,
False,
False,
),
Info(
"AliphaticCarbonXSNonHydrophobe",
"AliphaticCarbonXSNonHydrophobe",
"C",
6,
2.000000,
0.150000,
-0.001430,
33.510300,
0.770000,
1.900000,
False,
False,
False,
False,
),
Info(
"AromaticCarbonXSHydrophobe",
"AromaticCarbonXSHydrophobe",
"A",
6,
2.000000,
0.150000,
-0.000520,
33.510300,
0.770000,
1.900000,
True,
False,
False,
False,
),
Info(
"AromaticCarbonXSNonHydrophobe",
"AromaticCarbonXSNonHydrophobe",
"A",
6,
2.000000,
0.150000,
-0.000520,
33.510300,
0.770000,
1.900000,
False,
False,
False,
False,
),
Info(
"Nitrogen",
"Nitrogen",
"N",
7,
1.750000,
0.160000,
-0.001620,
22.449300,
0.750000,
1.800000,
False,
False,
False,
True,
),
Info(
"NitrogenXSDonor",
"NitrogenXSDonor",
"N",
7,
1.750000,
0.160000,
-0.001620,
22.449300,
0.750000,
1.800000,
False,
True,
False,
True,
),
Info(
"NitrogenXSDonorAcceptor",
"NitrogenXSDonorAcceptor",
"NA",
7,
1.750000,
0.160000,
-0.001620,
22.449300,
0.750000,
1.800000,
False,
True,
True,
True,
),
Info(
"NitrogenXSAcceptor",
"NitrogenXSAcceptor",
"NA",
7,
1.750000,
0.160000,
-0.001620,
22.449300,
0.750000,
1.800000,
False,
False,
True,
True,
),
Info(
"Oxygen",
"Oxygen",
"O",
8,
1.600000,
0.200000,
-0.002510,
17.157300,
0.730000,
1.700000,
False,
False,
False,
True,
),
Info(
"OxygenXSDonor",
"OxygenXSDonor",
"O",
8,
1.600000,
0.200000,
-0.002510,
17.157300,
0.730000,
1.700000,
False,
True,
False,
True,
),
Info(
"OxygenXSDonorAcceptor",
"OxygenXSDonorAcceptor",
"OA",
8,
1.600000,
0.200000,
-0.002510,
17.157300,
0.730000,
1.700000,
False,
True,
True,
True,
),
Info(
"OxygenXSAcceptor",
"OxygenXSAcceptor",
"OA",
8,
1.600000,
0.200000,
-0.002510,
17.157300,
0.730000,
1.700000,
False,
False,
True,
True,
),
Info(
"Sulfur",
"Sulfur",
"S",
16,
2.000000,
0.200000,
-0.002140,
33.510300,
1.020000,
2.000000,
False,
False,
False,
True,
),
Info(
"SulfurAcceptor",
"SulfurAcceptor",
"SA",
16,
2.000000,
0.200000,
-0.002140,
33.510300,
1.020000,
2.000000,
False,
False,
False,
True,
),
Info(
"Phosphorus",
"Phosphorus",
"P",
15,
2.100000,
0.200000,
-0.001100,
38.792400,
1.060000,
2.100000,
False,
False,
False,
True,
),
Info(
"Fluorine",
"Fluorine",
"F",
9,
1.545000,
0.080000,
-0.001100,
15.448000,
0.710000,
1.500000,
True,
False,
False,
True,
),
Info(
"Chlorine",
"Chlorine",
"Cl",
17,
2.045000,
0.276000,
-0.001100,
35.823500,
0.990000,
1.800000,
True,
False,
False,
True,
),
Info(
"Bromine",
"Bromine",
"Br",
35,
2.165000,
0.389000,
-0.001100,
42.566100,
1.140000,
2.000000,
True,
False,
False,
True,
),
Info(
"Iodine",
"Iodine",
"I",
53,
2.360000,
0.550000,
-0.001100,
55.058500,
1.330000,
2.200000,
True,
False,
False,
True,
),
Info(
"Magnesium",
"Magnesium",
"Mg",
12,
0.650000,
0.875000,
-0.001100,
1.560000,
1.300000,
1.200000,
False,
True,
False,
True,
),
Info(
"Manganese",
"Manganese",
"Mn",
25,
0.650000,
0.875000,
-0.001100,
2.140000,
1.390000,
1.200000,
False,
True,
False,
True,
),
Info(
"Zinc",
"Zinc",
"Zn",
30,
0.740000,
0.550000,
-0.001100,
1.700000,
1.310000,
1.200000,
False,
True,
False,
True,
),
Info(
"Calcium",
"Calcium",
"Ca",
20,
0.990000,
0.550000,
-0.001100,
2.770000,
1.740000,
1.200000,
False,
True,
False,
True,
),
Info(
"Iron",
"Iron",
"Fe",
26,
0.650000,
0.010000,
-0.001100,
1.840000,
1.250000,
1.200000,
False,
True,
False,
True,
),
Info(
"GenericMetal",
"GenericMetal",
"M",
0,
1.200000,
0.000000,
-0.001100,
22.449300,
1.750000,
1.200000,
False,
True,
False,
True,
),
# note AD4 doesn't have boron, so copying from carbon
Info(
"Boron",
"Boron",
"B",
5,
2.04,
0.180000,
-0.0011,
12.052,
0.90,
1.920000,
True,
False,
False,
False,
),
]
self.atom_types = [info.sm for info in self.atom_type_data]
def validate_bpform_bonds(form_type):
""" Validate bonds in alphabet
Args:
form_type (:obj:`type`): type of BpForm
Raises:
:obj:`ValueError`: if any of the bonds are invalid
"""
form = form_type()
element_table = openbabel.OBElementTable()
errors = []
# validate bonds to backbone
atom_types = [
['backbone', 'monomer_bond_atoms'],
['backbone', 'monomer_displaced_atoms'],
['bond', 'l_bond_atoms'],
['bond', 'r_bond_atoms'],
['bond', 'l_displaced_atoms'],
['bond', 'r_displaced_atoms'],
]
for molecule_md, atom_type in atom_types:
molecule = getattr(form, molecule_md)
selected_hydrogens = []
for atom_md in getattr(molecule, atom_type):
if atom_md.molecule == core.Backbone:
if form.backbone.structure:
n_backbone_atoms = form.backbone.structure.NumAtoms()
else:
n_backbone_atoms = 0
if atom_md.position < 1 or atom_md.position > n_backbone_atoms:
errors.append('Invalid position {} for {}.{}'.format(
atom_md.position, molecule_md, atom_type))
continue
atom = form.backbone.structure.GetAtom(atom_md.position)
if atom_md.element == 'H' and atom.GetAtomicNum() != 1:
atom = core.get_hydrogen_atom(atom, selected_hydrogens,
None)
if atom is None:
continue
if element_table.GetSymbol(
atom.GetAtomicNum()) != atom_md.element:
errors.append(
'Invalid element {} != {} at position {} for {}.{}'.
format(element_table.GetSymbol(atom.GetAtomicNum()),
atom_md.element, atom_md.position, molecule_md,
atom_type))
# validate bonds to monomer
atom_types = [
'backbone_bond_atoms',
'backbone_displaced_atoms',
'r_bond_atoms',
'l_bond_atoms',
'r_displaced_atoms',
'l_displaced_atoms',
]
for i_monomer, monomer in enumerate(form.alphabet.monomers.values()):
for atom_type in atom_types:
selected_hydrogens = []
for atom_md in getattr(monomer, atom_type):
if atom_md.molecule == core.Monomer:
if atom_md.position < 1 or atom_md.position > monomer.structure.NumAtoms(
):
errors.append(
'Invalid position {} for monomeric form:{} {}'.
format(atom_md.position, monomer.id, atom_type))
continue
atom = monomer.structure.GetAtom(atom_md.position)
if atom_md.element == 'H' and atom.GetAtomicNum() != 1:
atom = core.get_hydrogen_atom(atom, selected_hydrogens,
i_monomer)
if atom is None:
continue
if element_table.GetSymbol(
atom.GetAtomicNum()) != atom_md.element:
errors.append(
'Invalid element {} != {} at position {} for monomeric form:{} {}'
.format(
element_table.GetSymbol(atom.GetAtomicNum()),
atom_md.element, atom_md.position, monomer.id,
atom_type))
# validate monomeric forms and dimers
for monomer in form.alphabet.monomers.values():
monomer_form = form_type(seq=[monomer])
try:
monomer_structure = monomer_form.get_structure()[0]
if monomer_form.get_formula() != OpenBabelUtils.get_formula(
monomer_structure):
errors.append(
'Monomeric form of {} has incorrect formula: {} != {}'.
format(monomer.id, str(monomer_form.get_formula()),
str(OpenBabelUtils.get_formula(monomer_structure))))
continue
if monomer_form.get_charge() != monomer_structure.GetTotalCharge():
errors.append(
'Monomeric form of {} has incorrect charge: {} != {}'.
format(monomer.id, monomer_form.get_charge(),
monomer_structure.GetTotalCharge()))
continue
OpenBabelUtils.export(monomer_structure, 'smiles')
OpenBabelUtils.export(monomer_structure, 'inchi')
except Exception as error:
errors.append(
'Unable to create monomeric form of {}:\n {}'.format(
monomer.id, str(error)))
if form.can_monomer_bond_left(monomer) and form.can_monomer_bond_right(
monomer):
dimer_form = form_type(seq=[monomer, monomer])
try:
dimer_structure = dimer_form.get_structure()[0]
if dimer_form.get_formula() != OpenBabelUtils.get_formula(
dimer_structure):
errors.append(
'Dimer of {} has incorrect formula: {} != {}'.format(
monomer.id, str(dimer_form.get_formula()),
str(OpenBabelUtils.get_formula(dimer_structure))))
continue
if dimer_form.get_charge() != dimer_structure.GetTotalCharge():
errors.append(
'Dimer of {} has incorrect charge: {} != {}'.format(
monomer.id, dimer_form.get_charge(),
dimer_structure.GetTotalCharge()))
continue
OpenBabelUtils.export(dimer_structure, 'smiles')
OpenBabelUtils.export(dimer_structure, 'inchi')
except Exception as error:
errors.append('Unable to form dimer of {}:\n {}'.format(
monomer.id, str(error)))
# report errors
if errors:
raise ValueError('BpForm {} is invalid:\n {}'.format(
form_type.__name__, '\n '.join(errors)))
def itp_from_params(mol, q, eps, sig, dfrBonds, dfrAngles, dfrDihedrals,
dfrImpDih):
# table to convert atomic number to symbols
if not ob3:
etab = openbabel.OBElementTable()
# !!! units are converted as the reverse of: http://chembytes.wikidot.com/oplsaagro2tnk and based on GROMACS manual
# write header
fcontent = """
;
; Generated by dice2gromacs
; https://github.com/hmcezar/dicetools
;
[ atomtypes ]
;name bond_type mass charge ptype sigma epsilon
"""
# write the atomtypes
for i, atom in enumerate(mol.atoms):
if ob3:
fcontent += "att_%03d %s%03d %7.4f 0.000 A %.5e %.5e\n" % (
i + 1, openbabel.GetSymbol(atom.atomicnum), i + 1,
atom.atomicmass, a2nm(sig[i]), cal2j(eps[i]))
else:
fcontent += "att_%03d %s%03d %7.4f 0.000 A %.5e %.5e\n" % (
i + 1, etab.GetSymbol(atom.atomicnum), i + 1, atom.atomicmass,
a2nm(sig[i]), cal2j(eps[i]))
fcontent += """
[ moleculetype ]
;name nrexcl
UNL 3
[ atoms ]
; nr type resi res atom cgnr charge mass
"""
# write the atoms
for i, atom in enumerate(mol.atoms):
if ob3:
fcontent += "%6d att_%03d 1 UNL %s%03d 1 %.4f %7.4f\n" % (
i + 1, i + 1, openbabel.GetSymbol(
atom.atomicnum), i + 1, q[i], atom.atomicmass)
else:
fcontent += "%6d att_%03d 1 UNL %s%03d 1 %.4f %7.4f\n" % (
i + 1, i + 1, etab.GetSymbol(
atom.atomicnum), i + 1, q[i], atom.atomicmass)
# write the bonds
fcontent += """
[ bonds ]
; ai aj funct r k
"""
for bnd in dfrBonds:
ai, aj = [int(x) for x in bnd.split()]
fcontent += "%6d %6d 1 %.4f %.4f\n" % (
ai, aj, a2nm(dfrBonds[bnd][1]), cal2j(dfrBonds[bnd][0]) * 200.0)
# write the angles
fcontent += """
[ angles ]
; ai aj ak funct theta cth
"""
for ang in dfrAngles:
ai, aj, ak = [int(x) for x in ang.split()]
fcontent += "%6d %6d %6d 1 %.4f %.4f\n" % (
ai, aj, ak, dfrAngles[ang][1], cal2j(dfrAngles[ang][0]) * 2.0)
# write the proper dihedrals
fcontent += """
[ dihedrals ]
; proper dihedrals - converted to the RB form from Fourier type if OPLS
; ai aj ak al func params
"""
fimp = ""
for dih in dfrDihedrals:
ai, aj, ak, al = [int(x) for x in dih.split()]
if dfrDihedrals[dih][0].lower() == "amber":
# check if it's a proper or improper dihedral
bondIterator = openbabel.OBMolBondIter(mol.OBMol)
cnt = 0
for bond in bondIterator:
if ((ai == bond.GetBeginAtom().GetId() + 1) and
(aj == bond.GetEndAtom().GetId() + 1)) or (
(aj == bond.GetBeginAtom().GetId() + 1) and
(ai == bond.GetEndAtom().GetId() + 1)):
cnt += 1
elif ((aj == bond.GetBeginAtom().GetId() + 1) and
(ak == bond.GetEndAtom().GetId() + 1)) or (
(ak == bond.GetBeginAtom().GetId() + 1) and
(aj == bond.GetEndAtom().GetId() + 1)):
cnt += 1
elif ((ak == bond.GetBeginAtom().GetId() + 1) and
(al == bond.GetEndAtom().GetId() + 1)) or (
(al == bond.GetBeginAtom().GetId() + 1) and
(ak == bond.GetEndAtom().GetId() + 1)):
cnt += 1
if cnt == 3:
func = 9
fparam = [
cal2j(float(x)) / 2.0 for x in dfrDihedrals[dih][1:4]
]
for i, term in enumerate(fparam, 1):
if term != 0.0:
fcontent += "%6d %6d %6d %6d %1d %6.2f %9.5f %d\n" % (
ai, aj, ak, al, func,
float(dfrDihedrals[dih][i + 3]), term, i)
else:
func = 4
fparam = [
cal2j(float(x)) / 2.0 for x in dfrDihedrals[dih][1:4]
]
for i, term in enumerate(fparam, 1):
if term != 0.0:
fimp += "%6d %6d %6d %6d %1d %6.2f %9.5f %d\n" % (
ai, aj, ak, al, func,
float(dfrDihedrals[dih][i + 3]), term, i)
elif dfrDihedrals[dih][0].lower() == "opls":
fparam = [cal2j(float(x)) for x in dfrDihedrals[dih][1:4]]
c0 = fparam[1] + 0.5 * (fparam[0] + fparam[2])
c1 = 0.5 * (-fparam[0] + 3.0 * fparam[2])
c2 = -fparam[1]
c3 = -2.0 * fparam[2]
c4 = 0.0
c5 = 0.0
fcontent += "%6d %6d %6d %6d 3 %9.5f %9.5f %9.5f %9.5f %9.5f %9.5f\n" % (
ai, aj, ak, al, c0, c1, c2, c3, c4, c5)
else:
print(
"Error: Dihedral type (%s) found for dihedral %s in .dfr is not valid."
% (dfrDihedrals[dih][0], dih))
sys.exit(0)
# write the improper dihedrals
if fimp or dfrImpDih:
fcontent += """
[ dihedrals ]
; improper dihedrals
; ai aj ak al func params
"""
if fimp:
fcontent += fimp
for idih in dfrImpDih:
ai, aj, ak, al = [int(x) for x in idih.split()]
fcontent += "%6d %6d %6d %6d 4 %6.2f %9.5f %d\n" % (
ai, aj, ak, al, float(
dfrImpDih[idih][1]), cal2j(float(dfrImpDih[idih][0])), 2)
# write the pairs
fcontent += """
[ pairs ]
"""
pairs = get_pairs(mol)
for pair in pairs:
ai, aj = [int(x) for x in pair.split()]
fcontent += "%6d %6d 1\n" % (ai, aj)
return fcontent
def top2dfr(topfile, geomfile, flexfrag, eqgeom, savefrags, topcharges, ffname,
path):
if "amber" in ffname:
potname = "AMBER"
else:
potname = "OPLS"
# get the atomic positions from the geometry file
base, ext = os.path.splitext(geomfile)
etab = openbabel.OBElementTable()
mol = pybel.readfile(ext[1:], geomfile).__next__()
molxyzinfo = {}
for i, atom in enumerate(mol, 1):
molxyzinfo[i] = [etab.GetSymbol(atom.atomicnum), atom.coords]
# first pass through the topology file to get the data into a dictionary
tdata = {}
tdata["[ improper ]"] = []
with open(topfile) as f:
while True:
line = f.readline()
if not line:
break
if line.strip().startswith(";") or line.strip().startswith(
"#") or len(line.strip()) == 0:
continue
if strip_comment(line).startswith("[ "):
if strip_comment(line) not in tdata:
key = strip_comment(line)
tdata[key] = []
else:
# remove comments from the line
if "improper" not in line:
line = strip_comment(line)
if "improper" in line:
tdata["[ improper ]"].append(line)
else:
try:
tdata[key].append(line)
except:
print(
"You have a line (%s) of data before assigning a type of entry (such as [ atoms ], [ bonds ] ...)"
% (line))
sys.exit(0)
# check if atoms were found
if "[ atoms ]" not in tdata:
print(
"The [ atoms ] section was not found in your topology, make sure you're using a single file without #include"
)
sys.exit(0)
# get the atoms and its positions and parameters in a dictionary
atoms = OrderedDict()
rdfs = {}
atom_num = 1
rdf_label = 1
for line in tdata["[ atoms ]"]:
atomlbl = line.split()[0]
atoms[atomlbl] = []
fromatomtype = True
# if atomtypes are in the topology, use them. Otherwise get from ffnonbonded.itp
ffline = ""
if "[ atomtypes ]" in tdata:
ffline = lookup_ljparam_ifile(line.split()[1],
tdata["[ atomtypes ]"])
# get the data from the ffnonbonded.itp
if not ffline:
fromatomtype = False
ffline = lookup_ljparam(line.split()[1], path)
# if the atomtype was not found, stop
if not ffline:
print(
"Atom type %s was not found neither in the .itp or the force field directory"
% (line.split()[1]))
sys.exit(0)
# append the data to the list in the same order it will be written in the .dfr
atoms[atomlbl].append(str(atom_num))
atom_num += 1
if "amber" in ffname:
atomsp = ffline.split()[1]
else:
atomsp = ffline.split()[2]
if atomsp not in rdfs:
rdfs[atomsp] = str(rdf_label)
rdf_label += 1
atoms[atomlbl].append(rdfs[atomsp])
if "[ atomtypes ]" in tdata:
atoms[atomlbl].append(mol.atoms[int(atomlbl) - 1].atomicnum)
else:
atoms[atomlbl].append(atomsp)
x, y, z = molxyzinfo[int(atomlbl)][1]
atoms[atomlbl].append(str(x))
atoms[atomlbl].append(str(y))
atoms[atomlbl].append(str(z))
if ("amber" in ffname) or fromatomtype:
if topcharges:
atoms[atomlbl].append(line.split()[6])
else:
atoms[atomlbl].append(ffline.split()[-4])
atoms[atomlbl].append(str(j2cal(float(ffline.split()[-1]))))
atoms[atomlbl].append(str(nm2a(float(ffline.split()[-2]))))
# the last one will not be printed but is needed to retrieve the force constants
if ("[ atomtypes ]" in tdata) and ("opls" in ffname):
atoms[atomlbl].append(ffline.split()[1])
else:
atoms[atomlbl].append(ffline.split()[0])
else:
if topcharges:
atoms[atomlbl].append(line.split()[6])
else:
atoms[atomlbl].append(ffline.split()[4])
atoms[atomlbl].append(str(j2cal(float(ffline.split()[7]))))
atoms[atomlbl].append(str(nm2a(float(ffline.split()[6]))))
# the last one will not be printed but is needed to retrieve the opls-aa force constants
atoms[atomlbl].append(ffline.split()[1])
# print(atoms[atomlbl])
# now create the fragment connection list from this file and store it in fraginfo
# creates a temporary xyz using mkstemp (https://www.logilab.org/blogentry/17873)
# this file will be used to get the fragment data
fd, temp_path = tempfile.mkstemp(suffix=".xyz")
fxyz = os.fdopen(fd, 'w')
fxyz.write(mol.write("xyz"))
fxyz.close()
generate_fragfile(temp_path, "header")
base, ext = os.path.splitext(temp_path)
fraginfo = []
with open(base + ".dfr") as f:
while True:
line = f.readline()
if line.strip() != "$atoms fragments":
continue
else:
while line.strip() != "$end fragment connection":
if flexfrag:
fraginfo.append(line.replace("R", "M"))
else:
fraginfo.append(line)
line = f.readline()
fraginfo.append(line)
break
# remove the files
os.remove(temp_path)
os.remove(base + ".dfr")
os.remove(base + ".txt")
if savefrags:
shutil.move(
base + "_fragments",
os.path.join(
os.path.dirname(os.path.abspath(geomfile)),
os.path.splitext(os.path.basename(topfile))[0] + "_fragments"))
else:
shutil.rmtree(base + "_fragments")
# !!! units should be converted as in: http://chembytes.wikidot.com/oplsaagro2tnk !!!
# get the bond info
bonds = []
for line in tdata["[ bonds ]"]:
# get parameters from user's .itp
if (len(line.split()) == 5):
if eqgeom:
bonds.append(
line.split()[0] + " " + line.split()[1] + " \t" +
str(round(j2cal(float(line.split()[4])) / (200.0), 4)) +
"\t" + str(
round(
mol.OBMol.GetBond(int(line.split(
)[0]), int(line.split()[1])).GetLength(), 4)) +
"\n")
else:
bonds.append(
line.split()[0] + " " + line.split()[1] + " \t" +
str(round(j2cal(float(line.split()[4])) / (200.0), 4)) +
"\t" + str(nm2a(float(line.split()[3]))) + "\n")
# get parameters from ffbonded.itp
else:
ffline = lookup_ffbond(atoms[line.split()[0]][9],
atoms[line.split()[1]][9], path)
if ffline == "not found":
bonds.append(
line.split()[0] + " " + line.split()[1] + " \tXXX\t" +
str(
round(
mol.OBMol.GetBond(int(line.split(
)[0]), int(line.split()[1])).GetLength(), 4)) +
"\n")
elif eqgeom:
bonds.append(
line.split()[0] + " " + line.split()[1] + " \t" +
str(round(j2cal(float(ffline.split()[4])) / (200.0), 4)) +
"\t" + str(
round(
mol.OBMol.GetBond(int(line.split(
)[0]), int(line.split()[1])).GetLength(), 4)) +
"\n")
else:
bonds.append(
line.split()[0] + " " + line.split()[1] + " \t" +
str(round(j2cal(float(ffline.split()[4])) / (200.0), 4)) +
"\t" + str(nm2a(float(ffline.split()[3]))) + "\n")
# get the angle info
angles = []
for line in tdata["[ angles ]"]:
# get parameters from user's .itp
if (len(line.split()) == 6):
if eqgeom:
angles.append(
line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " \tharmonic\t" +
str(j2cal(float(line.split()[5])) / (2.0)) + "\t" + str(
round(
mol.OBMol.GetAngle(
mol.OBMol.GetAtom(int(line.split()[0])),
mol.OBMol.GetAtom(int(line.split()[1])),
mol.OBMol.GetAtom(int(line.split()[2]))), 4)) +
"\n")
else:
angles.append(line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " \tharmonic\t" +
str(j2cal(float(line.split()[5])) / (2.0)) +
"\t" + str(float(line.split()[4])) + "\n")
# get parameters from ffbonded.itp
else:
ffline = lookup_ffangle(atoms[line.split()[0]][9],
atoms[line.split()[1]][9],
atoms[line.split()[2]][9], path)
if ffline == "not found":
angles.append(
line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " \tharmonic\tXXX\t" + str(
round(
mol.OBMol.GetAngle(
mol.OBMol.GetAtom(int(line.split()[0])),
mol.OBMol.GetAtom(int(line.split()[1])),
mol.OBMol.GetAtom(int(line.split()[2]))), 4)) +
"\n")
elif eqgeom:
angles.append(
line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " \tharmonic\t" +
str(j2cal(float(ffline.split()[5])) / (2.0)) + "\t" + str(
round(
mol.OBMol.GetAngle(
mol.OBMol.GetAtom(int(line.split()[0])),
mol.OBMol.GetAtom(int(line.split()[1])),
mol.OBMol.GetAtom(int(line.split()[2]))), 4)) +
"\n")
else:
angles.append(line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " \tharmonic\t" +
str(j2cal(float(ffline.split()[5])) /
(2.0)) + "\t" + ffline.split()[4] + "\n")
# get the dihedrals info
dihedrals = []
pline = {}
ipline = {}
dih9 = False
dih4 = False
for rline in tdata["[ dihedrals ]"]:
line = strip_comment(rline)
ffline = ""
# get parameters from user's .itp
if (len(line.split()) == 11 and line.split()[4] == '3'):
V3 = round(-j2cal(float(line.split()[8]) / 2.0), 3)
V2 = round(-j2cal(float(line.split()[7])), 3)
V1 = round(-2.0 * j2cal(float(line.split()[6])) + 3.0 * V3, 3)
dihedrals.append(line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " " + line.split()[3] +
" \t" + potname + "\t" + str(V1) + "\t" +
str(V2) + "\t" + str(V3) + "\t0.0\t0.0\t0.0\n")
elif (len(line.split()) == 8 and line.split()[4] == '9'):
dih9 = True
dihline = "%s %s %s %s" % (line.split()[0], line.split()[1],
line.split()[2], line.split()[3])
if dihline in pline:
pline[dihline].append(line)
else:
pline[dihline] = [line]
elif (len(line.split()) == 8
and (line.split()[4] == '4' or line.split()[4] == '1')):
dih4 = True
dihline = "%s %s %s %s" % (line.split()[0], line.split()[1],
line.split()[2], line.split()[3])
if dihline in ipline:
ipline[dihline].append(line)
else:
ipline[dihline] = [line]
# get parameters from ffbonded.itp
elif len(line.split()) == 5:
ffline = lookup_ffdihedral(atoms[line.split()[0]][9],
atoms[line.split()[1]][9],
atoms[line.split()[2]][9],
atoms[line.split()[3]][9],
int(line.split()[4]), ffname, path)
else:
print(
"Error: something is wrong in dihedral line (%s) maybe the number of parameters?"
% (line))
sys.exit(0)
# parameters from ffbonded.itp need to be converted and stored
if ffline:
if ffline == "not found":
dihedrals.append(line.split()[0] + " " + line.split()[1] +
" " + line.split()[2] + " " +
line.split()[3] + " \t" + potname +
"\tXXX\tXXX\tXXX" + "\t0.0\t0.0\t0.0\n")
continue
if "amber" in ffname:
# parameters are already of Fourier type, just need to convert to cal
if float(ffline.split()[8]) == 0.:
dihedrals.append(
line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " " + line.split()[3] + " \t" +
potname + "\t" + str(j2cal(float(ffline.split()[5]))) +
"\t" + str(j2cal(float(ffline.split()[6]))) + "\t" +
str(j2cal(float(ffline.split()[7]))) + "\t" +
str(round(float(ffline.split()[9]), 1)) + "\t" +
str(round(float(ffline.split()[10]), 1)) + "\t" +
str(round(float(ffline.split()[11]), 1)) + "\n")
else:
dihedrals.append(
line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " " + line.split()[3] + " \t" +
potname + "\t" + str(j2cal(float(ffline.split()[5]))) +
"\t" + str(j2cal(float(ffline.split()[6]))) + "\t" +
str(j2cal(float(ffline.split()[7]))) + "\t" +
str(j2cal(float(ffline.split()[8]))) + "\t" +
str(round(float(ffline.split()[9]), 1)) + "\t" +
str(round(float(ffline.split()[10]), 1)) + "\t" +
str(round(float(ffline.split()[11]), 1)) + "\t" +
str(round(float(ffline.split()[12]), 1)) + "\n")
else:
if float(ffline.split()[9]) != 0.0 or float(
ffline.split()[10]) != 0.0:
print(
"Parameters for %s - %s - %s - %s dihedrals are undefined, please treat by hand!"
%
(atoms[line.split()[0]][9], atoms[line.split()[1]][9],
atoms[line.split()[2]][9], atoms[line.split()[3]][9]))
V3 = round(-j2cal(float(ffline.split()[8]) / 2.0), 3)
V2 = round(-j2cal(float(ffline.split()[7])), 3)
V1 = round(-2.0 * j2cal(float(ffline.split()[6])) + 3.0 * V3,
3)
dihedrals.append(line.split()[0] + " " + line.split()[1] +
" " + line.split()[2] + " " +
line.split()[3] + " \t" + potname + "\t" +
str(V1) + "\t" + str(V2) + "\t" + str(V3) +
"\t0.0\t0.0\t0.0\n")
# after reading all the dihedrals, if type 9 was used, we need to convert
if dih9:
for kdih in pline:
params = [0.0] * 6
for line in pline[kdih]:
n = int(line.split()[7])
# have to multiply by 2.0 since I use the 0.5*(...) version of the AMBER definition
params[n - 1] = 2.0 * float(line.split()[6])
params[n + 2] = float(line.split()[5])
retline = "%s\t%d\t%f\t%f\t%f\t%f\t%f\t%f" % (
kdih, 9, params[0], params[1], params[2], params[3], params[4],
params[5])
dihedrals.append(retline.split()[0] + " " + retline.split()[1] +
" " + retline.split()[2] + " " +
retline.split()[3] + " \t" + potname + "\t" +
str(j2cal(float(retline.split()[5]))) + "\t" +
str(j2cal(float(retline.split()[6]))) + "\t" +
str(j2cal(float(retline.split()[7]))) + "\t" +
retline.split()[8] + "\t" + retline.split()[9] +
"\t" + retline.split()[10] + "\n")
# finally, add the improper dihedrals described as proper dihedrals
if dih4:
for kdih in ipline:
params = [0.0] * 6
for line in ipline[kdih]:
n = int(line.split()[7])
# have to multiply by 2.0 since I use the 0.5*(...) version of the AMBER definition
params[n - 1] = 2.0 * float(line.split()[6])
params[n + 2] = float(line.split()[5])
retline = "%s\t%d\t%.3f\t%.3f\t%.3f\t%.2f\t%.2f\t%.2f" % (
kdih, 9, params[0], params[1], params[2], params[3], params[4],
params[5])
dihedrals.append(retline.split()[0] + " " + retline.split()[1] +
" " + retline.split()[2] + " " +
retline.split()[3] + " \t" + "AMBER" + "\t" +
str(j2cal(float(retline.split()[5]))) + "\t" +
str(j2cal(float(retline.split()[6]))) + "\t" +
str(j2cal(float(retline.split()[7]))) + "\t" +
retline.split()[8] + "\t" + retline.split()[9] +
"\t" + retline.split()[10] + "\n")
# get the improper dihedrals info
improper = []
for line in tdata["[ improper ]"]:
if len(strip_comment(line).split()) >= 7:
improper.append(line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " " + line.split()[3] + " \t" +
str(round(j2cal(float(line.split()[6])), 3)) +
"\t" + line.split()[5] + "\n")
else:
if "opls" in ffname:
ffline = lookup_ffimproper(line.split()[5], path)
improper.append(
line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " " + line.split()[3] + " \t" +
str(round(j2cal(float(ffline.split()[3])), 3)) + "\t" +
ffline.split()[2] + "\n")
else:
ffline = lookup_ffdihedral(atoms[line.split()[0]][9],
atoms[line.split()[1]][9],
atoms[line.split()[2]][9],
atoms[line.split()[3]][9], 4, path)
improper.append(line.split()[0] + " " + line.split()[1] + " " +
line.split()[2] + " " + line.split()[3] +
" \t" + "OPLS" + "\t" +
str(j2cal(float(ffline.split()[5]))) + "\t" +
str(j2cal(float(ffline.split()[6]))) + "\t" +
str(j2cal(float(ffline.split()[7]))) + "\t" +
ffline.split()[8] + "\t" + ffline.split()[9] +
"\t" + ffline.split()[10] + "\n")
# print everything to the output file
base, ext = os.path.splitext(geomfile)
with open(base + ".txt", "w") as f:
f.write("*\n1\n")
f.write(
str(len(atoms)) + " \t %s (generated with gromacs2dice)\n" %
(os.path.basename(base)))
for atom, data in atoms.items():
f.write(
"%2d %2d \t %7.4f \t %7.4f \t %7.4f \t %7.4f \t %7.4f \t %7.4f\n"
% (int(data[1]), int(data[2]), float(data[3]), float(
data[4]), float(data[5]), float(data[6]), float(
data[7]), float(data[8])))
f.write("$end\n")
with open(base + ".dfr", "w") as f:
for line in fraginfo:
f.write(line)
f.write("\n$bond\n")
for line in bonds:
f.write(line)
f.write("$end bond\n\n$angle\n")
for line in angles:
f.write(line)
f.write("$end angle\n\n$dihedral\n")
for line in dihedrals:
f.write(line)
if dih4:
for line in improper:
f.write(line)
f.write("$end dihedral\n")
if improper:
for line in improper:
print("\n$improper dihedral\n")
f.write(line)
f.write("$end improper dihedral\n")
if not flexfrag:
withoutba = clean_dofs(base + ".dfr")
with open(base + ".dfr", 'w') as f:
f.write(withoutba)
print("The files %s and %s were successfully generated." %
(base + ".txt", base + ".dfr"))
def draw(self, show=True, filename=None, update=False, usecoords=False,
method="mcdl"):
"""Create a 2D depiction of the molecule.
Optional parameters:
show -- display on screen (default is True)
filename -- write to file (default is None)
update -- update the coordinates of the atoms to those
determined by the structure diagram generator
(default is False)
usecoords -- don't calculate 2D coordinates, just use
the current coordinates (default is False)
method -- two methods are available for calculating the
2D coordinates: OpenBabel's "mcdl" (the default), or
"oasa" (from the OASA toolkit)
OASA is used for depiction. Tkinter and Python
Imaging Library are required for image display.
"""
etab = ob.OBElementTable()
if not oasa:
errormessage = ("OASA not found, but is required for depiction. "
"OASA is part of BKChem. "
"See installation instructions for more "
"information.")
raise ImportError(errormessage)
if method not in ["mcdl", "oasa"]:
raise ValueError("Method '%s' not recognised. Should be either"
" 'mcdl' or 'oasa'.")
workingmol = self
if method == "mcdl":
if not update: # Call gen2D on a clone
workingmol = Molecule(self)
if not usecoords:
_operations['gen2D'].Do(workingmol.OBMol)
usecoords = True # Use the workingmol's coordinates
mol = oasa.molecule()
for atom in workingmol.atoms:
v = mol.create_vertex()
v.symbol = etab.GetSymbol(atom.atomicnum)
v.charge = atom.formalcharge
if usecoords:
v.x, v.y, v.z = atom.coords[0] * 30., atom.coords[1] * 30., 0.0
mol.add_vertex(v)
for bond in ob.OBMolBondIter(workingmol.OBMol):
e = mol.create_edge()
e.order = bond.GetBO()
if bond.IsHash():
e.type = "h"
elif bond.IsWedge():
e.type = "w"
mol.add_edge(bond.GetBeginAtomIdx() - 1,
bond.GetEndAtomIdx() - 1,
e)
# I'm sure there's a more elegant way to do the following, but here goes...
# let's set the stereochemistry around double bonds
self.write("can") # Perceive UP/DOWNness
for bond in ob.OBMolBondIter(workingmol.OBMol):
ends = bond.GetBeginAtomIdx(), bond.GetEndAtomIdx()
if bond.GetBO() == 2:
stereobonds = [[b for b in ob.OBAtomBondIter(workingmol.OBMol.GetAtom(x)) if b.GetIdx() != bond.GetIdx() and (b.IsUp() or b.IsDown())]
for x in ends]
if stereobonds[0] and stereobonds[1]: # Needs to be defined at either end
if stereobonds[0][0].IsUp() == stereobonds[1][0].IsUp():
# Either both up or both down
stereo = oasa.stereochemistry.cis_trans_stereochemistry.SAME_SIDE
else:
stereo = oasa.stereochemistry.cis_trans_stereochemistry.OPPOSITE_SIDE
atomids = [(b[0].GetBeginAtomIdx(), b[0].GetEndAtomIdx()) for b in stereobonds]
extremes = []
for id, end in zip(ends, atomids):
if end[0] == id:
extremes.append(end[1])
else:
extremes.append(end[0])
center = mol.get_edge_between(mol.atoms[ends[0] - 1], mol.atoms[ends[1] - 1])
st = oasa.stereochemistry.cis_trans_stereochemistry(
center = center, value = stereo,
references = (mol.atoms[extremes[0] - 1], mol.atoms[ends[0] - 1],
mol.atoms[ends[1] - 1], mol.atoms[extremes[1] - 1]))
mol.add_stereochemistry(st)
mol.remove_unimportant_hydrogens()
if method == "oasa" and not usecoords:
oasa.coords_generator.calculate_coords(mol, bond_length=30)
if update:
newcoords = [(v.x / 30., v.y / 30., 0.0) for v in mol.vertices]
for atom, newcoord in zip(ob.OBMolAtomIter(self.OBMol), newcoords):
atom.SetVector(*newcoord)
if filename or show:
maxx = max([v.x for v in mol.vertices])
minx = min([v.x for v in mol.vertices])
maxy = max([v.y for v in mol.vertices])
miny = min([v.y for v in mol.vertices])
maxcoord = max(maxx - minx, maxy - miny)
fontsize = 16
bondwidth = 6
linewidth = 2
if maxcoord > 270: # 300 - margin * 2
for v in mol.vertices:
v.x *= 270. / maxcoord
v.y *= 270. / maxcoord
fontsize *= math.sqrt(270. / maxcoord)
bondwidth *= math.sqrt(270. / maxcoord)
linewidth *= math.sqrt(270. / maxcoord)
if filename:
filedes = None
else:
filedes, filename = tempfile.mkstemp()
canvas = oasa.cairo_out.cairo_out()
canvas.show_hydrogens_on_hetero = True
canvas.font_size = fontsize
canvas.bond_width = bondwidth
canvas.line_width = linewidth
canvas.mol_to_cairo(mol, filename)
if show:
if not tk:
errormessage = ("Tkinter or Python Imaging "
"Library not found, but is required for image "
"display. See installation instructions for "
"more information.")
raise ImportError(errormessage)
root = tk.Tk()
root.title((hasattr(self, "title") and self.title)
or self.__str__().rstrip())
frame = tk.Frame(root, colormap="new", visual='truecolor').pack()
image = PIL.open(filename)
imagedata = piltk.PhotoImage(image)
label = tk.Label(frame, image=imagedata).pack()
quitbutton = tk.Button(root, text="Close", command=root.destroy).pack(fill=tk.X)
root.mainloop()
if filedes:
os.close(filedes)
os.remove(filename)
#!/usr/bin/env python
# python modules
from omg import geom
import numpy as np
import openbabel as ob
#create a periodic table object
PERIODIC_TABLE = ob.OBElementTable()
class Atom(ob.OBAtom):
"""A container for all the information relative to an atom
Attributes:
type (str)
x (float)
y (float)
z (float)
Other info is kept in its own object inside atom:
- mm (molecular mechanics)
- oniom (oniom info)
- resinfo (information about the residue the atom belongs to)
- pdbinfo (information for pdb files)
"""
def __init__(self, atype, xyz):
"""
Args:
def __init__(self, add_hydrogens: bool = False) -> None:
self._add_hydrogens = add_hydrogens
self._edge_types = set()
self._node_types = set()
self._element_tabel = openbabel.OBElementTable()
class Compound(object):
_obElements = openbabel.OBElementTable()
def __init__(self, database, compound_id, inchi, pKas, majorMSpH7, nHs,
zs):
self.database = database
self.compound_id = compound_id
self.inchi = inchi
self.pKas = pKas
self.majorMSpH7 = majorMSpH7
self.nHs = nHs
self.zs = zs
@staticmethod
def from_kegg(cid):
inchi = Compound.get_inchi(cid)
if inchi is None:
pKas = []
majorMSpH7 = -1
nHs = []
zs = []
else:
pKas, majorMSpH7, nHs, zs = Compound.get_species_pka(inchi)
return Compound('KEGG', 'C%05d' % cid, inchi, pKas, majorMSpH7, nHs,
zs)
def to_json_dict(self):
return {
'database': self.database,
'id': self.compound_id,
'inchi': self.inchi,
'pKas': self.pKas,
'majorMSpH7': self.majorMSpH7,
'nHs': self.nHs,
'zs': self.zs
}
@staticmethod
def from_json_dict(d):
return Compound(d['database'], d['id'], d['inchi'], d['pKas'],
d['majorMSpH7'], d['nHs'], d['zs'])
@staticmethod
def get_inchi(cid):
s_mol = urllib.urlopen('http://rest.kegg.jp/get/cpd:C%05d/mol' %
cid).read()
return Compound.mol2inchi(s_mol)
@staticmethod
def mol2inchi(s):
openbabel.obErrorLog.SetOutputLevel(-1)
conv = openbabel.OBConversion()
conv.SetInAndOutFormats('mol', 'inchi')
conv.AddOption("F", conv.OUTOPTIONS)
conv.AddOption("T", conv.OUTOPTIONS)
conv.AddOption("x", conv.OUTOPTIONS, "noiso")
conv.AddOption("w", conv.OUTOPTIONS)
obmol = openbabel.OBMol()
if not conv.ReadString(obmol, s):
return None
inchi = conv.WriteString(obmol,
True) # second argument is trimWhitespace
if inchi == '':
return None
else:
return inchi
@staticmethod
def smiles2inchi(smiles):
openbabel.obErrorLog.SetOutputLevel(-1)
conv = openbabel.OBConversion()
conv.SetInAndOutFormats('smiles', 'inchi')
conv.AddOption("F", conv.OUTOPTIONS)
conv.AddOption("T", conv.OUTOPTIONS)
conv.AddOption("x", conv.OUTOPTIONS, "noiso")
conv.AddOption("w", conv.OUTOPTIONS)
obmol = openbabel.OBMol()
conv.ReadString(obmol, smiles)
inchi = conv.WriteString(obmol,
True) # second argument is trimWhitespace
if inchi == '':
return None
else:
return inchi
@staticmethod
def get_atom_bag_and_charge_from_inchi(inchi):
if inchi is None:
return {}, 0
fixed_charge = 0
for q in re.findall('/q([0-9\+\-\;]+)', inchi):
for s in q.split(';'):
if s:
fixed_charge += int(s)
atom_bag = {}
# the /f field gives the fixed-H structure
tokens = re.findall('/f([0-9A-Za-z\.]+/)', inchi)
# if /f is not given, use the main formula and
# adjust the number of protons according to the /p field
if len(tokens) == 0:
tokens = re.findall('InChI=1S?/([0-9A-Za-z\.]+)', inchi)
for p in re.findall('/p([0-9\+\-\;]+)', inchi):
for s in p.split(';'):
if s:
atom_bag['H'] = atom_bag.get('H', 0) + int(s)
if len(tokens) == 1:
formula = tokens[0]
elif len(tokens) > 1:
raise ValueError('Bad InChI: ' + inchi)
else:
formula = ''
for mol_formula_times in formula.split('.'):
for times, mol_formula in re.findall('^(\d+)?(\w+)',
mol_formula_times):
if not times:
times = 1
else:
times = int(times)
for atom, count in re.findall("([A-Z][a-z]*)([0-9]*)",
mol_formula):
if count == '':
count = 1
else:
count = int(count)
atom_bag[atom] = atom_bag.get(atom, 0) + count * times
return atom_bag, fixed_charge
@staticmethod
def get_species_pka(inchi):
if inchi is None:
return [], -1, [], []
try:
pKas, major_ms = GetDissociationConstants(inchi)
pKas = sorted(
[pka for pka in pKas if pka > MIN_PH and pka < MAX_PH],
reverse=True)
major_ms_inchi = Compound.smiles2inchi(major_ms)
except ChemAxonError:
logging.warning('chemaxon failed to find pKas for this inchi: ' +
inchi)
pKas = []
major_ms_inchi = inchi
atom_bag, major_ms_charge = Compound.get_atom_bag_and_charge_from_inchi(
major_ms_inchi)
major_ms_nH = atom_bag.get('H', 0)
n_species = len(pKas) + 1
if pKas == []:
majorMSpH7 = 0
else:
majorMSpH7 = len([1 for pka in pKas if pka > 7])
nHs = []
zs = []
for i in xrange(n_species):
zs.append((i - majorMSpH7) + major_ms_charge)
nHs.append((i - majorMSpH7) + major_ms_nH)
return pKas, majorMSpH7, nHs, zs
def __str__(self):
return "%s\nInChI: %s\npKas: %s\nmajor MS: nH = %d, charge = %d" % \
(self.compound_id, self.inchi, ', '.join(['%.2f' % p for p in self.pKas]),
self.nHs[self.majorMSpH7], self.zs[self.majorMSpH7])
def get_atom_bag_with_electrons(self):
"""
Calculates the number of electrons in a given molecule
Returns:
a dictionary of all element counts and also electron count ('e-')
"""
if self.inchi is None:
return None
atom_bag, charge = Compound.get_atom_bag_and_charge_from_inchi(
self.inchi)
n_protons = sum([
count * Compound._obElements.GetAtomicNum(str(elem))
for (elem, count) in atom_bag.iteritems()
])
atom_bag['e-'] = n_protons - charge
return atom_bag
def transform(self, pH, I, T):
if self.inchi is None:
return 0
elif self.pKas == []:
dG0s = np.zeros((1, 1))
else:
dG0s = -np.cumsum([0] + self.pKas) * R * T * np.log(10)
dG0s = dG0s - dG0s[self.majorMSpH7]
DH = debye_huckel((I, T))
# dG0' = dG0 + nH * (R T ln(10) pH + DH) - charge^2 * DH
pseudoisomers = np.vstack(
[dG0s, np.array(self.nHs),
np.array(self.zs)]).T
dG0_prime_vector = pseudoisomers[:, 0] + \
pseudoisomers[:, 1] * (R * T * np.log(10) * pH + DH) - \
pseudoisomers[:, 2]**2 * DH
return -R * T * logsumexp(dG0_prime_vector / (-R * T))
class Molecule(object):
# for more rendering options visit:
# http://www.ggasoftware.com/opensource/indigo/api/options#rendering
_obElements = openbabel.OBElementTable()
_obSmarts = openbabel.OBSmartsPattern()
@staticmethod
def GetNumberOfElements():
return Molecule._obElements.GetNumberOfElements()
@staticmethod
def GetAllElements():
return [Molecule._obElements.GetSymbol(i) for i in
xrange(Molecule.GetNumberOfElements())]
@staticmethod
def GetSymbol(atomic_num):
return Molecule._obElements.GetSymbol(atomic_num)
@staticmethod
def GetAtomicNum(elem):
if type(elem) == types.UnicodeType:
elem = str(elem)
return Molecule._obElements.GetAtomicNum(elem)
@staticmethod
def VerifySmarts(smarts):
return Molecule._obSmarts.Init(smarts)
def __init__(self):
self.title = None
self.obmol = openbabel.OBMol()
self.smiles = None
self.inchi = None
def __str__(self):
return self.title or self.smiles or self.inchi or ""
def __len__(self):
return self.GetNumAtoms()
def Clone(self):
tmp = Molecule()
tmp.title = self.title
tmp.obmol = openbabel.OBMol(self.obmol)
tmp.smiles = self.smiles
tmp.inchi = self.inchi
return tmp
def SetTitle(self, title):
self.title = title
@staticmethod
def FromSmiles(smiles):
m = Molecule()
m.smiles = smiles
obConversion = openbabel.OBConversion()
obConversion.AddOption("w", obConversion.OUTOPTIONS)
obConversion.SetInFormat("smiles")
if not obConversion.ReadString(m.obmol, m.smiles):
raise OpenBabelError("Cannot read the SMILES string: " + smiles)
try:
m.UpdateInChI()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from SMILES: " + smiles)
m.SetTitle(smiles)
return m
@staticmethod
def FromInChI(inchi):
m = Molecule()
m.inchi = inchi
obConversion = openbabel.OBConversion()
obConversion.AddOption("w", obConversion.OUTOPTIONS)
obConversion.SetInFormat("inchi")
obConversion.ReadString(m.obmol, m.inchi)
try:
m.UpdateSmiles()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from InChI: " + inchi)
m.SetTitle(inchi)
return m
@staticmethod
def FromMol(mol):
m = Molecule()
obConversion = openbabel.OBConversion()
obConversion.AddOption("w", obConversion.OUTOPTIONS)
obConversion.SetInFormat("mol")
obConversion.ReadString(m.obmol, mol)
try:
m.UpdateInChI()
m.UpdateSmiles()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from MOL file:\n" + mol)
m.SetTitle("")
return m
@staticmethod
def FromOBMol(obmol):
m = Molecule()
m.obmol = obmol
try:
m.UpdateInChI()
m.UpdateSmiles()
except OpenBabelError:
raise OpenBabelError("Failed to create Molecule from OBMol")
m.SetTitle("")
return m
@staticmethod
def _FromFormat(s, fmt='inchi'):
if fmt == 'smiles' or fmt == 'smi':
return Molecule.FromSmiles(s)
if fmt == 'inchi':
return Molecule.FromInChI(s)
if fmt == 'mol':
return Molecule.FromMol(s)
if fmt == 'obmol':
return Molecule.FromOBMol(s)
@staticmethod
def _ToFormat(obmol, fmt='inchi'):
obConversion = openbabel.OBConversion()
obConversion.AddOption("w", obConversion.OUTOPTIONS)
obConversion.SetOutFormat(fmt)
res = obConversion.WriteString(obmol)
if not res:
raise OpenBabelError("Cannot convert OBMol to %s" % fmt)
if fmt == 'smiles' or fmt == 'smi':
res = res.split()
if res == []:
raise OpenBabelError("Cannot convert OBMol to %s" % fmt)
else:
return res[0]
elif fmt == 'inchi':
return res.strip()
else:
return res
@staticmethod
def Smiles2InChI(smiles):
obConversion = openbabel.OBConversion()
obConversion.AddOption("w", obConversion.OUTOPTIONS)
obConversion.SetInAndOutFormats("smiles", "inchi")
obmol = openbabel.OBMol()
if not obConversion.ReadString(obmol, smiles):
raise OpenBabelError("Cannot read the SMILES string: " + smiles)
return obConversion.WriteString(obmol).strip()
@staticmethod
def InChI2Smiles(inchi):
obConversion = openbabel.OBConversion()
obConversion.AddOption("w", obConversion.OUTOPTIONS)
obConversion.SetInAndOutFormats("inchi", "smiles")
obmol = openbabel.OBMol()
if not obConversion.ReadString(obmol, inchi):
raise OpenBabelError("Cannot read the InChI string: " + inchi)
return obConversion.WriteString(obmol).split()[0]
def RemoveHydrogens(self):
self.obmol.DeleteHydrogens()
def RemoveAtoms(self, indices):
self.obmol.BeginModify()
for i in sorted(indices, reverse=True):
self.obmol.DeleteAtom(self.obmol.GetAtom(i+1))
self.obmol.EndModify()
self.smiles = None
self.inchi = None
def SetAtomicNum(self, index, new_atomic_num):
self.obmol.GetAtom(index+1).SetAtomicNum(new_atomic_num)
self.smiles = None
self.inchi = None
def ToOBMol(self):
return self.obmol
def ToFormat(self, fmt='inchi'):
return Molecule._ToFormat(self.obmol, fmt=fmt)
def ToMolfile(self):
return self.ToFormat('mol')
def UpdateInChI(self):
self.inchi = Molecule._ToFormat(self.obmol, 'inchi')
def ToInChI(self):
"""
Lazy storage of the InChI identifier (calculate once only when
asked for and store for later use).
"""
if not self.inchi:
self.UpdateInChI()
return self.inchi
def UpdateSmiles(self):
self.smiles = Molecule._ToFormat(self.obmol, 'smiles')
def ToSmiles(self):
"""
Lazy storage of the SMILES identifier (calculate once only when
asked for and store for later use).
"""
if not self.smiles:
self.UpdateSmiles()
return self.smiles
def GetFormula(self):
tokens = re.findall('InChI=1S?/([0-9A-Za-z\.]+)', self.ToInChI())
if len(tokens) == 1:
return tokens[0]
elif len(tokens) > 1:
raise ValueError('Bad InChI: ' + self.ToInChI())
else:
return ''
def GetExactMass(self):
return self.obmol.GetExactMass()
def GetAtomBagAndCharge(self):
inchi = self.ToInChI()
fixed_charge = 0
for s in re.findall('/q([0-9\+\-]+)', inchi):
fixed_charge += int(s)
fixed_protons = 0
for s in re.findall('/p([0-9\+\-]+)', inchi):
fixed_protons += int(s)
formula = self.GetFormula()
atom_bag = {}
for mol_formula_times in formula.split('.'):
for times, mol_formula in re.findall('^(\d+)?(\w+)', mol_formula_times):
if not times:
times = 1
else:
times = int(times)
for atom, count in re.findall("([A-Z][a-z]*)([0-9]*)", mol_formula):
if count == '':
count = 1
else:
count = int(count)
atom_bag[atom] = atom_bag.get(atom, 0) + count * times
if fixed_protons:
atom_bag['H'] = atom_bag.get('H', 0) + fixed_protons
fixed_charge += fixed_protons
return atom_bag, fixed_charge
def GetHydrogensAndCharge(self):
atom_bag, charge = self.GetAtomBagAndCharge()
return atom_bag.get('H', 0), charge
def GetNumElectrons(self):
"""Calculates the number of electrons in a given molecule."""
atom_bag, fixed_charge = self.GetAtomBagAndCharge()
n_protons = 0
for elem, count in atom_bag.iteritems():
n_protons += count * self._obElements.GetAtomicNum(elem)
return n_protons - fixed_charge
def GetNumAtoms(self):
return self.obmol.NumAtoms()
def GetAtoms(self):
return [self.obmol.GetAtom(i+1) for i in xrange(self.obmol.NumAtoms())]
def FindSmarts(self, smarts):
"""
Corrects the pyBel version of Smarts.findall() which returns results as tuples,
with 1-based indices even though Molecule.atoms is 0-based.
Args:
mol: the molecule to search in.
smarts_str: the SMARTS query to search for.
Returns:
The re-mapped list of SMARTS matches.
"""
Molecule._obSmarts.Init(smarts)
if Molecule._obSmarts.Match(self.obmol):
match_list = Molecule._obSmarts.GetMapList()
shift_left = lambda m: [(n - 1) for n in m]
return map(shift_left, match_list)
else:
return []
def GetAtomCharges(self):
"""
Returns:
A list of charges, according to the number of atoms
in the molecule
"""
return [atom.GetFormalCharge() for atom in self.GetAtoms()]
else:
ipython_notebook = False
except NameError:
ipython_notebook = False
try:
__version__ = check_output(['obabel', '-V']).split()[2].decode('ascii')
except Exception as e:
__version__ = ''
# setup typetable to translate atom types
typetable = OBTypeTable()
typetable.SetFromType('INT')
typetable.SetToType('SYB')
# setup ElementTable
elementtable = ob.OBElementTable()
# hash OB!
ob.obErrorLog.StopLogging()
def _filereader_mol2(filename, opt=None):
block = ''
data = ''
n = 0
with gzip.open(filename) if filename.split('.')[-1] == 'gz' else open(
filename) as f:
for line in f:
if line[:1] == '#':
data += line
elif line[:17] == '@<TRIPOS>MOLECULE':
def pic(self, filename, picformat='svg'):
"""
Generates a graphical file with 2D-representation of the resonance structure
"""
try:
import openbabel as ob
except:
print "Cannot import openbabel"
return
#ValEl = {'H':1, 'B':3,'C':4,'N':5,'O':6,'F':7,'S':6}
#ValEl = {'1':1, '5':3,'6':4,'7':5,'8':6,'9':7,'16':6}
# Import Element Numbers
ati = []
Sym2Num = ob.OBElementTable()
for a in self.symbols:
ElNum = Sym2Num.GetAtomicNum(a)
ati.append(ElNum)
# Import connections
conn = self.data
mol = ob.OBMol()
# Create atoms
for a in ati:
at = ob.OBAtom()
at.SetAtomicNum(a)
mol.AddAtom(at)
# Create connections
val = []
total_LP = 0
for i in range(len(conn)):
total_LP += conn[i][i]
for i in range(len(conn)):
val.append(conn[i][i] * 2)
for j in range(i):
if conn[i][j] == 0:
continue
val[i] += conn[i][j]
val[j] += conn[i][j]
atA = mol.GetAtomById(i)
atB = mol.GetAtomById(j)
b = ob.OBBond()
b.SetBegin(atA)
b.SetEnd(atB)
b.SetBO(int(conn[i][j]))
mol.AddBond(b)
for i in range(len(conn)):
atA = mol.GetAtomById(i)
atAN = atA.GetAtomicNum()
FormValEl = CountValenceEl(atAN)
#if total_LP == 0:
# if atAN == 1:
# FullShell = 2
# else:
# FullShell = 8
# FormCharge = FormValEl + int(val[i]) - FullShell
#else:
FormCharge = int(FormValEl - val[i])
#print "atAN, FormValEl, val[i], FullShell"
#print atAN, FormValEl, val[i], FullShell
#FormCharge = FormCharge % 2
atA.SetFormalCharge(FormCharge)
# Export file
mol.DeleteNonPolarHydrogens()
conv = ob.OBConversion()
conv.SetOutFormat(picformat)
conv.AddOption('C')
conv.WriteFile(mol, filename)