def read(self, f_handle, f_id="ncont"):
"""Read a contact file
Parameters
----------
f_handle
Open file handle [read permissions]
f_id : str, optional
Unique contact file identifier
Returns
-------
:obj:`~conkit.core.contactfile.ContactFile`
"""
contact_file = ContactFile(f_id)
contact_map = ContactMap("map_1")
contact_file.add(contact_map)
for line in f_handle:
line = line.strip()
if RE_CONTACT.match(line):
matches = RE_CONTACT.match(line)
res1_seq = int(matches.group(2))
res2_seq = int(matches.group(5))
lb = ub = float(matches.group(7))
if (res1_seq, res2_seq) in contact_map:
msg = (
"This parser cannot handle multiple atoms of the same residue. "
"If your contact map contains such entries, only the first will be stored!"
)
warnings.warn(msg, Warning)
continue
contact = Contact(res1_seq,
res2_seq,
1.0,
distance_bound=(lb, ub))
contact.res1_chain = matches.group(1)
contact.res2_chain = matches.group(4)
contact.res1 = matches.group(3)
contact.res2 = matches.group(6)
contact_map.add(contact)
contact_file.method = "Contact map generated using Ncont"
return contact_file
def read(self, f_handle, f_id="membrain"):
"""Read a contact file
Parameters
----------
f_handle
Open file handle [read permissions]
f_id : str, optional
Unique contact file identifier
Returns
-------
:obj:`~conkit.core.contactfile.ContactFile`
"""
hierarchy = ContactFile(f_id)
contact_map = ContactMap("map_1")
hierarchy.add(contact_map)
for line in f_handle:
line = line.rstrip()
if not line:
continue
if RE_HEADER.match(line):
continue
else:
_, res1_seq, res1, _, res2_seq, res2, raw_score = RE_SPLIT.split(
line)
contact = Contact(int(res1_seq), int(res2_seq),
float(raw_score))
contact.res1 = res1
contact.res2 = res2
contact_map.add(contact)
hierarchy.method = 'Contact map predicted using MemBrain'
return hierarchy
def read(self, f_handle, f_id="comsat"):
"""Read a contact file
Parameters
----------
f_handle
Open file handle [read permissions]
f_id : str, optional
Unique contact file identifier
Returns
-------
:obj:`~conkit.core.contactfile.ContactFile`
"""
contact_file = ContactFile(f_id)
contact_map = ContactMap("map_1")
contact_file.add(contact_map)
for line in f_handle:
line = line.rstrip()
if not line:
continue
else:
res1_seq, res1, res2_seq, res2, _ = RE_SPLIT.split(line)
contact = Contact(int(res1_seq), int(res2_seq), 0.0)
contact.res1 = res1
contact.res2 = res2
contact_map.add(contact)
contact_file.method = "Contact map predicted using COMSAT"
return contact_file
def test_res1_3(self):
contact = Contact(1, 2, 1.0)
with self.assertRaises(ValueError):
contact.res1 = "8"
def test_res1_2(self):
contact = Contact(1, 2, 1.0)
contact.res1 = "T"
self.assertEqual("T", contact.res1)
def test_res1_1(self):
contact = Contact(1, 2, 1.0)
contact.res1 = "ALA"
self.assertEqual("A", contact.res1)
def test_res1_2(self):
contact = Contact(1, 2, 1.0)
contact.res1 = 'T'
self.assertEqual('T', contact.res1)
def test_res1_1(self):
contact = Contact(1, 2, 1.0)
contact.res1 = 'ALA'
self.assertEqual('A', contact.res1)
def _read(self, structure, f_id, distance_cutoff, atom_type):
"""Read a contact file
Parameters
----------
structure
A :obj:`Structure <Bio.PDB.Structure.Structure>` instance
f_id : str
Unique contact file identifier
distance_cutoff : int
Distance cutoff for which to determine contacts
atom_type : str
Atom type between which distances are calculated
Returns
-------
:obj:`ContactFile <conkit.core.contactfile.ContactFile>`
"""
hierarchies = []
for model in structure:
hierarchy = ContactFile(f_id + '_' + str(model.id))
chains = list(chain for chain in model)
for chain in chains:
self._remove_hetatm(chain)
self._remove_atom(chain, atom_type)
for chain1, chain2 in itertools.product(chains, chains):
if chain1.id == chain2.id: # intra
contact_map = ContactMap(chain1.id)
else: # inter
contact_map = ContactMap(chain1.id + chain2.id)
for (atom1, atom2, distance) in self._chain_contacts(chain1, chain2):
contact = Contact(
atom1.resseq,
atom2.resseq,
round(1.0 - (distance / 100), 6),
distance_bound=(0., float(distance_cutoff)))
contact.res1_altseq = atom1.resseq_alt
contact.res2_altseq = atom2.resseq_alt
contact.res1 = atom1.resname
contact.res2 = atom2.resname
contact.res1_chain = atom1.reschain
contact.res2_chain = atom2.reschain
if distance_cutoff == 0 or distance < distance_cutoff:
contact.define_match()
contact_map.add(contact)
if contact_map.empty:
del contact_map
else:
if len(contact_map.id) == 1:
contact_map.sequence = self._build_sequence(chain1)
assert len(contact_map.sequence.seq) == len(chain1)
else:
contact_map.sequence = self._build_sequence(chain1) \
+ self._build_sequence(chain2)
assert len(contact_map.sequence.seq) \
== len(chain1) + len(chain2)
hierarchy.add(contact_map)
hierarchy.method = 'Contact map extracted from PDB ' + str(model.id)
hierarchy.remark = [
'The model id is the chain identifier, i.e XY equates to chain X and chain Y.',
'Residue numbers in column 1 are chain X, and numbers in column 2 are chain Y.'
]
hierarchies.append(hierarchy)
if len(hierarchies) > 1:
msg = "Super-level to contact file not yet implemented. " \
"Parser returns hierarchy for top model only!"
warnings.warn(msg, FutureWarning)
return hierarchies[0]
