def read_cartesian(self,fname):
molc=Molecule()
f = open(fname,'r')
while (True):
str = f.readline()
if (str==''):
break
sp = str.split()
if (len(sp) != 4):
continue
molc.add_atom(Atom(sp[0].lower(),0,float(sp[1]),float(sp[2]),float(sp[3])))
self.mol = molc
return molc
def load(file_name):
# we read a list of molecules
read_molecules = []
sdf_file = None
try:
sdf_file = open(file_name, "r")
except IOError as e:
# file not found or something... not much we could do other than print an error and return an empty
# molecule list
error(">>> Error while opening %s" % file_name)
error(">>> I/O error code %d: %s" % (e.errno, e.strerror))
error(">>> No molecules were loaded.")
return read_molecules
atomic_masses = {
"C": 12.0107,
"H": 1.00794,
"Cl": 35.4527,
"N": 14.00674,
"O": 15.9994,
"S": 32.066,
"F": 18.9984032
}
line_number = 1
molecule_name = ""
n_atoms = 0
n_bonds = 0
molecule = ''
absolute_line = 1
property_name = None
reading_property = False
for line in sdf_file:
# decide where are we based on the line number
if SDFReader.is_header(line_number):
# we append the header lines to the name of the molecule
if line_number > 1:
molecule_name += ','
molecule_name += line.strip()
elif SDFReader.is_counts_line(line_number):
# we can create the molecule and add its name
molecule = Molecule(molecule_name)
# reset the name
molecule_name = ''
# extract the number of atoms and bonds
# number of atoms are the first three characters and number of bonds the next three
n_atoms = int(line[0:3])
n_bonds = int(line[3:6])
elif SDFReader.is_atom_table(line_number, n_atoms):
# the coordinates are given in the first 30 characters, 10 per each dimension
x_pos = float(line[0:10])
y_pos = float(line[10:20])
z_pos = float(line[20:30])
# then the name of the atom
symbol = line[30:33].strip()
# insert the atom in the molecule
# TODO: use a table to find the atomic mass based on the atomic symbol!
atomic_mass = atomic_masses.get(symbol)
molecule.add_atom(
Atom(symbol, Point3D(x_pos, y_pos, z_pos), atomic_mass))
elif SDFReader.is_bond_table(line_number, n_atoms, n_bonds):
# the first three characters represent the first atom
atom_one_index = int(line[0:3])
# the 4th-6th characters represent the second atom
atom_two_index = int(line[3:6])
# the 7th-9th characters represent the bond type
bond_type = int(line[6:9])
# remember that in python indices start from 0, while in mol files, they start from 1!!!
molecule.add_bond(atom_one_index - 1, atom_two_index - 1,
bond_type)
elif SDFReader.is_properties_block(line_number, n_atoms, n_bonds):
if line == "M END":
# ignore
pass
elif line.startswith('>'):
# property
reading_property = True
# this is what the line looks like_
# > <PUBCHEM_COMPOUND_CID>
property_name = line[3:line.rfind('>')]
elif reading_property:
# the name has been read, we take the whole line as the value (remove line break)
# right now, no support for multi-line property values is available...
molecule.add_property(property_name, line.rstrip('\n'))
reading_property = False
elif line.strip().startswith("$$$$"):
# end of molecule!
read_molecules.append(molecule)
line_number = 0
# done reading this line
line_number += 1
absolute_line += 1
# done reading file, close it
try:
sdf_file.close()
except IOError:
# ignore the error
pass
return read_molecules
class GeomReader:
def read_cartesian(self,fname):
molc=Molecule()
f = open(fname,'r')
while (True):
str = f.readline()
if (str==''):
break
sp = str.split()
if (len(sp) != 4):
continue
molc.add_atom(Atom(sp[0].lower(),0,float(sp[1]),float(sp[2]),float(sp[3])))
self.mol = molc
return molc
def read_zmatrix(self,fname):
f = open(fname,'r')
self.file = f;
name = f.readline().strip()
self.mol = Molecule()
self.mol.add_atom(Atom(name.lower(),1,0,0,0,connect=0));
str = f.readline();
sp = str.split();
if (sp == []):
return self.mol
sp[0] = sp[0].lower()
if (sp==[]):
return self.mol
try:
bl = float(sp[2])
except:
bl = self.lookup_var(sp[2])
self.mol.add_atom(Atom(sp[0],2,0,0,bl,int(sp[1])))
self.mol.add_bond(2,int(sp[1]))
str = f.readline();
sp = str.split();
if (sp == []):
return self.mol
sp[0] = sp[0].lower()
try:
bl = float(sp[2]);
except:
bl = self.lookup_var(sp[2])
try:
ba = float(sp[4]);
except:
ba = self.lookup_var(sp[4])
a = Atom(sp[0],3,math.sin(ba*math.pi/180.0)*bl,0,self.mol.atoms[int(sp[1]) -1].z - math.cos(ba*math.pi/180.0)*bl,int(sp[1]))# BIG CHANGE
self.mol.add_atom(a)
self.mol.add_bond(3,int(sp[1]))
atom_number = 4;
while (True):
#print atom_number
str = f.readline();
sp = str.split();
if (sp==[]):
print 'end of matrix'
break
sp[0] = sp[0].lower()
try:
bl = float(sp[2]);
except:
bl = self.lookup_var(sp[2])
try:
ba = float(sp[4]);
except:
ba = self.lookup_var(sp[4])
try:
dh = float(sp[6]);
except:
dh = self.lookup_var(sp[6])
connect = int(sp[1])
angle_connect = int(sp[3])
dihed_connect = int(sp[5])
atoms = self.mol.atoms;
#print 'connect: %i angle_connect: %i' %(connect,angle_connect)
vector1 = vec_minus(atoms[connect-1].xyz,atoms[angle_connect-1].xyz );
vector2 = vec_minus(atoms[connect-1].xyz,atoms[dihed_connect-1].xyz );
norm1 = vec_cross(vector1,vector2)
norm2 = vec_cross(vector1,norm1)
norm1 = normalise(norm1)
norm2 = normalise(norm2)
norm1 =vec_times(norm1,-1*math.sin(dh*math.pi/180))
norm2 = vec_times(norm2,math.cos(dh*math.pi/180))
vector3 =vec_add(norm1,norm2)
vector3 =normalise(vector3)
vector3 = vec_times(vector3,bl*math.sin(ba*math.pi/180.0))
vector1 = normalise(vector1)
vector1 = vec_times(vector1,bl*math.cos(ba*math.pi/180.0))
vector2 = vec_add(atoms[connect - 1].xyz,vector3)
vector2 = vec_minus(vector2,vector1)
a = Atom(sp[0],atom_number,vector2[0],vector2[1],vector2[2],int(sp[1]))
self.mol.add_atom(a)
self.mol.add_bond(atom_number,int(sp[1]))
atom_number+=1;
return self.mol
def lookup_var(self,name):
# print 'Looking up ',name
mark = self.file.tell();
name.strip()
mul = 1;
if (name[0] == '-'):
mul = -1;
name = name[1:]
while (True):
str = self.file.readline();
if (str == ''):
print 'Lookup error'
break;
sp = str.split()
if (sp ==[]):
continue
if (sp[0]==name):
self.file.seek(mark);
return float(sp[1])*mul;
print 'Couldnt look up ', name;
self.file.seek(mark);
return None;
def __init__(self, fname):
if fname.endswith(".zmat"):
self.read_zmatrix(fname)
else:
self.read_cartesian(fname)
def write_atoms(self,fname):
f = open(fname,'w')
for at in self.mol.atoms:
f.write('%s %f %f %f\n'%(at.name,at.x,at.y,at.z))
f.close()
def print_mol_ats(self):
key = {}
for at in self.mol.atoms:
try:
key[at.name]+=1
except:
key[at.name]=1
for i in key:
print '%s %i' % (i,key[i])