def get_seqres_file(self, savefile='pdb_seqres.txt'):
"""Retrieves a (big) file containing all the sequences of PDB entries
and writes it to a file.
"""
print("Retrieving sequence file (takes over 110 MB).")
url = self.pdb_server + '/pub/pdb/derived_data/pdb_seqres.txt'
_urlretrieve(url, savefile)
def retrieve_pdb_file(self, pdb_code, obsolete=False, pdir=None):
""" Retrieves a PDB structure file from the PDB server and
stores it in a local file tree.
The PDB structure's file name is returned as a single string.
If obsolete == True, the file will be saved in a special file tree.
@param pdir: put the file in this directory (default: create a PDB-style directory tree)
@type pdir: string
@return: filename
@rtype: string
"""
# Get the compressed PDB structure
code = pdb_code.lower()
archive_fn = "pdb%s.ent.gz" % code
pdb_dir = "divided" if not obsolete else "obsolete"
url = (self.pdb_server + '/pub/pdb/data/structures/%s/pdb/%s/%s' %
(pdb_dir, code[1:3], archive_fn))
# Where does the final PDB file get saved?
if pdir is None:
path = self.local_pdb if not obsolete else self.obsolete_pdb
if not self.flat_tree: # Put in PDB-style directory tree
path = os.path.join(path, code[1:3])
else: # Put in specified directory
path = pdir
if not os.access(path, os.F_OK):
os.makedirs(path)
filename = os.path.join(path, archive_fn)
final_file = os.path.join(path, "pdb%s.ent" % code) # (decompressed)
# Skip download if the file already exists
if not self.overwrite:
if os.path.exists(final_file):
print("Structure exists: '%s' " % final_file)
return final_file
# Retrieve the file
print("Downloading PDB structure '%s'..." % pdb_code)
_urlretrieve(url, filename)
# Uncompress the archive, delete when done
#Can't use context manager with gzip.open until Python 2.7
gz = gzip.open(filename, 'rb')
with open(final_file, 'wb') as out:
out.writelines(gz)
gz.close()
os.remove(filename)
return final_file
def open_pdb(pdbid, pdb_url=None):
"""Make a local copy of an online pdb file and return a file handle."""
if pdb_url is None:
pdb_url = default_pdb_url
url = pdb_url % pdbid
fn, header = _urlretrieve(url)
return open(fn)
def get_seqres_file(self, savefile="pdb_seqres.txt"):
"""Retrieve and save a (big) file containing all the sequences of PDB entries."""
if self._verbose:
print("Retrieving sequence file (takes over 110 MB).")
url = self.pdb_server + "/pub/pdb/derived_data/pdb_seqres.txt"
_urlretrieve(url, savefile)
def retrieve_pdb_file(self,
pdb_code,
obsolete=False,
pdir=None,
file_format=None,
overwrite=False):
"""Fetch PDB structure file from PDB server, and store it locally.
The PDB structure's file name is returned as a single string.
If obsolete ``==`` True, the file will be saved in a special file tree.
NOTE. The default download format has changed from PDB to PDBx/mmCif
:param pdb_code: 4-symbols structure Id from PDB (e.g. 3J92).
:type pdb_code: string
:param file_format:
File format. Available options:
* "mmCif" (default, PDBx/mmCif file),
* "pdb" (format PDB),
* "xml" (PDBML/XML format),
* "mmtf" (highly compressed),
* "bundle" (PDB formatted archive for large structure}
:type file_format: string
:param overwrite: if set to True, existing structure files will be overwritten. Default: False
:type overwrite: bool
:param obsolete:
Has a meaning only for obsolete structures. If True, download the obsolete structure
to 'obsolete' folder, otherwise download won't be performed.
This option doesn't work for mmtf format as obsoleted structures aren't stored in mmtf.
Also doesn't have meaning when parameter pdir is specified.
Note: make sure that you are about to download the really obsolete structure.
Trying to download non-obsolete structure into obsolete folder will not work
and you face the "structure doesn't exists" error.
Default: False
:type obsolete: bool
:param pdir: put the file in this directory (default: create a PDB-style directory tree)
:type pdir: string
:return: filename
:rtype: string
"""
file_format = self._print_default_format_warning(
file_format) # Deprecation warning
# Get the compressed PDB structure
code = pdb_code.lower()
archive = {
"pdb": "pdb%s.ent.gz",
"mmCif": "%s.cif.gz",
"xml": "%s.xml.gz",
"mmtf": "%s",
"bundle": "%s-pdb-bundle.tar.gz"
}
archive_fn = archive[file_format] % code
if file_format not in archive.keys():
raise (
"Specified file_format %s doesn't exists or is not supported. Maybe a typo. "
"Please, use one of the following: mmCif, pdb, xml, mmtf, bundle"
% file_format)
if file_format in ("pdb", "mmCif", "xml"):
pdb_dir = "divided" if not obsolete else "obsolete"
file_type = "pdb" if file_format == "pdb" else "mmCIF" if file_format == "mmCif" else "XML"
url = (self.pdb_server + "/pub/pdb/data/structures/%s/%s/%s/%s" %
(pdb_dir, file_type, code[1:3], archive_fn))
elif file_format == "bundle":
url = (self.pdb_server +
"/pub/pdb/compatible/pdb_bundle/%s/%s/%s" %
(code[1:3], code, archive_fn))
else:
url = ("http://mmtf.rcsb.org/v1.0/full/%s" % code)
# Where does the final PDB file get saved?
if pdir is None:
path = self.local_pdb if not obsolete else self.obsolete_pdb
if not self.flat_tree: # Put in PDB-style directory tree
path = os.path.join(path, code[1:3])
else: # Put in specified directory
path = pdir
if not os.access(path, os.F_OK):
os.makedirs(path)
filename = os.path.join(path, archive_fn)
final = {
"pdb": "pdb%s.ent",
"mmCif": "%s.cif",
"xml": "%s.xml",
"mmtf": "%s.mmtf",
"bundle": "%s-pdb-bundle.tar"
}
final_file = os.path.join(path, final[file_format] % code)
# Skip download if the file already exists
if not overwrite:
if os.path.exists(final_file):
if self._verbose:
print("Structure exists: '%s' " % final_file)
return final_file
# Retrieve the file
if self._verbose:
print("Downloading PDB structure '%s'..." % pdb_code)
try:
_urlcleanup()
_urlretrieve(url, filename)
except IOError:
print("Desired structure doesn't exists")
else:
with gzip.open(filename, "rb") as gz:
with open(final_file, "wb") as out:
out.writelines(gz)
os.remove(filename)
return final_file
def main():
try:
opts, args = getopt.getopt(
sys.argv[1:], "hp:o:i:",
["help", "usage", "pdb=", "output=", "input="])
except getopt.GetoptError:
# show help information and exit:
usage()
sys.exit(2)
input = None
in_handle = None
output = None
pdb_url = None
cla_url = None
raf_url = None
for o, a in opts:
if o in ("-h", "--help", "--usage"):
usage()
sys.exit()
elif o in ("-o", "--output"):
output = a
elif o in ("-i", "--input"):
input = a
elif o in ("-p", "--pdb"):
pdb_url = a
if len(args) < 2:
sys.stderr.write(
"Not enough arguments. Try --help for more details.\n")
sys.exit(2)
raf_url = args[0]
cla_url = args[1]
(raf_filename, headers) = _urlretrieve(raf_url)
seqMapIndex = Raf.SeqMapIndex(raf_filename)
(cla_filename, headers) = _urlretrieve(cla_url)
claIndex = Cla.Index(cla_filename)
if input is None:
sids = args[2:]
elif input == '-':
sids = sys.stdin
else:
in_handle = open(input)
sids = in_handle
try:
for sid in sids:
if not sid or sid[0:1] == '#':
continue
id = sid[0:7]
pdbid = id[1:5]
s = pdbid[0:1]
if s == '0' or s == 's':
sys.stderr.write("No coordinates for domain %s\n" % id)
continue
if output is None:
filename = id + ".ent"
out_handle = open(filename, "w+")
elif output == '-':
out_handle = sys.stdout
else:
out_handle = open(output, "w+")
try:
try:
claRec = claIndex[id]
residues = claRec.residues
seqMap = seqMapIndex.getSeqMap(residues)
pdbid = residues.pdbid
f = open_pdb(pdbid, pdb_url)
try:
seqMap.getAtoms(f, out_handle)
finally:
f.close()
except (IOError, KeyError, RuntimeError) as e:
sys.stderr.write("I cannot do SCOP domain %s : %s\n" %
(id, e))
finally:
out_handle.close()
finally:
if in_handle is not None:
in_handle.close()
def open_pdb(pdbid, pdb_url=None):
if pdb_url is None:
pdb_url = default_pdb_url
url = pdb_url % pdbid
fn, header = _urlretrieve(url)
return open(fn)
def retrieve_pdb_file(
self,
pdb_code,
obsolete=False,
pdir=None,
file_format=None,
overwrite=False,
biounit=False):
"""
Replacement for Bio.PDB.PDBList.retrieve_pdb_file to support
MMB PDB API. Defaults to Biopython super() if standard server is used.
"""
if self.pdb_server != 'mmb':
return super().retrieve_pdb_file(
pdb_code, obsolete, pdir, file_format, overwrite
)
self._verbose = True
code = pdb_code.lower()
if file_format in ('pdb', 'mmCif', 'xml'):
if file_format == 'mmCif':
file_format = 'cif'
if not biounit:
url = (URL_PREFIX + '/%s.%s' % (code, file_format))
else:
file_format = 'pdb'
url = (URL_PREFIX + '/%s_bn%s.pdb' % (code, biounit))
else:
print('Error: MMB Server: File format', file_format, 'not supported')
sys.exit(1)
#Where does the final PDB file get saved?
if pdir is None:
path = self.local_pdb if not obsolete else self.obsolete_pdb
if not self.flat_tree: # Put in PDB-style directory tree
path = os.path.join(path, code[1:3])
else: # Put in specified directory
path = pdir
if not os.access(path, os.F_OK):
os.makedirs(path)
if biounit:
final = {
'pdb': '%s_%s.pdb',
'mmCif': '%s_%s.cif',
'cif': '%s_%s.cif',
'xml': '%s_%s.xml'
}
final_file = os.path.join(path, final[file_format] % (code, biounit))
else:
final = {
'pdb': '%s.pdb',
'mmCif': '%s.cif',
'cif': '%s.cif',
'xml': '%s.xml'
}
final_file = os.path.join(path, final[file_format] % code)
# Skip download if the file already exists
if not overwrite:
if os.path.exists(final_file):
if self._verbose:
print("Structure exists: '%s' " % final_file)
return final_file
# Retrieve the file
if self._verbose:
if biounit:
print("Downloading PDB structure '%s.%s'..." % (pdb_code, biounit))
else:
print("Downloading PDB structure '%s'..." % pdb_code)
try:
_urlcleanup()
_urlretrieve(url, final_file)
except IOError:
print("Desired structure doesn't exists")
return final_file