def all_pdbids_from_file(filename):
"""Returns a set of all PDB ID:chains in the specified SCOP class file.
"""
with open(filename, 'r') as f:
res = set()
for record in cla_parse(f):
if record.residues.fragments and record.residues.fragments[0]:
res.add(record.residues.pdbid + ':' + record.residues.fragments[0][0])
return res
def all_pdbids_from_file_in(filename, target_key, target_value):
"""Returns a set of all PDB ID:chainss in the specified SCOP classification file
in the hierarchy at the specified level (target_key) and with the specified
SCOP unique identifying number (target_value).
target_key should be one of the members of the Keys class.
"""
pdbids = set(())
with open(filename, 'r') as f:
# iterate over each record in the SCOP Classification file
for record in cla_parse(f):
# iterate over each key/value pair in that record's hierarchy
for key, value in record.hierarchy:
# if this record is of the correct level of the hierarchy
if key == target_key and value == target_value and record.residues.fragments and record.residues.fragments[0]:
pdbids.add(record.residues.pdbid + ':' + record.residues.fragments[0][0])
return pdbids
def hierarchy_sets_from_file(filename, target_key, target_value):
"""Reads PDB ids from the specified SCOP classification file and returns a
map from elements of the hierarchy level beneath the specified target key
to a set containing (PDB ID, chain) tupless of all proteins described by
that hierarchy classification.
For example, if target_key is Keys.SUPERFAMILY and target_value is 50156
(the PDZ domain superfamily), this function returns a map from family IDs
to sets of all protein PDB IDs which are in that family.
"""
# TODO change this to return the specific chain:range_start-range_end
# instead of the 'to' in the map being PDB IDs, it should be to (maybe) pdbCHAIN:start-end
# FOR NOW this returns a tuple of pdbid,chain (chain may be None)
result = {}
logger.debug("opening " + filename + " in hierarchy_sets_from_file. target: " + target_key + ':' + str(target_value))
with open(filename, 'r') as hierarchy_file:
try:
if Keys.order.index(target_key) == len(Keys.order):
raise Error('Cannot get sets for lowest level of hierarchy')
except ValueError:
raise ValueError('Key "' + str(target_key) + '" is not a ' + \
'known hierarchy key')
# iterate over each record in the SCOP Classification file
for record in cla_parse(hierarchy_file):
# iterate over each key/value pair in that record's hierarchy
for key, value in record.hierarchy:
# if this record is of the correct level of the hierarchy
if key == target_key and value == target_value:
# get the key for the next level down in the hierarchy
next_target_key = Keys.order[Keys.order.index(key) + 1]
# iterate over the record's hierarchy AGAIN
for key2, value2 in record.hierarchy:
# if we are looking at the next target key
if key2 == next_target_key:
# add this record's pdbid
# and chain to the result dictionary
# we get the fragments (residues.fragments)
# and add pdbid + fragments[0]
if record.residues.fragments:
# we have a fragment, use first chain
newfrag = str(record.residues.fragments[0][0])
else:
newfrag = None
newval = (str(record.residues.pdbid), newfrag)
if value2 in result:
result[value2].add(newval)
else:
result[value2] = set(())
result[value2].add(newval)
return result
