def _getPolymers(lines):
"""Returns list of polymers (macromolecules)."""
pdbid = lines['pdbid']
polymers = dict()
for i, line in lines['SEQRES']:
ch = line[11]
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.sequence += ''.join(getSequence(line[19:].split()))
for i, line in lines['DBREF ']:
i += 1
ch = line[12]
if ch == ' ':
if not len(polymers) == 1:
LOGGER.warn('DBREF chain identifier is not specified '
'({0}:{1})'.format(pdbid, i))
continue
else:
ch = list(polymers)[0]
dbabbr = line[26:32].strip()
dbref = DBRef()
dbref.dbabbr = dbabbr
dbref.database = _PDB_DBREF.get(dbabbr, 'Unknown')
dbref.accession = line[33:41].strip()
dbref.idcode = line[42:54].strip()
try:
first = int(line[14:18])
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'initial sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
last = int(line[20:24])
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'ending sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
dbref.first = (first, line[18], int(line[56:60]))
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'initial sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
dbref.last = (last, line[24].strip(), int(line[62:67]))
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'ending sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.dbrefs.append(dbref)
dbref1 = lines['DBREF1']
dbref2 = lines['DBREF2']
if len(dbref1) != len(dbref2):
LOGGER.warn('DBREF1 and DBREF1 records are not complete')
dbref12 = []
else:
dbref12 = zip(dbref1, dbref2) # PY3K: OK
for dbref1, dbref2 in dbref12:
i, line = dbref1
i += 1
ch = line[12]
dbabbr = line[26:32].strip()
dbref = DBRef()
dbref.dbabbr = dbabbr
dbref.database = _PDB_DBREF.get(dbabbr, 'Unknown')
dbref.idcode = line[47:67].strip()
try:
first = int(line[14:18])
except:
LOGGER.warn('DBREF1 for chain {2}: failed to parse '
'initial sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
last = int(line[20:24])
except:
LOGGER.warn('DBREF1 for chain {2}: failed to parse '
'ending sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
i, line = dbref2
i += 1
if line[12] == ' ':
LOGGER.warn('DBREF2 chain identifier is not specified '
'({0}:{1})'.format(pdbid, i, ch))
elif line[12] != ch:
LOGGER.warn('DBREF1 and DBREF2 chain id mismatch'
'({0}:{1})'.format(pdbid, i, ch))
dbref.accession = line[18:40].strip()
try:
dbref.first = (first, line[18].strip(), int(line[45:55]))
except:
LOGGER.warn('DBREF2 for chain {2}: failed to parse '
'initial sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
dbref.last = (last, line[24].strip(), int(line[57:67]))
except:
LOGGER.warn('DBREF2 for chain {2}: failed to parse '
'ending sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.dbrefs.append(dbref)
for poly in polymers.values(): # PY3K: OK
resnum = []
for dbref in poly.dbrefs:
dbabbr = dbref.dbabbr
if dbabbr == 'PDB':
if not (pdbid == dbref.accession == dbref.idcode):
LOGGER.warn('DBREF for chain {2} refers to PDB '
'entry {3} ({0}:{1})'.format(
pdbid, i, ch, dbref.accession))
else:
if pdbid == dbref.accession or pdbid == dbref.idcode:
LOGGER.warn('DBREF for chain {2} is {3}, '
'expected PDB ({0}:{1})'.format(
pdbid, i, ch, dbabbr))
dbref.database = 'PDB'
resnum.append((dbref.first[0], dbref.last[0]))
resnum.sort()
last = -10000
for first, temp in resnum:
if first <= last:
LOGGER.warn('DBREF records overlap for chain {0} ({1})'.format(
poly.chid, pdbid))
last = temp
for i, line in lines['MODRES']:
ch = line[16]
if ch == ' ':
if not len(polymers) == 1:
LOGGER.warn('MODRES chain identifier is not specified '
'({0}:{1})'.format(pdbid, i))
continue
else:
ch = list(polymers)[0]
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
if poly.modified is None:
poly.modified = []
poly.modified.append(
(line[12:15].strip(), line[18:22].strip() + line[22].strip(),
line[24:27].strip(), line[29:70].strip()))
for i, line in lines['SEQADV']:
i += 1
ch = line[16]
if ch == ' ':
if not len(polymers) == 1:
LOGGER.warn('MODRES chain identifier is not specified '
'({0}:{1})'.format(pdbid, i))
continue
else:
ch = list(polymers)[0]
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
dbabbr = line[24:28].strip()
resname = line[12:15].strip()
try:
resnum = int(line[18:22].strip())
except:
continue
LOGGER.warn('SEQADV for chain {2}: failed to parse PDB sequence '
'number ({0}:{1})'.format(pdbid, i, ch))
icode = line[22].strip()
try:
dbnum = int(line[43:48].strip())
except:
continue
LOGGER.warn('SEQADV for chain {2}: failed to parse database '
'sequence number ({0}:{1})'.format(pdbid, i, ch))
comment = line[49:70].strip()
match = False
for dbref in poly.dbrefs:
if not dbref.first[0] <= resnum <= dbref.last[0]:
continue
match = True
if dbref.dbabbr != dbabbr:
LOGGER.warn('SEQADV for chain {2}: reference database '
'mismatch, expected {3} parsed {4} '
'({0}:{1})'.format(pdbid, i,
ch, repr(dbref.dbabbr),
repr(dbabbr)))
continue
dbacc = line[29:38].strip()
if dbref.accession[:9] != dbacc[:9]:
LOGGER.warn('SEQADV for chain {2}: accession code '
'mismatch, expected {3} parsed {4} '
'({0}:{1})'.format(pdbid, i, ch,
repr(dbref.accession),
repr(dbacc)))
continue
dbref.diff.append((resname, resnum, icode, dbnum, dbnum, comment))
if not match:
continue
LOGGER.warn('SEQADV for chain {2}: database sequence reference '
'not found ({0}:{1})'.format(pdbid, i, ch))
string = ' '.join([line[10:].strip() for i, line in lines['COMPND']])
if string.startswith('MOL_ID'):
dict_ = {}
for molecule in string[6:].split('MOL_ID'):
dict_.clear()
for token in molecule.split(';'):
token = token.strip()
if not token:
continue
items = token.split(':', 1)
if len(items) == 2:
key, value = items
dict_[key.strip()] = value.strip()
chains = dict_.pop('CHAIN', '').strip()
if not chains:
continue
for ch in chains.split(','):
ch = ch.strip()
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.name = dict_.get('MOLECULE', '')
poly.fragment = dict_.get('FRAGMENT', '')
poly.comments = dict_.get('OTHER_DETAILS', '')
val = dict_.get('SYNONYM', '')
poly.synonyms = [s.strip()
for s in val.split(',')] if val else []
val = dict_.get('EC', '')
poly.ec = [s.strip() for s in val.split(',')] if val else []
poly.engineered = dict_.get('ENGINEERED', '') == 'YES'
poly.mutation = dict_.get('MUTATION', '') == 'YES'
return list(polymers.values())
def _getPolymers(lines):
"""Returns list of polymers (macromolecules)."""
pdbid = lines['pdbid']
polymers = dict()
for i, line in lines['SEQRES']:
ch = line[11]
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.sequence += ''.join(getSequence(line[19:].split()))
for i, line in lines['DBREF ']:
i += 1
ch = line[12]
if ch == ' ':
if not len(polymers) == 1:
LOGGER.warn('DBREF chain identifier is not specified '
'({0}:{1})'.format(pdbid, i))
continue
else:
ch = list(polymers)[0]
dbabbr = line[26:32].strip()
dbref = DBRef()
dbref.dbabbr = dbabbr
dbref.database = _PDB_DBREF.get(dbabbr, 'Unknown')
dbref.accession = line[33:41].strip()
dbref.idcode = line[42:54].strip()
try:
first = int(line[14:18])
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'initial sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
last = int(line[20:24])
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'ending sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
dbref.first = (first, line[18], int(line[56:60]))
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'initial sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
dbref.last = (last, line[24].strip(), int(line[62:67]))
except:
LOGGER.warn('DBREF for chain {2}: failed to parse '
'ending sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.dbrefs.append(dbref)
dbref1 = lines['DBREF1']
dbref2 = lines['DBREF2']
if len(dbref1) != len(dbref2):
LOGGER.warn('DBREF1 and DBREF1 records are not complete')
dbref12 = []
else:
dbref12 = zip(dbref1, dbref2) # PY3K: OK
for dbref1, dbref2 in dbref12:
i, line = dbref1
i += 1
ch = line[12]
dbabbr = line[26:32].strip()
dbref = DBRef()
dbref.dbabbr = dbabbr
dbref.database = _PDB_DBREF.get(dbabbr, 'Unknown')
dbref.idcode = line[47:67].strip()
try:
first = int(line[14:18])
except:
LOGGER.warn('DBREF1 for chain {2}: failed to parse '
'initial sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
last = int(line[20:24])
except:
LOGGER.warn('DBREF1 for chain {2}: failed to parse '
'ending sequence number of the PDB sequence '
'({0}:{1})'.format(pdbid, i, ch))
i, line = dbref2
i += 1
if line[12] == ' ':
LOGGER.warn('DBREF2 chain identifier is not specified '
'({0}:{1})'.format(pdbid, ch))
elif line[12] != ch:
LOGGER.warn('DBREF1 and DBREF2 chain id mismatch'
'({0}:{1})'.format(pdbid, ch))
dbref.accession = line[18:40].strip()
try:
dbref.first = (first, line[18].strip(), int(line[45:55]))
except:
LOGGER.warn('DBREF2 for chain {2}: failed to parse '
'initial sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
try:
dbref.last = (last, line[24].strip(), int(line[57:67]))
except:
LOGGER.warn('DBREF2 for chain {2}: failed to parse '
'ending sequence number of the database sequence '
'({0}:{1})'.format(pdbid, i, ch))
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.dbrefs.append(dbref)
for poly in polymers.values(): # PY3K: OK
resnum = []
for dbref in poly.dbrefs:
dbabbr = dbref.dbabbr
if dbabbr == 'PDB':
if not (pdbid == dbref.accession == dbref.idcode):
LOGGER.warn('DBREF for chain {2} refers to PDB '
'entry {3} ({0}:{1})'
.format(pdbid, i, ch, dbref.accession))
else:
if pdbid == dbref.accession or pdbid == dbref.idcode:
LOGGER.warn('DBREF for chain {2} is {3}, '
'expected PDB ({0}:{1})'
.format(pdbid, i, ch, dbabbr))
dbref.database = 'PDB'
resnum.append((dbref.first[0], dbref.last[0]))
resnum.sort()
last = -10000
for first, temp in resnum:
if first <= last:
LOGGER.warn('DBREF records overlap for chain {0} ({1})'
.format(poly.chid, pdbid))
last = temp
for i, line in lines['MODRES']:
ch = line[16]
if ch == ' ':
if not len(polymers) == 1:
LOGGER.warn('MODRES chain identifier is not specified '
'({0}:{1})'.format(pdbid, i))
continue
else:
ch = list(polymers)[0]
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
if poly.modified is None:
poly.modified = []
poly.modified.append((line[12:15].strip(), line[18:22].strip() +
line[22].strip(), line[24:27].strip(),
line[29:70].strip()))
for i, line in lines['SEQADV']:
i += 1
ch = line[16]
if ch == ' ':
if not len(polymers) == 1:
LOGGER.warn('MODRES chain identifier is not specified '
'({0}:{1})'.format(pdbid, i))
continue
else:
ch = list(polymers)[0]
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
dbabbr = line[24:28].strip()
resname = line[12:15].strip()
try:
resnum = int(line[18:22].strip())
except:
#LOGGER.warn('SEQADV for chain {2}: failed to parse PDB sequence '
# 'number ({0}:{1})'.format(pdbid, i, ch))
continue
icode = line[22].strip()
try:
dbnum = int(line[43:48].strip())
except:
#LOGGER.warn('SEQADV for chain {2}: failed to parse database '
# 'sequence number ({0}:{1})'.format(pdbid, i, ch))
continue
comment = line[49:70].strip()
match = False
for dbref in poly.dbrefs:
if not dbref.first[0] <= resnum <= dbref.last[0]:
continue
match = True
if dbref.dbabbr != dbabbr:
LOGGER.warn('SEQADV for chain {2}: reference database '
'mismatch, expected {3} parsed {4} '
'({0}:{1})'.format(pdbid, i, ch,
repr(dbref.dbabbr), repr(dbabbr)))
continue
dbacc = line[29:38].strip()
if dbref.accession[:9] != dbacc[:9]:
LOGGER.warn('SEQADV for chain {2}: accession code '
'mismatch, expected {3} parsed {4} '
'({0}:{1})'.format(pdbid, i, ch,
repr(dbref.accession), repr(dbacc)))
continue
dbref.diff.append((resname, resnum, icode, dbnum, dbnum, comment))
if not match:
LOGGER.warn('SEQADV for chain {2}: database sequence reference '
'not found ({0}:{1})'.format(pdbid, i, ch))
continue
string = ' '.join([line[10:].strip() for i, line in lines['COMPND']])
if string.startswith('MOL_ID'):
dict_ = {}
for molecule in string[6:].split('MOL_ID'):
dict_.clear()
for token in molecule.split(';'):
token = token.strip()
if not token:
continue
items = token.split(':', 1)
if len(items) == 2:
key, value = items
dict_[key.strip()] = value.strip()
chains = dict_.pop('CHAIN', '').strip()
if not chains:
continue
for ch in chains.split(','):
ch = ch.strip()
poly = polymers.get(ch, Polymer(ch))
polymers[ch] = poly
poly.name = dict_.get('MOLECULE', '')
poly.fragment = dict_.get('FRAGMENT', '')
poly.comments = dict_.get('OTHER_DETAILS', '')
val = dict_.get('SYNONYM', '')
poly.synonyms = [s.strip() for s in val.split(',')
] if val else []
val = dict_.get('EC', '')
poly.ec = [s.strip() for s in val.split(',')] if val else []
poly.engineered = dict_.get('ENGINEERED', '') == 'YES'
poly.mutation = dict_.get('MUTATION', '') == 'YES'
return list(polymers.values())