def performDSSP(pdb, parseall=False, stderr=True):
"""Perform DSSP calculations and parse results. DSSP data is returned
in an :class:`~.AtomGroup` instance. See also :func:`execDSSP`
and :func:`parseDSSP`."""
pdb = fetchPDB(pdb, compressed=False)
return parseDSSP(execDSSP(pdb, stderr=stderr), parsePDB(pdb), parseall)
def parsePDB(pdb, **kwargs):
"""Return an :class:`.AtomGroup` and/or dictionary containing header data
parsed from a PDB file.
This function extends :func:`.parsePDBStream`.
|example| See :ref:`parsepdb` for a detailed example.
:arg pdb: a PDB identifier or a filename
If needed, PDB files are downloaded using :func:`.fetchPDB()` function.
"""
title = kwargs.get('title', None)
if not os.path.isfile(pdb):
if len(pdb) == 4 and pdb.isalnum():
if title is None:
title = pdb
kwargs['title'] = title
filename = fetchPDB(pdb)
if filename is None:
raise IOError('PDB file for {0:s} could not be downloaded.'
.format(pdb))
pdb = filename
else:
raise IOError('{0:s} is not a valid filename or a valid PDB '
'identifier.'.format(pdb))
if title is None:
title, ext = os.path.splitext(os.path.split(pdb)[1])
if ext == '.gz':
title, ext = os.path.splitext(title)
kwargs['title'] = title
pdb = openFile(pdb)
result = parsePDBStream(pdb, **kwargs)
pdb.close()
return result
def performSTRIDE(pdb):
"""Perform STRIDE calculations and parse results. STRIDE data is
returned in an :class:`~.AtomGroup` instance. See also
:func:`execSTRIDE` and :func:`parseSTRIDE`."""
pdb = fetchPDB(pdb, compressed=False)
return parseSTRIDE(execSTRIDE(pdb), parsePDB(pdb))
def execDSSP(pdb, outputname=None, outputdir=None, stderr=True):
"""Execute DSSP for given *pdb*. *pdb* can be a PDB identifier or a PDB
file path. If *pdb* is a compressed file, it will be decompressed using
Python :mod:`gzip` library. When no *outputname* is given, output name
will be :file:`pdb.dssp`. :file:`.dssp` extension will be appended
automatically to *outputname*. If :file:`outputdir` is given, DSSP
output and uncompressed PDB file will be written into this folder.
Upon successful execution of :command:`dssp pdb > out` command, output
filename is returned. On Linux platforms, when *stderr* is false,
standard error messages are suppressed, i.e.
``dssp pdb > outputname 2> /dev/null``.
For more information on DSSP see http://swift.cmbi.ru.nl/gv/dssp/.
If you benefited from DSSP, please consider citing [WK83]_."""
dssp = which('dssp')
if dssp is None:
raise EnvironmentError('command not found: dssp executable is not '
'found in one of system paths')
assert outputname is None or isinstance(outputname, str),\
'outputname must be a string'
assert outputdir is None or isinstance(outputdir, str),\
'outputdir must be a string'
if not os.path.isfile(pdb):
pdb = fetchPDB(pdb, compressed=False)
if pdb is None:
raise ValueError('pdb is not a valid PDB identifier or filename')
if os.path.splitext(pdb)[1] == '.gz':
if outputdir is None:
pdb = gunzip(pdb, os.path.splitext(pdb)[0])
else:
pdb = gunzip(pdb, os.path.join(outputdir,
os.path.split(os.path.splitext(pdb)[0])[1]))
if outputdir is None:
outputdir = '.'
if outputname is None:
out = os.path.join(outputdir,
os.path.splitext(os.path.split(pdb)[1])[0] + '.dssp')
else:
out = os.path.join(outputdir, outputname + '.dssp')
if not stderr and PLATFORM != 'Windows':
status = os.system('{0:s} {1:s} > {2:s} 2> /dev/null'.format(
dssp, pdb, out))
else:
status = os.system('{0:s} {1:s} > {2:s}'.format(dssp, pdb, out))
if status == 0:
return out
def execSTRIDE(pdb, outputname=None, outputdir=None):
"""Execute STRIDE program for given *pdb*. *pdb* can be an identifier or
a PDB file path. If *pdb* is a compressed file, it will be decompressed
using Python :mod:`gzip` library. When no *outputname* is given, output
name will be :file:`pdb.stride`. :file:`.stride` extension will be
appended automatically to *outputname*. If :file:`outputdir` is given,
STRIDE output and uncompressed PDB file will be written into this folder.
Upon successful execution of :command:`stride pdb > out` command, output
filename is returned.
For more information on STRIDE see http://webclu.bio.wzw.tum.de/stride/.
If you benefited from STRIDE, please consider citing [DF95]_."""
stride = which('stride')
if stride is None:
raise EnvironmentError('command not found: stride executable is not '
'found in one of system paths')
assert outputname is None or isinstance(outputname, str),\
'outputname must be a string'
assert outputdir is None or isinstance(outputdir, str),\
'outputdir must be a string'
if not os.path.isfile(pdb):
pdb = fetchPDB(pdb, compressed=False)
if pdb is None:
raise ValueError('pdb is not a valid PDB identifier or filename')
if os.path.splitext(pdb)[1] == '.gz':
if outputdir is None:
pdb = gunzip(pdb, os.path.splitext(pdb)[0])
else:
pdb = gunzip(pdb, os.path.join(outputdir,
os.path.split(os.path.splitext(pdb)[0])[1]))
if outputdir is None:
outputdir = '.'
if outputname is None:
out = os.path.join(outputdir,
os.path.splitext(os.path.split(pdb)[1])[0] + '.stride')
else:
out = os.path.join(outputdir, outputname + '.stride')
status = os.system('{0:s} {1:s} > {2:s}'.format(stride, pdb, out))
if status == 0:
return out
def parsePDBHeader(pdb, *keys):
"""Return header data dictionary for *pdb*. This function is equivalent to
``parsePDB(pdb, header=True, model=0, meta=False)``, likewise *pdb* may be
an identifier or a filename.
List of header records that are parsed.
============ ================= ============================================
Record type Dictionary key(s) Description
============ ================= ============================================
HEADER | classification | molecule classification
| deposition_date | deposition date
| identifier | PDB identifier
TITLE title title for the experiment or analysis
SPLIT split list of PDB entries that make up the whole
structure when combined with this one
COMPND polymers see :class:`Polymer`
EXPDTA experiment information about the experiment
NUMMDL n_models number of models
MDLTYP model_type additional structural annotation
AUTHOR authors list of contributors
JRNL reference reference information dictionary:
* *authors*: list of authors
* *title*: title of the article
* *editors*: list of editors
* *issn*:
* *reference*: journal, vol, issue, etc.
* *publisher*: publisher information
* *pmid*: pubmed identifier
* *doi*: digital object identifier
DBREF[1|2] polymers see :class:`Polymer` and :class:`DBRef`
SEQADV polymers see :class:`Polymer`
SEQRES polymers see :class:`Polymer`
MODRES polymers see :class:`Polymer`
HELIX polymers see :class:`Polymer`
SHEET polymers see :class:`Polymer`
HET chemicals see :class:`Chemical`
HETNAM chemicals see :class:`Chemical`
HETSYN chemicals see :class:`Chemical`
FORMUL chemicals see :class:`Chemical`
REMARK 2 resolution resolution of structures, when applicable
REMARK 4 version PDB file version
REMARK 350 biomoltrans biomolecular transformation lines
(unprocessed)
============ ================= ============================================
Header records that are not parsed are: OBSLTE, CAVEAT, SOURCE, KEYWDS,
REVDAT, SPRSDE, SSBOND, LINK, CISPEP, CRYST1, ORIGX1, ORIGX2, ORIGX3,
MTRIX1, MTRIX2, MTRIX3, and REMARK X not mentioned above.
"""
if not os.path.isfile(pdb):
if len(pdb) == 4 and pdb.isalnum():
filename = fetchPDB(pdb)
if filename is None:
raise IOError('PDB file for {0:s} could not be downloaded.'
.format(pdb))
pdb = filename
else:
raise IOError('{0:s} is not a valid filename or a valid PDB '
'identifier.'.format(pdb))
pdb = openFile(pdb)
header, _ = getHeaderDict(pdb, *keys)
pdb.close()
return header
