def parsePQR(filename, **kwargs):
"""Returns an :class:`.AtomGroup` containing data parsed from PDB lines.
:arg filename: a PQR filename
:type filename: str"""
title = kwargs.get('title', kwargs.get('name'))
model = 1
header = False
chain = kwargs.get('chain')
subset = kwargs.get('subset')
altloc = kwargs.get('altloc', 'A')
max_n_atoms = kwargs.get('max_n_atoms', 1e5)
if not os.path.isfile(filename):
raise IOError('No such file: {0}'.format(repr(filename)))
if title is None:
fn, ext = os.path.splitext(os.path.split(filename)[1])
if ext == '.gz':
fn, ext = os.path.splitext(fn)
title = fn.lower()
title_suffix = ''
if subset:
try:
subset = _PDBSubsets[subset.lower()]
except AttributeError:
raise TypeError('subset must be a string')
except KeyError:
raise ValueError('{0} is not a valid subset'
.format(repr(subset)))
title_suffix = '_' + subset
if chain is not None:
if not isinstance(chain, str):
raise TypeError('chain must be a string')
elif len(chain) == 0:
raise ValueError('chain must not be an empty string')
title_suffix = '_' + chain + title_suffix
if 'ag' in kwargs:
ag = kwargs['ag']
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
n_csets = ag.numCoordsets()
else:
ag = AtomGroup(title + title_suffix)
n_csets = 0
pqr = openFile(filename, 'rt')
lines = pqr.readlines()
pqr.close()
LOGGER.timeit()
ag = _parsePDBLines(ag, lines, split=0, model=1, chain=chain,
subset=subset, altloc_torf=False, format='pqr',
max_n_atoms=max_n_atoms)
if ag.numAtoms() > 0:
LOGGER.report('{0} atoms and {1} coordinate sets were '
'parsed in %.2fs.'.format(ag.numAtoms(),
ag.numCoordsets() - n_csets))
return ag
else:
return None
def parsePQR(filename, **kwargs):
"""Returns an :class:`.AtomGroup` containing data parsed from PDB lines.
:arg filename: a PQR filename
:type filename: str"""
title = kwargs.get('title', kwargs.get('name'))
chain = kwargs.get('chain')
subset = kwargs.get('subset')
if not os.path.isfile(filename):
raise IOError('No such file: {0}'.format(repr(filename)))
if title is None:
fn, ext = os.path.splitext(os.path.split(filename)[1])
if ext == '.gz':
fn, ext = os.path.splitext(fn)
title = fn.lower()
title_suffix = ''
if subset:
try:
subset = _PDBSubsets[subset.lower()]
except AttributeError:
raise TypeError('subset must be a string')
except KeyError:
raise ValueError('{0} is not a valid subset'.format(repr(subset)))
title_suffix = '_' + subset
if chain is not None:
if not isinstance(chain, str):
raise TypeError('chain must be a string')
elif len(chain) == 0:
raise ValueError('chain must not be an empty string')
title_suffix = '_' + chain + title_suffix
if 'ag' in kwargs:
ag = kwargs['ag']
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
n_csets = ag.numCoordsets()
else:
ag = AtomGroup(title + title_suffix)
n_csets = 0
pqr = openFile(filename, 'rt')
lines = pqr.readlines()
pqr.close()
LOGGER.timeit()
ag = _parsePDBLines(ag,
lines,
split=0,
model=1,
chain=chain,
subset=subset,
altloc_torf=False,
format='pqr')
if ag.numAtoms() > 0:
LOGGER.report('{0} atoms and {1} coordinate sets were '
'parsed in %.2fs.'.format(ag.numAtoms(),
ag.numCoordsets() - n_csets))
return ag
else:
return None
def parseEMDStream(stream, **kwargs):
""" Returns an :class:`.AtomGroup` containing EMD data parsed from a stream of EMD file.
:arg stream: Anything that implements the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
cutoff = float(kwargs.get('cutoff', 1.20))
n_nodes = int(kwargs.get('n_nodes', 1000))
num_iter = int(kwargs.get('num_iter', 20))
ag = None
title_suffix = ''
if 'ag' in kwargs:
ag = kwargs['ag']
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
n_csets = ag.numCoordsets()
else:
ag = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
n_csets = 0
biomol = kwargs.get('biomol', False)
hd = None
LOGGER.warn('Building coordinates from electron density map. This may take a while.')
LOGGER.timeit()
_parseEMDLines(ag, stream, cutoff=cutoff, n_nodes=n_nodes, num_iter=num_iter, format='EMD')
LOGGER.report('{0} atoms and {1} coordinate sets were '
'parsed in %.2fs.'.format(ag.numAtoms(),
ag.numCoordsets() - n_csets))
return ag
def parseEMDStream(stream, **kwargs):
""" Returns an :class:`.AtomGroup` containing EMD data parsed from a stream of EMD file.
:arg stream: Any object with the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
cutoff = kwargs.get('cutoff', None)
if cutoff is not None:
cutoff = float(cutoff)
n_nodes = kwargs.get('n_nodes', 0)
num_iter = int(kwargs.get('num_iter', 20))
map = kwargs.get('map', False)
make_nodes = kwargs.get('make_nodes', False)
if n_nodes > 0:
make_nodes = True
n_nodes = int(n_nodes)
if map is False and make_nodes is False:
LOGGER.warn(
'At least one of map and make_nodes should be True. '
'Setting map to False was an intentional change from the default '
'behaviour so make_nodes has been set to True with n_nodes=1000.')
make_nodes = True
n_nodes = 1000
title_suffix = kwargs.get('title_suffix', '')
atomgroup = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
atomgroup._n_atoms = n_nodes
if make_nodes:
LOGGER.info(
'Building coordinates from electron density map. This may take a while.'
)
LOGGER.timeit()
if map:
emd, atomgroup = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, map=map, make_nodes=make_nodes)
else:
atomgroup = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, map=map, make_nodes=make_nodes)
LOGGER.report('{0} pseudoatoms were fitted in %.2fs.'.format(
atomgroup.numAtoms(), atomgroup.numCoordsets()))
else:
emd = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, map=map, make_nodes=make_nodes)
if make_nodes:
if map:
return emd, atomgroup
else:
return atomgroup
else:
return emd
def parseEMDStream(stream, **kwargs):
""" Returns an :class:`.AtomGroup` containing EMD data parsed from a stream of EMD file.
:arg stream: Anything that implements the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
cutoff = kwargs.get('cutoff', None)
if cutoff is not None:
cutoff = float(cutoff)
n_nodes = int(kwargs.get('n_nodes', 1000))
num_iter = int(kwargs.get('num_iter', 20))
return_map = kwargs.get('return_map',False)
make_nodes = kwargs.get('make_nodes',False)
if return_map is False and make_nodes is False:
LOGGER.warn('At least one of return_map and make_nodes should be True. '
'Setting make_nodes to False was an intentional change from the default '
'so return_map has been set to True.')
kwargs['return_map'] = True
title_suffix = kwargs.get('title_suffix','')
atomgroup = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
if make_nodes:
LOGGER.info('Building coordinates from electron density map. This may take a while.')
LOGGER.timeit()
if return_map:
emd, atomgroup = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, return_map=return_map, \
make_nodes=make_nodes)
else:
atomgroup = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, return_map=return_map, \
make_nodes=make_nodes)
LOGGER.report('{0} atoms and {1} coordinate sets were '
'parsed in %.2fs.'.format(atomgroup.numAtoms(), atomgroup.numCoordsets()))
else:
emd = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, return_map=return_map, \
make_nodes=make_nodes)
if make_nodes:
if return_map:
return emd, atomgroup
else:
return atomgroup
else:
return emd
def parseEMDStream(stream, **kwargs):
""" Returns an :class:`.AtomGroup` containing EMD data parsed from a stream of EMD file.
:arg stream: Any object with the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
cutoff = kwargs.get('cutoff', None)
if cutoff is not None:
cutoff = float(cutoff)
n_nodes = int(kwargs.get('n_nodes', 1000))
num_iter = int(kwargs.get('num_iter', 20))
map = kwargs.get('map',True)
make_nodes = kwargs.get('make_nodes',False)
if map is False and make_nodes is False:
LOGGER.warn('At least one of map and make_nodes should be True. '
'Setting map to False was an intentional change from the default '
'behaviour so make_nodes has been set to True.')
make_nodes = True
title_suffix = kwargs.get('title_suffix','')
atomgroup = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
atomgroup._n_atoms = n_nodes
if make_nodes:
LOGGER.info('Building coordinates from electron density map. This may take a while.')
LOGGER.timeit()
if map:
emd, atomgroup = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, map=map, make_nodes=make_nodes)
else:
atomgroup = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, map=map, make_nodes=make_nodes)
LOGGER.report('{0} atoms and {1} coordinate sets were '
'parsed in %.2fs.'.format(atomgroup.numAtoms(), atomgroup.numCoordsets()))
else:
emd = _parseEMDLines(atomgroup, stream, cutoff=cutoff, n_nodes=n_nodes, \
num_iter=num_iter, map=map, make_nodes=make_nodes)
if make_nodes:
if map:
return emd, atomgroup
else:
return atomgroup
else:
return emd
def parsePSF(filename, title=None, ag=None):
"""Returns an :class:`.AtomGroup` instance storing data parsed from X-PLOR
format PSF file *filename*. Atom and bond information is parsed from the
file. If *title* is not given, *filename* will be set as the title of the
:class:`.AtomGroup` instance. An :class:`.AtomGroup` instance may be
provided as *ag* argument. When provided, *ag* must have the same number
of atoms in the same order as the file. Data from PSF file will be added
to the *ag*. This may overwrite present data if it overlaps with PSF file
content.
This function now includes the angles, dihedrals, and impropers sections
as well as donors, acceptors and crossterms!"""
if ag is not None:
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
psf = openFile(filename, 'rb')
line = psf.readline()
while line:
line = line.strip()
if line.endswith(b'!NATOM'):
n_atoms = int(line.split(b'!')[0])
break
line = psf.readline()
if title is None:
title = os.path.splitext(os.path.split(filename)[1])[0]
else:
title = str(title)
if ag is None:
ag = AtomGroup(title)
else:
if n_atoms != ag.numAtoms():
raise ValueError('ag and PSF file must have same number of atoms')
serials = zeros(n_atoms, ATOMIC_FIELDS['serial'].dtype)
segnames = zeros(n_atoms, ATOMIC_FIELDS['segment'].dtype)
resnums = zeros(n_atoms, ATOMIC_FIELDS['resnum'].dtype)
resnames = zeros(n_atoms, ATOMIC_FIELDS['resname'].dtype)
atomnames = zeros(n_atoms, ATOMIC_FIELDS['name'].dtype)
atomtypes = zeros(n_atoms, ATOMIC_FIELDS['type'].dtype)
charges = zeros(n_atoms, ATOMIC_FIELDS['charge'].dtype)
masses = zeros(n_atoms, ATOMIC_FIELDS['mass'].dtype)
n = 0
n_bonds = 0
for line in psf:
if line.strip() == b'':
continue
if b'!NBOND:' in line.upper():
items = line.split()
n_bonds = int(items[0])
break
if n + 1 > n_atoms:
continue
if len(line) <= 71:
serials[n] = line[:8]
segnames[n] = line[9:13].strip()
resnums[n] = line[14:19]
resnames[n] = line[19:23].strip()
atomnames[n] = line[24:28].strip()
atomtypes[n] = line[29:35].strip()
charges[n] = line[35:44]
masses[n] = line[50:60]
else:
items = line.split()
serials[n] = items[0]
segnames[n] = items[1]
resnums[n] = items[2]
resnames[n] = items[3]
atomnames[n] = items[4]
atomtypes[n] = items[5]
charges[n] = items[6]
masses[n] = items[7]
n += 1
if n < n_atoms:
raise IOError(
'number of lines in PSF atoms block is less than the number of '
'atoms')
n_angles = 0
lines = []
for i, line in enumerate(psf):
if line.strip() == b'':
continue
if b'!NTHETA' in line:
items = line.split()
n_angles = int(items[0])
break
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
b_array = fromstring(lines, count=n_bonds * 2, dtype=int, sep=' ')
if len(b_array) != n_bonds * 2:
raise IOError('number of bonds expected and parsed do not match')
n_dihedrals = 0
lines = []
for i, line in enumerate(psf):
if line.strip() == b'':
continue
if b'!NPHI' in line:
items = line.split()
n_dihedrals = int(items[0])
break
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
a_array = fromstring(lines, count=n_angles * 3, dtype=int, sep=' ')
if len(a_array) != n_angles * 3:
raise IOError('number of angles expected and parsed do not match')
n_impropers = 0
lines = []
for i, line in enumerate(psf):
if line.strip() == b'':
continue
if b'!NIMPHI' in line:
items = line.split()
n_impropers = int(items[0])
break
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
d_array = fromstring(lines, count=n_dihedrals * 4, dtype=int, sep=' ')
if len(d_array) != n_dihedrals * 4:
raise IOError('number of dihedrals expected and parsed do not match')
n_donors = 0
lines = []
for i, line in enumerate(psf):
if line.strip() == b'':
continue
if b'!NDON' in line:
items = line.split()
n_donors = int(items[0])
break
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
i_array = fromstring(lines, count=n_impropers * 4, dtype=int, sep=' ')
if len(i_array) != n_impropers * 4:
raise IOError('number of impropers expected and parsed do not match')
n_acceptors = 0
lines = []
for i, line in enumerate(psf):
if line.strip() == b'':
continue
if b'!NACC' in line:
items = line.split()
n_acceptors = int(items[0])
break
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
do_array = fromstring(lines, count=n_donors * 2, dtype=int, sep=' ')
if len(do_array) != n_donors * 2:
raise IOError('number of donors expected and parsed do not match')
n_exclusions = 0
lines = []
for i, line in enumerate(psf):
if line.strip() == b'':
continue
if b'!NNB' in line:
items = line.split()
n_exclusions = int(items[0])
break
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
ac_array = fromstring(lines, count=n_acceptors * 2, dtype=int, sep=' ')
if len(ac_array) != n_acceptors * 2:
raise IOError('number of acceptors expected and parsed do not match')
lines = []
for i, line in enumerate(psf):
if line.strip() == b'':
continue
if b'!' in line:
break
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
nbe_array = fromstring(lines, count=n_exclusions * 2, dtype=int, sep=' ')
if len(nbe_array) != n_exclusions * 2:
raise IOError(
'number of nonbonded exclusions expected and parsed do not match')
n_crossterms = 0
for i, line in enumerate(psf):
if b'!NCRTERM' in line:
items = line.split()
n_crossterms = int(items[0])
break
lines = []
for i, line in enumerate(psf):
lines.append(line.decode(encoding='UTF-8'))
lines = ''.join(lines)
c_array = fromstring(lines, count=n_crossterms * 4, dtype=int, sep=' ')
if len(c_array) != n_crossterms * 4:
raise IOError('number of crossterms expected and parsed do not match')
psf.close()
ag.setSerials(serials)
ag.setSegnames(segnames)
ag.setResnums(resnums)
ag.setResnames(resnames)
ag.setNames(atomnames)
ag.setTypes(atomtypes)
ag.setCharges(charges)
ag.setMasses(masses)
if n_bonds > 0:
b_array = add(b_array, -1, b_array)
ag.setBonds(b_array.reshape((n_bonds, 2)))
if n_angles > 0:
a_array = add(a_array, -1, a_array)
ag.setAngles(a_array.reshape((n_angles, 3)))
if n_dihedrals > 0:
d_array = add(d_array, -1, d_array)
ag.setDihedrals(d_array.reshape((n_dihedrals, 4)))
if n_impropers > 0:
i_array = add(i_array, -1, i_array)
ag.setImpropers(i_array.reshape((n_impropers, 4)))
if n_donors > 0:
do_array = add(do_array, -1, do_array)
ag.setDonors(do_array.reshape((n_donors, 2)))
if n_acceptors > 0:
ac_array = add(ac_array, -1, ac_array)
ag.setAcceptors(ac_array.reshape((n_acceptors, 2)))
if n_exclusions > 0:
nbe_array = add(nbe_array, -1, nbe_array)
ag.setNBExclusions(nbe_array.reshape((n_exclusions, 2)))
if n_crossterms > 0:
c_array = add(c_array, -1, c_array)
ag.setCrossterms(c_array.reshape((n_crossterms, 4)))
return ag
def parsePSF(filename, title=None, ag=None):
"""Return an :class:`.AtomGroup` instance storing data parsed from X-PLOR
format PSF file *filename*. Atom and bond information is parsed from the
file. If *title* is not given, *filename* will be set as the title of the
:class:`.AtomGroup` instance. An :class:`.AtomGroup` instance may be
provided as *ag* argument. When provided, *ag* must have the same number
of atoms in the same order as the file. Data from PSF file will be added
to the *ag*. This may overwrite present data if it overlaps with PSF file
content. Note that this function does not evaluate angles, dihedrals, and
impropers sections."""
if ag is not None:
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
psf = openFile(filename, 'rb')
line = psf.readline()
i_line = 1
while line:
line = line.strip()
if line.endswith('!NATOM'):
n_atoms = int(line.split('!')[0])
break
line = psf.readline()
i_line += 1
if title is None:
title = os.path.splitext(os.path.split(filename)[1])[0]
else:
title = str(title)
if ag is None:
ag = AtomGroup(title)
else:
if n_atoms != ag.numAtoms():
raise ValueError('ag and PSF file must have same number of atoms')
serials = zeros(n_atoms, ATOMIC_FIELDS['serial'].dtype)
segnames = zeros(n_atoms, ATOMIC_FIELDS['segment'].dtype)
resnums = zeros(n_atoms, ATOMIC_FIELDS['resnum'].dtype)
resnames = zeros(n_atoms, ATOMIC_FIELDS['resname'].dtype)
atomnames = zeros(n_atoms, ATOMIC_FIELDS['name'].dtype)
atomtypes = zeros(n_atoms, ATOMIC_FIELDS['type'].dtype)
charges = zeros(n_atoms, ATOMIC_FIELDS['charge'].dtype)
masses = zeros(n_atoms, ATOMIC_FIELDS['mass'].dtype)
lines = psf.readlines(71 * (n_atoms + 5))
if len(lines) < n_atoms:
raise IOError('number of lines in PSF is less than the number of '
'atoms')
for i, line in enumerate(lines):
if i == n_atoms:
break
i_line += 1
if len(line) <= 71:
serials[i] = line[:8]
segnames[i] = line[9:13].strip()
resnums[i] = line[14:19]
resnames[i] = line[19:23].strip()
atomnames[i] = line[24:28].strip()
atomtypes[i] = line[29:35].strip()
charges[i] = line[35:44]
masses[i] = line[50:60]
else:
items = line.split()
serials[i] = items[0]
segnames[i] = items[1]
resnums[i] = items[2]
resnames[i] = items[3]
atomnames[i] = items[4]
atomtypes[i] = items[5]
charges[i] = items[6]
masses[i] = items[7]
i = n_atoms
while 1:
line = lines[i].split()
if len(line) >= 2 and line[1] == '!NBOND:':
n_bonds = int(line[0])
break
i += 1
lines = ''.join(lines[i+1:]) + psf.read(n_bonds/4 * 71)
array = fromstring(lines, count=n_bonds*2, dtype=int, sep=' ')
if len(array) != n_bonds*2:
raise IOError('number of bonds expected and parsed do not match')
psf.close()
ag.setSerials(serials)
ag.setSegnames(segnames)
ag.setResnums(resnums)
ag.setResnames(resnames)
ag.setNames(atomnames)
ag.setTypes(atomtypes)
ag.setCharges(charges)
ag.setMasses(masses)
array = add(array, -1, array)
ag.setBonds(array.reshape((n_bonds, 2)))
return ag
def parsePDBStream(stream, **kwargs):
"""Returns an :class:`.AtomGroup` and/or dictionary containing header data
parsed from a stream of PDB lines.
:arg stream: Anything that implements the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
model = kwargs.get('model')
header = kwargs.get('header', False)
assert isinstance(header, bool), 'header must be a boolean'
chain = kwargs.get('chain')
subset = kwargs.get('subset')
altloc = kwargs.get('altloc', 'A')
if model is not None:
if isinstance(model, Integral):
if model < 0:
raise ValueError('model must be greater than 0')
else:
raise TypeError('model must be an integer, {0} is invalid'
.format(str(model)))
title_suffix = ''
if subset:
try:
subset = _PDBSubsets[subset.lower()]
except AttributeError:
raise TypeError('subset must be a string')
except KeyError:
raise ValueError('{0} is not a valid subset'
.format(repr(subset)))
title_suffix = '_' + subset
if chain is not None:
if not isinstance(chain, str):
raise TypeError('chain must be a string')
elif len(chain) == 0:
raise ValueError('chain must not be an empty string')
title_suffix = chain + title_suffix
ag = kwargs.pop('ag', None)
if ag is not None:
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
n_csets = ag.numCoordsets()
elif model != 0:
ag = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
n_csets = 0
biomol = kwargs.get('biomol', False)
auto_secondary = None
secondary = kwargs.get('secondary')
if not secondary:
auto_secondary = SETTINGS.get('auto_secondary')
secondary = auto_secondary
split = 0
hd = None
if model != 0:
LOGGER.timeit()
try:
lines = stream.readlines()
except AttributeError as err:
try:
lines = stream.read().split('\n')
except AttributeError:
raise err
if not len(lines):
raise ValueError('empty PDB file or stream')
if header or biomol or secondary:
hd, split = getHeaderDict(lines)
_parsePDBLines(ag, lines, split, model, chain, subset, altloc)
if ag.numAtoms() > 0:
LOGGER.report('{0} atoms and {1} coordinate set(s) were '
'parsed in %.2fs.'.format(ag.numAtoms(),
ag.numCoordsets() - n_csets))
else:
ag = None
LOGGER.warn('Atomic data could not be parsed, please '
'check the input file.')
elif header:
hd, split = getHeaderDict(stream)
if ag is not None and isinstance(hd, dict):
if secondary:
if auto_secondary:
try:
ag = assignSecstr(hd, ag)
except ValueError:
pass
else:
ag = assignSecstr(hd, ag)
if biomol:
ag = buildBiomolecules(hd, ag)
if isinstance(ag, list):
LOGGER.info('Biomolecular transformations were applied, {0} '
'biomolecule(s) are returned.'.format(len(ag)))
else:
LOGGER.info('Biomolecular transformations were applied to the '
'coordinate data.')
if model != 0:
if header:
return ag, hd
else:
return ag
else:
return hd
def parsePDBStream(stream, **kwargs):
"""Returns an :class:`.AtomGroup` and/or dictionary containing header data
parsed from a stream of PDB lines.
:arg stream: Anything that implements the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
model = kwargs.get('model')
header = kwargs.get('header', False)
assert isinstance(header, bool), 'header must be a boolean'
chain = kwargs.get('chain')
subset = kwargs.get('subset')
altloc = kwargs.get('altloc', 'A')
if model is not None:
if isinstance(model, Integral):
if model < 0:
raise ValueError('model must be greater than 0')
else:
raise TypeError('model must be an integer, {0} is invalid'.format(
str(model)))
title_suffix = ''
if subset:
try:
subset = _PDBSubsets[subset.lower()]
except AttributeError:
raise TypeError('subset must be a string')
except KeyError:
raise ValueError('{0} is not a valid subset'.format(repr(subset)))
title_suffix = '_' + subset
if chain is not None:
if not isinstance(chain, str):
raise TypeError('chain must be a string')
elif len(chain) == 0:
raise ValueError('chain must not be an empty string')
title_suffix = chain + title_suffix
ag = None
if 'ag' in kwargs:
ag = kwargs['ag']
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
n_csets = ag.numCoordsets()
elif model != 0:
ag = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
n_csets = 0
biomol = kwargs.get('biomol', False)
auto_secondary = None
secondary = kwargs.get('secondary')
if not secondary:
auto_secondary = SETTINGS.get('auto_secondary')
secondary = auto_secondary
split = 0
hd = None
if model != 0:
LOGGER.timeit()
try:
lines = stream.readlines()
except AttributeError as err:
try:
lines = stream.read().split('\n')
except AttributeError:
raise err
if not len(lines):
raise ValueError('empty PDB file or stream')
if header or biomol or secondary:
hd, split = getHeaderDict(lines)
_parsePDBLines(ag, lines, split, model, chain, subset, altloc)
if ag.numAtoms() > 0:
LOGGER.report('{0} atoms and {1} coordinate set(s) were '
'parsed in %.2fs.'.format(
ag.numAtoms(),
ag.numCoordsets() - n_csets))
else:
ag = None
LOGGER.warn('Atomic data could not be parsed, please '
'check the input file.')
elif header:
hd, split = getHeaderDict(stream)
if ag is not None and isinstance(hd, dict):
if secondary:
if auto_secondary:
try:
ag = assignSecstr(hd, ag)
except ValueError:
pass
else:
ag = assignSecstr(hd, ag)
if biomol:
ag = buildBiomolecules(hd, ag)
if isinstance(ag, list):
LOGGER.info('Biomolecular transformations were applied, {0} '
'biomolecule(s) are returned.'.format(len(ag)))
else:
LOGGER.info('Biomolecular transformations were applied to the '
'coordinate data.')
if model != 0:
if header:
return ag, hd
else:
return ag
else:
return hd
def parseMMCIFStream(stream, **kwargs):
"""Returns an :class:`.AtomGroup` and/or a class:`.StarDict`
containing header data parsed from a stream of CIF lines.
:arg stream: Anything that implements the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
model = kwargs.get('model')
subset = kwargs.get('subset')
chain = kwargs.get('chain')
altloc = kwargs.get('altloc', 'A')
header = kwargs.get('header', False)
if model is not None:
if isinstance(model, int):
if model < 0:
raise ValueError('model must be greater than 0')
else:
raise TypeError('model must be an integer, {0} is invalid'.format(
str(model)))
title_suffix = ''
if subset:
try:
subset = _PDBSubsets[subset.lower()]
except AttributeError:
raise TypeError('subset must be a string')
except KeyError:
raise ValueError('{0} is not a valid subset'.format(repr(subset)))
title_suffix = '_' + subset
if chain is not None:
if not isinstance(chain, str):
raise TypeError('chain must be a string')
elif len(chain) == 0:
raise ValueError('chain must not be an empty string')
title_suffix = '_' + chain + title_suffix
ag = None
if 'ag' in kwargs:
ag = kwargs['ag']
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
n_csets = ag.numCoordsets()
elif model != 0:
ag = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
n_csets = 0
if model != 0:
LOGGER.timeit()
try:
lines = stream.readlines()
except AttributeError as err:
try:
lines = stream.read().split('\n')
except AttributeError:
raise err
if not len(lines):
raise ValueError('empty PDB file or stream')
if header:
ag, header = _parseMMCIFLines(ag, lines, model, chain, subset,
altloc, header)
else:
ag = _parseMMCIFLines(ag, lines, model, chain, subset, altloc,
header)
if ag.numAtoms() > 0:
LOGGER.report('{0} atoms and {1} coordinate set(s) were '
'parsed in %.2fs.'.format(
ag.numAtoms(),
ag.numCoordsets() - n_csets))
else:
ag = None
LOGGER.warn('Atomic data could not be parsed, please '
'check the input file.')
if header:
return ag, StarDict(*header,
title=str(kwargs.get('title', 'Unknown')))
return ag
def parsePSF(filename, title=None, ag=None):
"""Return an :class:`.AtomGroup` instance storing data parsed from X-PLOR
format PSF file *filename*. Atom and bond information is parsed from the
file. If *title* is not given, *filename* will be set as the title of the
:class:`.AtomGroup` instance. An :class:`.AtomGroup` instance may be
provided as *ag* argument. When provided, *ag* must have the same number
of atoms in the same order as the file. Data from PSF file will be added
to the *ag*. This may overwrite present data if it overlaps with PSF file
content. Note that this function does not evaluate angles, dihedrals, and
impropers sections."""
if ag is not None:
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
psf = openFile(filename, 'rb')
line = psf.readline()
i_line = 1
while line:
line = line.strip()
if line.endswith('!NATOM'):
n_atoms = int(line.split('!')[0])
break
line = psf.readline()
i_line += 1
if title is None:
title = os.path.splitext(os.path.split(filename)[1])[0]
else:
title = str(title)
if ag is None:
ag = AtomGroup(title)
else:
if n_atoms != ag.numAtoms():
raise ValueError('ag and PSF file must have same number of atoms')
serials = zeros(n_atoms, ATOMIC_FIELDS['serial'].dtype)
segnames = zeros(n_atoms, ATOMIC_FIELDS['segment'].dtype)
resnums = zeros(n_atoms, ATOMIC_FIELDS['resnum'].dtype)
resnames = zeros(n_atoms, ATOMIC_FIELDS['resname'].dtype)
atomnames = zeros(n_atoms, ATOMIC_FIELDS['name'].dtype)
atomtypes = zeros(n_atoms, ATOMIC_FIELDS['type'].dtype)
charges = zeros(n_atoms, ATOMIC_FIELDS['charge'].dtype)
masses = zeros(n_atoms, ATOMIC_FIELDS['mass'].dtype)
lines = psf.readlines(71 * (n_atoms + 5))
if len(lines) < n_atoms:
raise IOError('number of lines in PSF is less than the number of '
'atoms')
for i, line in enumerate(lines):
if i == n_atoms:
break
i_line += 1
if len(line) <= 71:
serials[i] = line[:8]
segnames[i] = line[9:13].strip()
resnums[i] = line[14:19]
resnames[i] = line[19:23].strip()
atomnames[i] = line[24:28].strip()
atomtypes[i] = line[29:35].strip()
charges[i] = line[35:44]
masses[i] = line[50:60]
else:
items = line.split()
serials[i] = items[0]
segnames[i] = items[1]
resnums[i] = items[2]
resnames[i] = items[3]
atomnames[i] = items[4]
atomtypes[i] = items[5]
charges[i] = items[6]
masses[i] = items[7]
i = n_atoms
while 1:
line = lines[i].split()
if len(line) >= 2 and line[1] == '!NBOND:':
n_bonds = int(line[0])
break
i += 1
lines = ''.join(lines[i + 1:]) + psf.read(n_bonds / 4 * 71)
array = fromstring(lines, count=n_bonds * 2, dtype=int, sep=' ')
if len(array) != n_bonds * 2:
raise IOError('number of bonds expected and parsed do not match')
psf.close()
ag.setSerials(serials)
ag.setSegnames(segnames)
ag.setResnums(resnums)
ag.setResnames(resnames)
ag.setNames(atomnames)
ag.setTypes(atomtypes)
ag.setCharges(charges)
ag.setMasses(masses)
array = add(array, -1, array)
ag.setBonds(array.reshape((n_bonds, 2)))
return ag
def parsePSF(filename, title=None, ag=None):
"""Returns an :class:`.AtomGroup` instance storing data parsed from X-PLOR
format PSF file *filename*. Atom and bond information is parsed from the
file. If *title* is not given, *filename* will be set as the title of the
:class:`.AtomGroup` instance. An :class:`.AtomGroup` instance may be
provided as *ag* argument. When provided, *ag* must have the same number
of atoms in the same order as the file. Data from PSF file will be added
to the *ag*. This may overwrite present data if it overlaps with PSF file
content. Note that this function does not evaluate angles, dihedrals, and
impropers sections."""
if ag is not None:
if not isinstance(ag, AtomGroup):
raise TypeError("ag must be an AtomGroup instance")
psf = openFile(filename, "rb")
line = psf.readline()
i_line = 1
while line:
line = line.strip()
if line.endswith(b"!NATOM"):
n_atoms = int(line.split(b"!")[0])
break
line = psf.readline()
i_line += 1
if title is None:
title = os.path.splitext(os.path.split(filename)[1])[0]
else:
title = str(title)
if ag is None:
ag = AtomGroup(title)
else:
if n_atoms != ag.numAtoms():
raise ValueError("ag and PSF file must have same number of atoms")
serials = zeros(n_atoms, ATOMIC_FIELDS["serial"].dtype)
segnames = zeros(n_atoms, ATOMIC_FIELDS["segment"].dtype)
resnums = zeros(n_atoms, ATOMIC_FIELDS["resnum"].dtype)
resnames = zeros(n_atoms, ATOMIC_FIELDS["resname"].dtype)
atomnames = zeros(n_atoms, ATOMIC_FIELDS["name"].dtype)
atomtypes = zeros(n_atoms, ATOMIC_FIELDS["type"].dtype)
charges = zeros(n_atoms, ATOMIC_FIELDS["charge"].dtype)
masses = zeros(n_atoms, ATOMIC_FIELDS["mass"].dtype)
# lines = psf.readlines(71 * (n_atoms + 5))
n = 0
n_bonds = 0
for i, line in enumerate(psf):
if line.strip() == b"":
continue
if b"!NBOND:" in line.upper():
items = line.split()
n_bonds = int(items[0])
break
if n + 1 > n_atoms:
continue
if len(line) <= 71:
serials[n] = line[:8]
segnames[n] = line[9:13].strip()
resnums[n] = line[14:19]
resnames[n] = line[19:23].strip()
atomnames[n] = line[24:28].strip()
atomtypes[n] = line[29:35].strip()
charges[n] = line[35:44]
masses[n] = line[50:60]
else:
items = line.split()
serials[n] = items[0]
segnames[n] = items[1]
resnums[n] = items[2]
resnames[n] = items[3]
atomnames[n] = items[4]
atomtypes[n] = items[5]
charges[n] = items[6]
masses[n] = items[7]
n += 1
if n < n_atoms:
raise IOError("number of lines in PSF is less than the number of " "atoms")
# i = n_atoms
# while 1:
# line = lines[i].split()
# if len(line) >= 2 and line[1] == '!NBOND:':
# n_bonds = int(line[0])
# break
# i += 1
# lines = ''.join(lines[i+1:]) + psf.read(n_bonds/4 * 71)
lines = []
for i, line in enumerate(psf):
if line.strip() == b"":
continue
if b"!" in line:
break
lines.append(line.decode(encoding="UTF-8"))
lines = "".join(lines)
array = fromstring(lines, count=n_bonds * 2, dtype=int, sep=" ")
if len(array) != n_bonds * 2:
raise IOError("number of bonds expected and parsed do not match")
psf.close()
ag.setSerials(serials)
ag.setSegnames(segnames)
ag.setResnums(resnums)
ag.setResnames(resnames)
ag.setNames(atomnames)
ag.setTypes(atomtypes)
ag.setCharges(charges)
ag.setMasses(masses)
array = add(array, -1, array)
ag.setBonds(array.reshape((n_bonds, 2)))
return ag
def parseCIFStream(stream, **kwargs):
"""Returns an :class:`.AtomGroup` and/or dictionary containing header data
parsed from a stream of CIF lines.
:arg stream: Anything that implements the method ``readlines``
(e.g. :class:`file`, buffer, stdin)"""
model = kwargs.get('model')
subset = kwargs.get('subset')
chain = kwargs.get('chain')
altloc = kwargs.get('altloc', 'A')
if model is not None:
if isinstance(model, int):
if model < 0:
raise ValueError('model must be greater than 0')
else:
raise TypeError('model must be an integer, {0} is invalid'
.format(str(model)))
title_suffix = ''
if subset:
try:
subset = _PDBSubsets[subset.lower()]
except AttributeError:
raise TypeError('subset must be a string')
except KeyError:
raise ValueError('{0} is not a valid subset'
.format(repr(subset)))
title_suffix = '_' + subset
if chain is not None:
if not isinstance(chain, str):
raise TypeError('chain must be a string')
elif len(chain) == 0:
raise ValueError('chain must not be an empty string')
title_suffix = '_' + chain + title_suffix
ag = None
if 'ag' in kwargs:
ag = kwargs['ag']
if not isinstance(ag, AtomGroup):
raise TypeError('ag must be an AtomGroup instance')
n_csets = ag.numCoordsets()
elif model != 0:
ag = AtomGroup(str(kwargs.get('title', 'Unknown')) + title_suffix)
n_csets = 0
if model != 0:
LOGGER.timeit()
try:
lines = stream.readlines()
except AttributeError as err:
try:
lines = stream.read().split('\n')
except AttributeError:
raise err
if not len(lines):
raise ValueError('empty PDB file or stream')
ag = _parseCIFLines(ag, lines, model, chain, subset, altloc)
if ag.numAtoms() > 0:
LOGGER.report('{0} atoms and {1} coordinate set(s) were '
'parsed in %.2fs.'.format(ag.numAtoms(),
ag.numCoordsets() - n_csets))
else:
ag = None
LOGGER.warn('Atomic data could not be parsed, please '
'check the input file.')
return ag
