def get_structure(pdb_id):
'''Returns a PDB structure.'''
source_url = 'http://www.rcsb.org/pdb/files/' + pdb_id + '.pdb'
target_filename = os.path.join(os.path.expanduser('~'), _DIR, _PDB_DIR,
pdb_id + '.pdb')
with open(io_utils.get_file(source_url, target_filename)) as pdb_file:
parser = PDBParser(QUIET=True)
return parser.get_structure(pdb_id, pdb_file.name)
def get_seq_structs(pdb_ids=None):
'''Returns sequence and structure.'''
seq_structs = {}
pdb_ids = sorted(pdb_ids) if pdb_ids is not None else None
in_field = False
tokens = ()
str_data = ''
source_url = 'http://www.rcsb.org/pdb/files/ss.txt'
target_filename = os.path.join(os.path.expanduser('~'), _DIR,
'ss.txt')
with open(io_utils.get_file(source_url, target_filename)) as fle:
for line in fle:
if line.startswith('>'):
pdb_id = re.search('(?<=\\>)[^:]*', line).group(0)
if pdb_ids is None or pdb_id in pdb_ids:
if in_field:
if tokens[:2] not in seq_structs:
seq_structs[tokens[:2]] = [None, None]
seq_structs[tokens[:2]][0 if tokens[2] == 'sequence'
else 1] = str_data
str_data = ''
tokens = tuple(re.split('>|:', line.strip())[1:])
in_field = True
elif in_field:
if tokens[:2] not in seq_structs:
seq_structs[tokens[:2]] = [None, None]
seq_structs[tokens[:2]][0 if tokens[2] == 'sequence'
else 1] = str_data
str_data = ''
in_field = False
tokens = ()
str_data = ''
elif in_field:
str_data += line[:-1]
return {key: value for key, value in seq_structs.iteritems()
if all(val is not None for val in value)}
def get_pdb_ids(max_ids=None, local_only=False):
'''Returns PDB ids.'''
if local_only:
# Returns local PDB ids.
pdb_dir = os.path.join(os.path.expanduser('~'), _DIR, _PDB_DIR)
ids = [filename[:-4].upper()
for _, _, files in os.walk(pdb_dir)
for filename in files if filename.endswith('.pdb')]
else:
# Returns all PDB ids.
source_url = 'http://www.uniprot.org/uniprot/?query=database:pdb' \
+ '&format=tab&columns=id,database(PDB)'
target_filename = os.path.join(os.path.expanduser('~'), _DIR,
'pdb_ids.txt')
with open(io_utils.get_file(source_url, target_filename)) as fle:
ids = [x for line in fle
for x in line.split()[1].split(';')
if len(x) > 0 and x != 'Cross-reference']
return ids if max_ids is None \
else random.sample(ids, min(len(ids), max_ids))
