def find_disulfides(self):
''' Finds disulfide bridges.
'''
if self.path_to_pdb=='':
raise AssertionError('Run find_files function first to locate the model source file.')
disulfide_pairs = []
with open(self.path_to_pdb, 'r') as f:
lines = f.readlines()
c=1
for line in lines:
if c>10:
break
if line.startswith('SSBOND'):
pdb_re = re.search('SSBOND\s+\d+\s+CYS\sR\s+(\d+)\s+CYS\sR\s+(\d+)', line)
num1 = pdb_re.group(1)
num2 = pdb_re.group(2)
res1, res2 = list(Residue.objects.filter(protein_conformation__protein__entry_name=self.receptor, sequence_number__in=[num1, num2]))
if res1.display_generic_number!=None:
gn1 = ggn(res1.display_generic_number.label)
else:
gn1 = str(res1.sequence_number)
if res2.display_generic_number!=None:
gn2 = ggn(res2.display_generic_number.label)
else:
gn2 = str(res2.sequence_number)
disulfide_pairs.append([gn1, gn2])
c+=1
print('FIND disulfides: {}'.format(disulfide_pairs))
return disulfide_pairs
def find_disulfides(self):
''' Finds disulfide bridges.
'''
if self.path_to_pdb=='':
raise AssertionError('Run find_files function first to locate the model source file.')
disulfide_pairs = []
with open(self.path_to_pdb, 'r') as f:
lines = f.readlines()
c=1
for line in lines:
if c>10:
break
if line.startswith('SSBOND'):
pdb_re = re.search('SSBOND\s+\d+\s+CYS\sR\s+(\d+)\s+CYS\sR\s+(\d+)', line)
num1 = pdb_re.group(1)
num2 = pdb_re.group(2)
res1, res2 = list(Residue.objects.filter(protein_conformation__protein__entry_name=self.receptor, sequence_number__in=[num1, num2]))
if res1.display_generic_number!=None:
gn1 = ggn(res1.display_generic_number.label)
else:
gn1 = str(res1.sequence_number)
if res2.display_generic_number!=None:
gn2 = ggn(res2.display_generic_number.label)
else:
gn2 = str(res2.sequence_number)
disulfide_pairs.append([gn1, gn2])
c+=1
print('FIND disulfides: {}'.format(disulfide_pairs))
return disulfide_pairs
def fetch_residues_from_pdb(self, structure, generic_numbers, modify_bulges=False, just_nums=False):
''' Fetches specific lines from pdb file by generic number (if generic number is
not available then by residue number). Returns nested OrderedDict()
with generic numbers as keys in the outer dictionary, and atom names as keys
in the inner dictionary.
@param structure: Structure, Structure object where residues should be fetched from \n
@param generic_numbers: list, list of generic numbers to be fetched \n
@param modify_bulges: boolean, set it to true when used for bulge switching. E.g. you want a 5x461
residue to be considered a 5x46 residue.
'''
output = OrderedDict()
atoms_list = []
for gn in generic_numbers:
rotamer=None
if 'x' in str(gn):
rotamer = list(Rotamer.objects.filter(structure__protein_conformation=structure.protein_conformation,
residue__display_generic_number__label=dgn(gn,structure.protein_conformation),
structure__preferred_chain=structure.preferred_chain))
else:
rotamer = list(Rotamer.objects.filter(structure__protein_conformation=structure.protein_conformation,
residue__sequence_number=gn, structure__preferred_chain=structure.preferred_chain))
if just_nums==False:
try:
gn = ggn(Residue.objects.get(protein_conformation=structure.protein_conformation,
sequence_number=gn).display_generic_number.label)
except:
pass
if len(rotamer)>1:
for i in rotamer:
if i.pdbdata.pdb.startswith('COMPND')==False:
if i.pdbdata.pdb[21] in structure.preferred_chain:
rotamer = i
break
else:
rotamer = rotamer[0]
io = StringIO(rotamer.pdbdata.pdb)
rota_struct = PDB.PDBParser(QUIET=True).get_structure('structure', io)[0]
for chain in rota_struct:
for residue in chain:
for atom in residue:
atoms_list.append(atom)
if modify_bulges==True and len(gn)==5:
output[gn.replace('x','.')[:-1]] = atoms_list
else:
try:
output[gn.replace('x','.')] = atoms_list
except:
output[str(gn)] = atoms_list
atoms_list = []
return output
def fetch_residues_from_pdb(self, structure, generic_numbers, modify_bulges=False, just_nums=False):
''' Fetches specific lines from pdb file by generic number (if generic number is
not available then by residue number). Returns nested OrderedDict()
with generic numbers as keys in the outer dictionary, and atom names as keys
in the inner dictionary.
@param structure: Structure, Structure object where residues should be fetched from \n
@param generic_numbers: list, list of generic numbers to be fetched \n
@param modify_bulges: boolean, set it to true when used for bulge switching. E.g. you want a 5x461
residue to be considered a 5x46 residue.
'''
output = OrderedDict()
atoms_list = []
for gn in generic_numbers:
rotamer=None
if 'x' in str(gn):
rotamer = list(Rotamer.objects.filter(structure__protein_conformation=structure.protein_conformation,
residue__display_generic_number__label=dgn(gn,structure.protein_conformation),
structure__preferred_chain=structure.preferred_chain))
else:
rotamer = list(Rotamer.objects.filter(structure__protein_conformation=structure.protein_conformation,
residue__sequence_number=gn, structure__preferred_chain=structure.preferred_chain))
if just_nums==False:
try:
gn = ggn(Residue.objects.get(protein_conformation=structure.protein_conformation,
sequence_number=gn).display_generic_number.label)
except:
pass
if len(rotamer)>1:
for i in rotamer:
if i.pdbdata.pdb.startswith('COMPND')==False:
if i.pdbdata.pdb[21] in structure.preferred_chain:
rotamer = i
break
else:
rotamer = rotamer[0]
io = StringIO(rotamer.pdbdata.pdb)
rota_struct = PDB.PDBParser(QUIET=True).get_structure('structure', io)[0]
for chain in rota_struct:
for residue in chain:
for atom in residue:
atoms_list.append(atom)
if modify_bulges==True and len(gn)==5:
output[gn.replace('x','.')[:-1]] = atoms_list
else:
try:
output[gn.replace('x','.')] = atoms_list
except:
output[str(gn)] = atoms_list
atoms_list = []
return output
def pdb_array_creator(self, structure=None, filename=None):
''' Creates an OrderedDict() from the pdb of a Structure object where residue numbers/generic numbers are
keys for the residues, and atom names are keys for the Bio.PDB.Residue objects.
@param structure: Structure, Structure object of protein. When using structure, leave filename=None. \n
@param filename: str, filename of pdb to be parsed. When using filename, leave structure=None).
'''
# seq_nums_overwrite_cutoff_dict = {'4PHU':2000, '4LDL':1000, '4LDO':1000, '4QKX':1000, '5JQH':1000, '5TZY':2000, '5KW2':2000}
if structure != None and filename == None:
io = StringIO(structure.pdb_data.pdb)
else:
io = filename
gn_array = []
residue_array = []
# pdb_struct = PDB.PDBParser(QUIET=True).get_structure(structure.pdb_code.index, io)[0]
residues = Residue.objects.filter(
protein_conformation=structure.protein_conformation)
gn_list = []
for i in residues:
try:
gn_list.append(
ggn(i.display_generic_number.label).replace('x', '.'))
except:
pass
ssno = StructureSeqNumOverwrite(structure)
ssno.seq_num_overwrite('pdb')
if len(ssno.pdb_wt_table) > 0:
residues = residues.filter(protein_segment__slug__in=[
'TM1', 'TM2', 'TM3', 'TM4', 'TM5', 'TM6', 'TM7', 'H8'
]).order_by('sequence_number')
output = OrderedDict()
for r in residues:
if r.protein_segment.slug == None:
continue
if r.protein_segment.slug not in output:
output[r.protein_segment.slug] = OrderedDict()
rotamer = Rotamer.objects.filter(residue=r)
rotamer = self.right_rotamer_select(rotamer)
rota_io = StringIO(rotamer.pdbdata.pdb)
p = PDB.PDBParser()
parsed_rota = p.get_structure('rota', rota_io)
for chain in parsed_rota[0]:
for res in chain:
atom_list = []
for atom in res:
# Skip hydrogens
if atom.get_id().startswith('H'):
continue
if atom.get_id() == 'N':
bw, gn = r.display_generic_number.label.split(
'x')
atom.set_bfactor(bw)
elif atom.get_id() == 'CA':
bw, gn = r.display_generic_number.label.split(
'x')
gn = "{}.{}".format(bw.split('.')[0], gn)
if len(gn.split('.')[1]) == 3:
gn = '-' + gn[:-1]
atom.set_bfactor(gn)
atom_list.append(atom)
output[r.protein_segment.slug][ggn(
r.display_generic_number.label).replace(
'x', '.')] = atom_list
pprint.pprint(output)
return output
else:
assign_gn = as_gn.GenericNumbering(
pdb_file=io,
pdb_code=structure.pdb_code.index,
sequence_parser=True)
pdb_struct = assign_gn.assign_generic_numbers_with_sequence_parser(
)
pref_chain = structure.preferred_chain
parent_prot_conf = ProteinConformation.objects.get(
protein=structure.protein_conformation.protein.parent)
parent_residues = Residue.objects.filter(
protein_conformation=parent_prot_conf)
last_res = list(parent_residues)[-1].sequence_number
if len(pref_chain) > 1:
pref_chain = pref_chain[0]
for residue in pdb_struct[pref_chain]:
if 'CA' in residue and -9.1 < residue['CA'].get_bfactor(
) < 9.1:
use_resid = False
gn = str(residue['CA'].get_bfactor())
if len(gn.split('.')[1]) == 1:
gn = gn + '0'
if gn[0] == '-':
gn = gn[1:] + '1'
# Exceptions
if structure.pdb_code.index == '3PBL' and residue.get_id(
)[1] == 331:
use_resid = True
elif structure.pdb_code.index == '6QZH' and residue.get_id(
)[1] == 1434:
use_resid = True
elif structure.pdb_code.index == '7M3E':
use_resid = True
#################################################
elif gn in gn_list:
gn_array.append(gn)
residue_array.append(residue.get_list())
else:
use_resid = True
if use_resid:
gn_array.append(str(residue.get_id()[1]))
residue_array.append(residue.get_list())
output = OrderedDict()
for num, label in self.segment_coding.items():
output[label] = OrderedDict()
if len(gn_array) != len(residue_array):
raise AssertionError()
for gn, res in zip(gn_array, residue_array):
if '.' in gn:
seg_num = int(gn.split('.')[0])
seg_label = self.segment_coding[seg_num]
if seg_num == 8 and len(output['TM7']) == 0:
continue
else:
output[seg_label][gn] = res
else:
try:
found_res, found_gn = None, None
try:
found_res = Residue.objects.get(
protein_conformation=structure.
protein_conformation,
sequence_number=gn)
except:
# Exception for res 317 in 5VEX, 5VEW
if structure.pdb_code.index in [
'5VEX', '5VEW'
] and gn == '317' and res[0].get_parent(
).get_resname() == 'CYS':
found_res = Residue.objects.get(
protein_conformation=parent_prot_conf,
sequence_number=gn)
#####################################
found_gn = str(
ggn(found_res.display_generic_number.label)
).replace('x', '.')
# Exception for res 318 in 5VEX, 5VEW
if structure.pdb_code.index in [
'5VEX', '5VEW'
] and gn == '318' and res[0].get_parent().get_resname(
) == 'ILE' and found_gn == '5.47':
found_gn = '5.48'
#####################################
if -9.1 < float(found_gn) < 9.1:
if len(res) == 1:
continue
if int(gn) > last_res:
continue
seg_label = self.segment_coding[int(
found_gn.split('.')[0])]
output[seg_label][found_gn] = res
except:
if res[0].get_parent().get_resname() == 'YCM' or res[
0].get_parent().get_resname() == 'CSD':
try:
found_res = Residue.objects.get(
protein_conformation=parent_prot_conf,
sequence_number=gn)
except:
continue
if found_res.protein_segment.slug[0] not in [
'T', 'H'
]:
continue
try:
found_gn = str(
ggn(found_res.display_generic_number.label)
).replace('x', '.')
except:
found_gn = str(gn)
output[
found_res.protein_segment.slug][found_gn] = res
return output
def pdb_array_creator(self, structure=None, filename=None):
''' Creates an OrderedDict() from the pdb of a Structure object where residue numbers/generic numbers are
keys for the residues, and atom names are keys for the Bio.PDB.Residue objects.
@param structure: Structure, Structure object of protein. When using structure, leave filename=None. \n
@param filename: str, filename of pdb to be parsed. When using filename, leave structure=None).
'''
seq_nums_overwrite_cutoff_dict = {'4PHU':2000, '4LDL':1000, '4LDO':1000, '4QKX':1000, '5JQH':1000, '5TZY':2000, '5KW2':2000}
if structure!=None and filename==None:
io = StringIO(structure.pdb_data.pdb)
else:
io = filename
gn_array = []
residue_array = []
# pdb_struct = PDB.PDBParser(QUIET=True).get_structure(structure.pdb_code.index, io)[0]
residues = Residue.objects.filter(protein_conformation=structure.protein_conformation)
gn_list = []
for i in residues:
try:
gn_list.append(ggn(i.display_generic_number.label).replace('x','.'))
except:
pass
assign_gn = as_gn.GenericNumbering(pdb_file=io, pdb_code=structure.pdb_code.index, sequence_parser=True)
pdb_struct = assign_gn.assign_generic_numbers_with_sequence_parser()
pref_chain = structure.preferred_chain
parent_prot_conf = ProteinConformation.objects.get(protein=structure.protein_conformation.protein.parent)
parent_residues = Residue.objects.filter(protein_conformation=parent_prot_conf)
last_res = list(parent_residues)[-1].sequence_number
if len(pref_chain)>1:
pref_chain = pref_chain[0]
for residue in pdb_struct[pref_chain]:
try:
if -9.1 < residue['CA'].get_bfactor() < 9.1:
gn = str(residue['CA'].get_bfactor())
if len(gn.split('.')[1])==1:
gn = gn+'0'
if gn[0]=='-':
gn = gn[1:]+'1'
# Exception for 3PBL 331, gn get's assigned wrong
if structure.pdb_code.index=='3PBL' and residue.get_id()[1]==331:
raise Exception()
#################################################
if gn in gn_list:
if int(residue.get_id()[1])>1000:
if structure.pdb_code.index in seq_nums_overwrite_cutoff_dict and int(residue.get_id()[1])>=seq_nums_overwrite_cutoff_dict[structure.pdb_code.index]:
gn_array.append(gn)
residue_array.append(residue.get_list())
else:
raise Exception()
else:
gn_array.append(gn)
residue_array.append(residue.get_list())
else:
raise Exception()
else:
raise Exception()
except:
if structure!=None and structure.pdb_code.index in seq_nums_overwrite_cutoff_dict:
if int(residue.get_id()[1])>seq_nums_overwrite_cutoff_dict[structure.pdb_code.index]:
gn_array.append(str(int(str(residue.get_id()[1])[1:])))
else:
gn_array.append(str(residue.get_id()[1]))
else:
gn_array.append(str(residue.get_id()[1]))
residue_array.append(residue.get_list())
output = OrderedDict()
for num, label in self.segment_coding.items():
output[label] = OrderedDict()
if len(gn_array)!=len(residue_array):
raise AssertionError()
for gn, res in zip(gn_array,residue_array):
if '.' in gn:
seg_num = int(gn.split('.')[0])
seg_label = self.segment_coding[seg_num]
if seg_num==8 and len(output['TM7'])==0:
continue
else:
output[seg_label][gn] = res
else:
try:
found_res, found_gn = None, None
try:
found_res = Residue.objects.get(protein_conformation=structure.protein_conformation,
sequence_number=gn)
except:
# Exception for res 317 in 5VEX, 5VEW
if structure.pdb_code.index in ['5VEX','5VEW'] and gn=='317' and res[0].get_parent().get_resname()=='CYS':
found_res = Residue.objects.get(protein_conformation=parent_prot_conf,
sequence_number=gn)
#####################################
found_gn = str(ggn(found_res.display_generic_number.label)).replace('x','.')
# Exception for res 318 in 5VEX, 5VEW
if structure.pdb_code.index in ['5VEX','5VEW'] and gn=='318' and res[0].get_parent().get_resname()=='ILE' and found_gn=='5.47':
found_gn = '5.48'
#####################################
if -9.1 < float(found_gn) < 9.1:
if len(res)==1:
continue
if int(gn)>last_res:
continue
seg_label = self.segment_coding[int(found_gn.split('.')[0])]
output[seg_label][found_gn] = res
except:
if res[0].get_parent().get_resname()=='YCM' or res[0].get_parent().get_resname()=='CSD':
found_res = Residue.objects.get(protein_conformation=parent_prot_conf, sequence_number=gn)
if found_res.protein_segment.slug[0] not in ['T','H']:
continue
try:
found_gn = str(ggn(found_res.display_generic_number.label)).replace('x','.')
except:
found_gn = str(gn)
output[found_res.protein_segment.slug][found_gn] = res
return output