def __init__(self, universe, selectionstring, cutoff=15.0, pbc=False, sparse=None):
"""Initialize from a *universe* or pdb file.
:Arguments:
*universe*
:class:`MDAnalysis.Universe` or a PDB file name.
*selection*
:class:`MDAnalysis.core.AtomGroup.AtomGroup` or a
:meth:`MDAnalysis.Universe.select_atoms` selection string
for atoms that define the lipid head groups, e.g.
universe.atoms.PO4 or "name PO4" or "name P*"
:Keywords:
*cutoff*
head group-defining atoms within a distance of *cutoff*
Angstroms are deemed to be in the same leaflet [15.0]
*pbc*
take periodic boundary conditions into account (only works
for orthorhombic boxes) [``False``]
*sparse*
``None``: use fastest possible routine; ``True``: use slow
sparse matrix implementation (for large systems); ``False``:
use fast :func:`~MDAnalysis.analysis.distances.distance_array`
implementation [``None``].
"""
universe = MDAnalysis.as_Universe(universe)
self.universe = universe
self.selectionstring = selectionstring
if type(self.selectionstring) == MDAnalysis.core.AtomGroup.AtomGroup:
self.selection = self.selectionstring
else:
self.selection = universe.select_atoms(self.selectionstring)
self.pbc = pbc
self.sparse = sparse
self._init_graph(cutoff)
def __init__(self, universe, selectionstring, cutoff=15.0, pbc=False, sparse=None):
"""Initialize from a *universe* or pdb file.
:Arguments:
*universe*
:class:`MDAnalysis.Universe` or a PDB file name.
*selection*
:class:`MDAnalysis.core.AtomGroup.AtomGroup` or a
:meth:`MDAnalysis.Universe.select_atoms` selection string
for atoms that define the lipid head groups, e.g.
universe.atoms.PO4 or "name PO4" or "name P*"
:Keywords:
*cutoff*
head group-defining atoms within a distance of *cutoff*
Angstroms are deemed to be in the same leaflet [15.0]
*pbc*
take periodic boundary conditions into account (only works
for orthorhombic boxes) [``False``]
*sparse*
``None``: use fastest possible routine; ``True``: use slow
sparse matrix implementation (for large systems); ``False``:
use fast :func:`~MDAnalysis.analysis.distances.distance_array`
implementation [``None``].
"""
universe = MDAnalysis.as_Universe(universe)
self.universe = universe
self.selectionstring = selectionstring
if type(self.selectionstring) == MDAnalysis.core.AtomGroup.AtomGroup:
self.selection = self.selectionstring
else:
self.selection = universe.select_atoms(self.selectionstring)
self.pbc = pbc
self.sparse = sparse
self._init_graph(cutoff)
def __init__(self, pdb, delta=1.0, select='resname HOH and name O',
metadata=None, padding=1.0, sigma=None):
"""Construct the density from psf and pdb and the select.
Parameters
----------
pdb : str
PDB file or :class:`MDAnalysis.Universe`;
select : str
selection string (MDAnalysis syntax) for the species to be analyzed
delta : float
bin size for the density grid in Angstrom (same in x,y,z) [1.0]
metadata : dict
dictionary of additional data to be saved with the object
padding : float
increase histogram dimensions by padding (on top of initial box size)
sigma : float
width (in Angstrom) of the gaussians that are used to build up the
density; if ``None`` (the default) then uses B-factors from pdb
Notes
-----
For assigning X-ray waters to MD densities one might have to use a sigma
of about 0.5 A to obtain a well-defined and resolved x-ray water density
that can be easily matched to a broader density distribution.
.. versionchanged:: 1.0.0
Changed `selection` keyword to `select`
Examples
--------
The following creates the density with the B-factors from the pdb file::
DC = BfactorDensityCreator(pdb, delta=1.0, select="name HOH",
padding=2, sigma=None)
density = DC.Density()
See Also
--------
:func:`density_from_PDB` for a convenience function
"""
u = MDAnalysis.as_Universe(pdb)
group = u.select_atoms(select)
coord = group.positions
logger.info("Selected {0:d} atoms ({1!s}) out of {2:d} total.".format(coord.shape[0], select, len(u.atoms)))
smin = np.min(coord, axis=0) - padding
smax = np.max(coord, axis=0) + padding
BINS = fixedwidth_bins(delta, smin, smax)
arange = list(zip(BINS['min'], BINS['max']))
bins = BINS['Nbins']
# get edges by doing a fake run
grid, self.edges = np.histogramdd(np.zeros((1, 3)),
bins=bins, range=arange, normed=False)
self.delta = np.diag([(e[-1] - e[0]) / (len(e) - 1) for e in self.edges])
self.midpoints = [0.5 * (e[:-1] + e[1:]) for e in self.edges]
self.origin = [m[0] for m in self.midpoints]
n_frames = 1
if sigma is None:
# histogram individually, and smear out at the same time
# with the appropriate B-factor
if np.any(group.bfactors == 0.0):
wmsg = "Some B-factors are Zero (will be skipped)."
logger.warning(wmsg)
warnings.warn(wmsg, category=MissingDataWarning)
rmsf = Bfactor2RMSF(group.bfactors)
grid *= 0.0 # reset grid
self.g = self._smear_rmsf(coord, grid, self.edges, rmsf)
else:
# histogram 'delta functions'
grid, self.edges = np.histogramdd(coord, bins=bins, range=arange, normed=False)
logger.info("Histogrammed {0:6d} atoms from pdb.".format(len(group.atoms)))
# just a convolution of the density with a Gaussian
self.g = self._smear_sigma(grid, sigma)
try:
metadata['pdb'] = pdb
except TypeError:
metadata = {'pdb': pdb}
metadata['select'] = select
metadata['n_frames'] = n_frames
metadata['sigma'] = sigma
self.metadata = metadata
logger.info("Histogram completed (initial density in Angstrom**-3)")
def __init__(self, pdb, delta=1.0, atomselection='resname HOH and name O',
metadata=None, padding=1.0, sigma=None):
"""Construct the density from psf and pdb and the atomselection.
Parameters
----------
pdb : str
PDB file or :class:`MDAnalysis.Universe`;
atomselection : str
selection string (MDAnalysis syntax) for the species to be analyzed
delta : float
bin size for the density grid in Angstroem (same in x,y,z) [1.0]
metadata : dict
dictionary of additional data to be saved with the object
padding : float
increase histogram dimensions by padding (on top of initial box size)
sigma : float
width (in Angstrom) of the gaussians that are used to build up the
density; if ``None`` (the default) then uses B-factors from pdb
Notes
-----
For assigning X-ray waters to MD densities one might have to use a sigma
of about 0.5 A to obtain a well-defined and resolved x-ray water density
that can be easily matched to a broader density distribution.
Examples
--------
The following creates the density with the B-factors from the pdb file::
DC = BfactorDensityCreator(pdb, delta=1.0, atomselection="name HOH",
padding=2, sigma=None)
density = DC.Density()
See Also
--------
:func:`density_from_PDB` for a convenience function
"""
u = MDAnalysis.as_Universe(pdb)
group = u.select_atoms(atomselection)
coord = group.positions
logger.info("Selected {0:d} atoms ({1!s}) out of {2:d} total.".format(coord.shape[0], atomselection, len(u.atoms)))
smin = np.min(coord, axis=0) - padding
smax = np.max(coord, axis=0) + padding
BINS = fixedwidth_bins(delta, smin, smax)
arange = list(zip(BINS['min'], BINS['max']))
bins = BINS['Nbins']
# get edges by doing a fake run
grid, self.edges = np.histogramdd(np.zeros((1, 3)),
bins=bins, range=arange, normed=False)
self.delta = np.diag([(e[-1] - e[0]) / (len(e) - 1) for e in self.edges])
self.midpoints = [0.5 * (e[:-1] + e[1:]) for e in self.edges]
self.origin = [m[0] for m in self.midpoints]
n_frames = 1
if sigma is None:
# histogram individually, and smear out at the same time
# with the appropriate B-factor
if np.any(group.bfactors == 0.0):
wmsg = "Some B-factors are Zero (will be skipped)."
logger.warning(wmsg)
warnings.warn(wmsg, category=MissingDataWarning)
rmsf = Bfactor2RMSF(group.bfactors)
grid *= 0.0 # reset grid
self.g = self._smear_rmsf(coord, grid, self.edges, rmsf)
else:
# histogram 'delta functions'
grid, self.edges = np.histogramdd(coord, bins=bins, range=arange, normed=False)
logger.info("Histogrammed {0:6d} atoms from pdb.".format(len(group.atoms)))
# just a convolution of the density with a Gaussian
self.g = self._smear_sigma(grid, sigma)
try:
metadata['pdb'] = pdb
except TypeError:
metadata = {'pdb': pdb}
metadata['atomselection'] = atomselection
metadata['n_frames'] = n_frames
metadata['sigma'] = sigma
self.metadata = metadata
logger.info("Histogram completed (initial density in Angstrom**-3)")
def __init__(self, *args, **kwargs):
"""Create a density grid from a trajectory.
density_from_trajectory(PSF, DCD, delta=1.0, atomselection='name OH2', ...) --> density
or
density_from_trajectory(PDB, XTC, delta=1.0, atomselection='name OH2', ...) --> density
:Arguments:
psf/pdb/gro
topology file
dcd/xtc/trr/pdb
trajectory; if reading a single PDB file it is sufficient to just provide it
once as a single argument
:Keywords:
mode
'solvent', 'bulk' or 'all' ('all' does both 'solvent' and \bulk' at the
same time and thus :meth:`DensityCreator.Density`` returns a list of
densities; this saves time!) ['all']
atomselection
selection string (MDAnalysis syntax) for the species to be analyzed
["name OH2"]
delta
approximate bin size for the density grid in Angstroem (same in x,y,z)
(It is slightly adjusted when the box length is not an integer multiple
of delta.) [1.0]
metadata
dictionary of additional data to be saved with the object
padding
increase histogram dimensions by padding (on top of initial box size)
in Angstroem [2.0]
soluteselection
MDAnalysis selection for the solute, e.g. "protein" [``None``]
cutoff
With *cutoff*, select '<atomsel> NOT WITHIN <cutoff> OF <soluteselection>'
(Special routines that are faster than the standard AROUND selection) [0]
verbosity: int
level of chattiness; 0 is silent, 3 is verbose [3]
:Returns: :class:`hop.sitemap.Density`
:TODO:
* Should be able to also set skip and start/stop for data collection.
.. Note::
* In order to calculate the bulk density, use
atomselection='name OH2',soluteselection='protein and not name H*',cutoff=3.5
This will select water oxygens not within 3.5 A of the protein heavy atoms.
Alternatively, use the VMD-based :func:`density_from_volmap` function.
* The histogramming grid is determined by the initial frames min and max.
* metadata will be populated with psf, dcd, and a few other items.
This allows more compact downstream processing.
"""
_kwargs = self.defaults.copy()
_kwargs.update(kwargs)
kwargs = _kwargs
# workaround for python 2.5 *args,**kwargs only allowed:
universe_kwargs = {"permissive": kwargs.pop("permissive", False)}
self.universe = MDAnalysis.as_Universe(*args, **universe_kwargs)
self.mode = kwargs.pop("mode", "all") # 'all' runs modes[1:]
if not self.mode in self.modes:
errmsg = "DensityCreator: mode must be one of %r, not %r" % (self.modes, self.mode)
logger.fatal(errmsg)
raise ValueError(errmsg)
if self.mode == "all":
modes = self.modes[1:]
else:
modes = [self.mode]
self.collectors = []
min_coords = []
max_coords = []
for mode in modes:
modeargs = kwargs.copy()
if mode == "solvent":
modeargs["soluteselection"] = None
modeargs["cutoff"] = 0
c = DensityCollector(mode, self.universe, **modeargs)
self.collectors.append(c)
min_coords.append(c.min_coordinates()) # with default padding from modeargs
max_coords.append(c.max_coordinates())
# determine maximum bounding box from initial positions of solvent
# (add generous padding... probably more than my default 2 A)
smin = numpy.sort(min_coords, axis=0)[0] # the three smallest values
smax = numpy.sort(max_coords, axis=0)[-1] # the three largest values
for c in self.collectors:
c.init_histogram(smin=smin, smax=smax) # also guarantees compatible grid
self.densities = {} # densities will be stored with mode as key
def __init__(self, pdb, delta=1.0, atomselection='resname HOH and name O',
metadata=None, padding=1.0, sigma=None):
"""Construct the density from psf and pdb and the atomselection.
DC = BfactorDensityCreator(pdb, delta=<delta>, atomselection=<MDAnalysis selection>,
metadata=<dict>, padding=2, sigma=None)
density = DC.Density()
:Arguments:
pdb
PDB file or :class:`MDAnalysis.Universe`; a PDB is read with the
simpl PDB reader. If the Bio.PDB reader is required, either set
the *permissive_pdb_reader* flag to ``False`` in
:data:`MDAnalysis.core.flags` or supply a Universe
that was created with the `permissive` = ``False`` keyword.
atomselection
selection string (MDAnalysis syntax) for the species to be analyzed
delta
bin size for the density grid in Angstroem (same in x,y,z) [1.0]
metadata
dictionary of additional data to be saved with the object
padding
increase histogram dimensions by padding (on top of initial box size)
sigma
width (in Angstrom) of the gaussians that are used to build up the
density; if None then uses B-factors from pdb
For assigning X-ray waters to MD densities one might have to use a sigma
of about 0.5 A to obtain a well-defined and resolved x-ray water density
that can be easily matched to a broader density distribution.
"""
u = MDAnalysis.as_Universe(pdb)
group = u.select_atoms(atomselection)
coord = group.coordinates()
logger.info("Selected {0:d} atoms ({1!s}) out of {2:d} total.".format(coord.shape[0], atomselection, len(u.atoms)))
smin = np.min(coord, axis=0) - padding
smax = np.max(coord, axis=0) + padding
BINS = fixedwidth_bins(delta, smin, smax)
arange = zip(BINS['min'], BINS['max'])
bins = BINS['Nbins']
# get edges by doing a fake run
grid, self.edges = np.histogramdd(np.zeros((1, 3)),
bins=bins, range=arange, normed=False)
self.delta = np.diag(map(lambda e: (e[-1] - e[0]) / (len(e) - 1), self.edges))
self.midpoints = map(lambda e: 0.5 * (e[:-1] + e[1:]), self.edges)
self.origin = map(lambda m: m[0], self.midpoints)
n_frames = 1
if sigma is None:
# histogram individually, and smear out at the same time
# with the appropriate B-factor
if np.any(group.bfactors == 0.0):
wmsg = "Some B-factors are Zero (will be skipped)."
logger.warn(wmsg)
warnings.warn(wmsg, category=MissingDataWarning)
rmsf = Bfactor2RMSF(group.bfactors)
grid *= 0.0 # reset grid
self.g = self._smear_rmsf(coord, grid, self.edges, rmsf)
else:
# histogram 'delta functions'
grid, self.edges = np.histogramdd(coord, bins=bins, range=arange, normed=False)
logger.info("Histogrammed {0:6d} atoms from pdb.".format(len(group.atoms)))
# just a convolution of the density with a Gaussian
self.g = self._smear_sigma(grid, sigma)
try:
metadata['pdb'] = pdb
except TypeError:
metadata = {'pdb': pdb}
metadata['atomselection'] = atomselection
metadata['n_frames'] = n_frames
metadata['sigma'] = sigma
self.metadata = metadata
logger.info("Histogram completed (initial density in Angstrom**-3)")
def __init__(self, *args, **kwargs):
"""Calculate native contacts within a group or between two groups.
:Arguments:
*topology*
psf or pdb file
*trajectory*
dcd or xtc/trr file
*universe*
instead of a topology/trajectory combination, one can also supply
a :class:`MDAnalysis.Universe`
:Keywords:
*selection*
selection string that determines which distances are calculated; if
this is a tuple or list with two entries then distances are
calculated between these two different groups ["name CA or name
B*"]
*refgroup*
reference group, either a single
:class:`~MDAnalysis.core.AtomGroup.AtomGroup` (if there is only a
single *selection*) or a list of two such groups. The reference
contacts are directly computed from *refgroup* and hence the atoms
in the reference group(s) must be equivalent to the ones produced
by the *selection* on the input trajectory.
*radius*
contacts are deemed any atoms within radius [8.0 A]
*outfile*
name of the output file; with the gz or bz2 suffix, a compressed
file is written. The average <q> is written to a second, gzipped
file that has the same name with 'array' included. E.g. for the
default name "q1.dat.gz" the <q> file will be "q1.array.gz". The
format is the matrix in column-row format, i.e. selection 1
residues are the columns and selection 2 residues are rows. The
file can be read with :func:`np.loadtxt`. ["q1.dat.gz"]
The function calculates the percentage of native contacts q1
along a trajectory. "Contacts" are defined as the number of atoms
within *radius* of a given other atom. *q1* is the fraction of contacts
relative to the reference state 1 (typically the starting conformation
of the trajectory).
The timeseries is written to a file *outfile* and is also accessible as
the attribute :attr:`ContactAnalysis1.timeseries`.
.. deprecated: 0.14.0
"""
# XX or should I use as input
# sel = (group1, group2), ref = (refgroup1, refgroup2)
# and get the universe from sel?
# Currently it's a odd hybrid.
#
# Enhancements:
# - select contact pairs to write out as a timecourse
# - make this selection based on qavg
from os.path import splitext
warnings.warn(
"ContactAnalysis1 is deprecated and will be removed "
"in 1.0. Use Contacts instead.",
category=DeprecationWarning)
self.selection_strings = self._return_tuple2(
kwargs.pop('selection', "name CA or name B*"), "selection")
self.references = self._return_tuple2(kwargs.pop('refgroup', None),
"refgroup")
self.radius = kwargs.pop('radius', 8.0)
self.targetdir = kwargs.pop('targetdir', os.path.curdir)
self.output = kwargs.pop('outfile', "q1.dat.gz")
self.outarray = splitext(splitext(self.output)[0])[0] + ".array.gz"
self.force = kwargs.pop('force', False)
self.timeseries = None # final result
self.filenames = args
self.universe = MDAnalysis.as_Universe(*args, **kwargs)
self.selections = [
self.universe.select_atoms(s) for s in self.selection_strings
]
# sanity checkes
for x in self.references:
if x is None:
raise ValueError("a reference AtomGroup must be supplied")
for ref, sel, s in zip(self.references, self.selections,
self.selection_strings):
if ref.atoms.n_atoms != sel.atoms.n_atoms:
raise ValueError("selection=%r: Number of atoms differ "
"between reference (%d) and trajectory (%d)" %
(s, ref.atoms.n_atoms, sel.atoms.n_atoms))
# compute reference contacts
dref = MDAnalysis.lib.distances.distance_array(
self.references[0].positions, self.references[1].positions)
self.qref = self.qarray(dref)
self.nref = self.qref.sum()
# setup arrays for the trajectory
self.d = np.zeros_like(dref)
self.q = self.qarray(self.d)
self._qtmp = np.zeros_like(self.q) # pre-allocated array
self.qavg = np.zeros(shape=self.q.shape, dtype=np.float64)
def __init__(self, *args, **kwargs):
"""Calculate native contacts within a group or between two groups.
:Arguments:
*topology*
psf or pdb file
*trajectory*
dcd or xtc/trr file
*universe*
instead of a topology/trajectory combination, one can also supply
a :class:`MDAnalysis.Universe`
:Keywords:
*selection*
selection string that determines which distances are calculated; if this
is a tuple or list with two entries then distances are calculated between
these two different groups ["name CA or name B*"]
*refgroup*
reference group, either a single :class:`~MDAnalysis.core.AtomGroup.AtomGroup`
(if there is only a single *selection*) or a list of two such groups.
The reference contacts are directly computed from *refgroup* and hence
the atoms in the reference group(s) must be equivalent to the ones produced
by the *selection* on the input trajectory.
*radius*
contacts are deemed any atoms within radius [8.0 A]
*outfile*
name of the output file; with the gz or bz2 suffix, a compressed
file is written. The average <q> is written to a second, gzipped
file that has the same name with 'array' included. E.g. for the
default name "q1.dat.gz" the <q> file will be "q1.array.gz". The
format is the matrix in column-row format, i.e. selection 1
residues are the columns and selection 2 residues are rows. The
file can be read with :func:`np.loadtxt`. ["q1.dat.gz"]
The function calculates the percentage of native contacts q1
along a trajectory. "Contacts" are defined as the number of atoms
within *radius* of a given other atom. *q1* is the fraction of contacts
relative to the reference state 1 (typically the starting conformation
of the trajectory).
The timeseries is written to a file *outfile* and is also accessible as
the attribute :attr:`ContactAnalysis1.timeseries`.
"""
# XX or should I use as input
# sel = (group1, group2), ref = (refgroup1, refgroup2)
# and get the universe from sel?
# Currently it's a odd hybrid.
#
# Enhancements:
# - select contact pairs to write out as a timecourse
# - make this selection based on qavg
from os.path import splitext
self.selection_strings = self._return_tuple2(kwargs.pop("selection", "name CA or name B*"), "selection")
self.references = self._return_tuple2(kwargs.pop("refgroup", None), "refgroup")
self.radius = kwargs.pop("radius", 8.0)
self.targetdir = kwargs.pop("targetdir", os.path.curdir)
self.output = kwargs.pop("outfile", "q1.dat.gz")
self.outarray = splitext(splitext(self.output)[0])[0] + ".array.gz"
self.force = kwargs.pop("force", False)
self.timeseries = None # final result
self.filenames = args
self.universe = MDAnalysis.as_Universe(*args, **kwargs)
self.selections = [self.universe.select_atoms(s) for s in self.selection_strings]
# sanity checkes
for x in self.references:
if x is None:
raise ValueError("a reference AtomGroup must be supplied")
for ref, sel, s in izip(self.references, self.selections, self.selection_strings):
if ref.atoms.n_atoms != sel.atoms.n_atoms:
raise ValueError(
"selection=%r: Number of atoms differ between "
"reference (%d) and trajectory (%d)" % (s, ref.atoms.n_atoms, sel.atoms.n_atoms)
)
# compute reference contacts
dref = MDAnalysis.lib.distances.distance_array(
self.references[0].coordinates(), self.references[1].coordinates()
)
self.qref = self.qarray(dref)
self.nref = self.qref.sum()
# setup arrays for the trajectory
self.d = np.zeros_like(dref)
self.q = self.qarray(self.d)
self._qtmp = np.zeros_like(self.q) # pre-allocated array
self.qavg = np.zeros(shape=self.q.shape, dtype=np.float64)
def __init__(self, *args, **kwargs):
"""Create a density grid from a trajectory.
density_from_trajectory(PSF, DCD, delta=1.0, atomselection='name OH2', ...) --> density
or
density_from_trajectory(PDB, XTC, delta=1.0, atomselection='name OH2', ...) --> density
:Arguments:
psf/pdb/gro
topology file
dcd/xtc/trr/pdb
trajectory; if reading a single PDB file it is sufficient to just provide it
once as a single argument
:Keywords:
mode
'solvent', 'bulk' or 'all' ('all' does both 'solvent' and \bulk' at the
same time and thus :meth:`DensityCreator.Density`` returns a list of
densities; this saves time!) ['all']
atomselection
selection string (MDAnalysis syntax) for the species to be analyzed
["name OH2"]
delta
approximate bin size for the density grid in Angstroem (same in x,y,z)
(It is slightly adjusted when the box length is not an integer multiple
of delta.) [1.0]
metadata
dictionary of additional data to be saved with the object
padding
increase histogram dimensions by padding (on top of initial box size)
in Angstroem [2.0]
soluteselection
MDAnalysis selection for the solute, e.g. "protein" [``None``]
cutoff
With *cutoff*, select '<atomsel> NOT WITHIN <cutoff> OF <soluteselection>'
(Special routines that are faster than the standard AROUND selection) [0]
verbosity: int
level of chattiness; 0 is silent, 3 is verbose [3]
:Returns: :class:`hop.sitemap.Density`
:TODO:
* Should be able to also set skip and start/stop for data collection.
.. Note::
* In order to calculate the bulk density, use
atomselection='name OH2',soluteselection='protein and not name H*',cutoff=3.5
This will select water oxygens not within 3.5 A of the protein heavy atoms.
Alternatively, use the VMD-based :func:`density_from_volmap` function.
* The histogramming grid is determined by the initial frames min and max.
* metadata will be populated with psf, dcd, and a few other items.
This allows more compact downstream processing.
"""
_kwargs = self.defaults.copy()
_kwargs.update(kwargs)
kwargs = _kwargs
# workaround for python 2.5 *args,**kwargs only allowed:
universe_kwargs = {'permissive': kwargs.pop('permissive', False)}
self.universe = MDAnalysis.as_Universe(*args, **universe_kwargs)
self.mode = kwargs.pop("mode", "all") # 'all' runs modes[1:]
if not self.mode in self.modes:
errmsg = "DensityCreator: mode must be one of %r, not %r" % (
self.modes, self.mode)
logger.fatal(errmsg)
raise ValueError(errmsg)
if self.mode == "all":
modes = self.modes[1:]
else:
modes = [self.mode]
self.collectors = []
min_coords = []
max_coords = []
for mode in modes:
modeargs = kwargs.copy()
if mode == "solvent":
modeargs['soluteselection'] = None
modeargs['cutoff'] = 0
c = DensityCollector(mode, self.universe, **modeargs)
self.collectors.append(c)
min_coords.append(
c.min_coordinates()) # with default padding from modeargs
max_coords.append(c.max_coordinates())
# determine maximum bounding box from initial positions of solvent
# (add generous padding... probably more than my default 2 A)
smin = numpy.sort(min_coords, axis=0)[0] # the three smallest values
smax = numpy.sort(max_coords, axis=0)[-1] # the three largest values
for c in self.collectors:
c.init_histogram(smin=smin,
smax=smax) # also guarantees compatible grid
self.densities = {} # densities will be stored with mode as key
def __init__(self,
pdb,
delta=1.0,
atomselection='resname HOH and name O',
metadata=None,
padding=1.0,
sigma=None):
"""Construct the density from psf and pdb and the atomselection.
DC = BfactorDensityCreator(pdb, delta=<delta>, atomselection=<MDAnalysis selection>,
metadata=<dict>, padding=2, sigma=None)
density = DC.Density()
:Arguments:
pdb
PDB file or :class:`MDAnalysis.Universe`; a PDB is read with the
simpl PDB reader. If the Bio.PDB reader is required, either set
the *permissive_pdb_reader* flag to ``False`` in
:data:`MDAnalysis.core.flags` or supply a Universe
that was created with the `permissive` = ``False`` keyword.
atomselection
selection string (MDAnalysis syntax) for the species to be analyzed
delta
bin size for the density grid in Angstroem (same in x,y,z) [1.0]
metadata
dictionary of additional data to be saved with the object
padding
increase histogram dimensions by padding (on top of initial box size)
sigma
width (in Angstrom) of the gaussians that are used to build up the
density; if None then uses B-factors from pdb
For assigning X-ray waters to MD densities one might have to use a sigma
of about 0.5 A to obtain a well-defined and resolved x-ray water density
that can be easily matched to a broader density distribution.
"""
u = MDAnalysis.as_Universe(pdb)
group = u.select_atoms(atomselection)
coord = group.coordinates()
logger.info("Selected {0:d} atoms ({1!s}) out of {2:d} total.".format(
coord.shape[0], atomselection, len(u.atoms)))
smin = np.min(coord, axis=0) - padding
smax = np.max(coord, axis=0) + padding
BINS = fixedwidth_bins(delta, smin, smax)
arange = zip(BINS['min'], BINS['max'])
bins = BINS['Nbins']
# get edges by doing a fake run
grid, self.edges = np.histogramdd(np.zeros((1, 3)),
bins=bins,
range=arange,
normed=False)
self.delta = np.diag(
map(lambda e: (e[-1] - e[0]) / (len(e) - 1), self.edges))
self.midpoints = map(lambda e: 0.5 * (e[:-1] + e[1:]), self.edges)
self.origin = map(lambda m: m[0], self.midpoints)
n_frames = 1
if sigma is None:
# histogram individually, and smear out at the same time
# with the appropriate B-factor
if np.any(group.bfactors == 0.0):
wmsg = "Some B-factors are Zero (will be skipped)."
logger.warn(wmsg)
warnings.warn(wmsg, category=MissingDataWarning)
rmsf = Bfactor2RMSF(group.bfactors)
grid *= 0.0 # reset grid
self.g = self._smear_rmsf(coord, grid, self.edges, rmsf)
else:
# histogram 'delta functions'
grid, self.edges = np.histogramdd(coord,
bins=bins,
range=arange,
normed=False)
logger.info("Histogrammed {0:6d} atoms from pdb.".format(
len(group.atoms)))
# just a convolution of the density with a Gaussian
self.g = self._smear_sigma(grid, sigma)
try:
metadata['pdb'] = pdb
except TypeError:
metadata = {'pdb': pdb}
metadata['atomselection'] = atomselection
metadata['n_frames'] = n_frames
metadata['sigma'] = sigma
self.metadata = metadata
logger.info("Histogram completed (initial density in Angstrom**-3)")
