def go(self, dict):
"""
Calculate rmsd values for trajectory and plot them.
@param dict: dictionary with path to pdb files as keys
@type dict: dict
@return: dictionary mapping input:output names
@rtype: dict
"""
d = {}
print "working on :",
for pdbIn, out in dict.items():
print T.stripFilename( pdbIn )
## set file name for pickled Structure object
dir = os.path.dirname( pdbIn )
if self.params['out'] <> None:
dir = self.params['out']
out = dir + '/' + T.stripFilename( pdbIn) + '.model'
try:
m = PDBModel( pdbIn, skipRes=self.params['skipRes'] )
if self.params['amber']:
self.renameAmberRes( m )
self.renameToAmberAtoms( m )
if self.params['sort']:
m = m.sort()
m.saveAs( out )
except:
if self.params.has_key('report') and self.params['report']:
f = open( out[:-6] + '.error', 'w' )
f.write( T.lastError() )
f.close()
else:
T.errWriteln("Error" + T.lastError() )
out = ''
d[pdbIn] = out
print "Done."
return d
def go(self, dict):
"""
Calculate rmsd values for trajectory and plot them.
@param dict: dictionary with path to pdb files as keys
@type dict: dict
@return: dictionary mapping input:output names
@rtype: dict
"""
d = {}
print "working on :",
for pdbIn, out in dict.items():
print T.stripFilename(pdbIn)
## set file name for pickled Structure object
dir = os.path.dirname(pdbIn)
if self.params['out'] <> None:
dir = self.params['out']
out = dir + '/' + T.stripFilename(pdbIn) + '.model'
try:
m = PDBModel(pdbIn, skipRes=self.params['skipRes'])
if self.params['amber']:
self.renameAmberRes(m)
self.renameToAmberAtoms(m)
if self.params['sort']:
m = m.sort()
m.saveAs(out)
except:
if self.params.has_key('report') and self.params['report']:
f = open(out[:-6] + '.error', 'w')
f.write(T.lastError())
f.close()
else:
T.errWriteln("Error" + T.lastError())
out = ''
d[pdbIn] = out
print "Done."
return d
def setResValues( self, model, values, key='temperature_factor',
lastOnly=0 ):
"""
Add numeric value per residue to all atoms of all Structures
or the last Structure in a model. The values will be written to
either the B- (temperature_factor) or Q-factor 'occupancy' column
in the temporary pdb-file.
These values can then be used to display properties in PyMol
via commands like 'color_b' and 'color_q'. See also
L{setAtomValues}.
@param model: model name
@type model: str
@param values: list of numbers, len( values ) == number of residues
@type values: [float]
@param key: key for Atom.properties dictionary
(default: temperature_factor)
@type key: occupancy|temperature_factor
@param lastOnly: 0 - add to all in model OR
1 - add only to last Structure (default: 0)
@type lastOnly: 1|0
"""
if lastOnly:
self.dic[ model ][-1].addResProperty( values, key )
else:
for m in self.dic[ model ]:
try:
m.addResProperty( values, key )
except:
T.errWriteln( "Warning: error while adding properties.")
T.errWriteln( "Key: "+str( key )+" values: "+str( values ) )
T.errWriteln( T.lastError() )
def array_or_list(self, prof, asarray):
"""
Convert to array or list depending on asarray option
Beware: empty lists will be upgraded to empty Float arrays.
@param prof: profile
@type prof: list OR array
@param asarray: 1.. autodetect type, 0.. force list, 2.. force array
@type asarray: 2|1|0
@return: profile
@rtype: list OR array
@raise ProfileError:
"""
try:
## autodetect type
if asarray == 1:
if isinstance(prof, N.ndarray):
return self.__picklesave_array(prof)
if type(prof) is str: # tolerate strings as profiles
return list(prof)
if len(prof) == 0: # don't create arrays from empty lists
return list(prof)
p = self.__picklesave_array(N.array(prof))
if p.dtype.char not in ['O', 'c',
'S']: ## no char or object arrays!
return p
return list(prof)
## force list
if asarray == 0:
if isinstance(prof, N.ndarray):
return prof.tolist()
return list(prof)
## force array
if asarray == 2:
if isinstance(prof, N.ndarray):
return self.__picklesave_array(prof)
return self.__picklesave_array(N.array(prof))
except TypeError, why:
## Numeric bug: N.array(['','','']) raises TypeError
if asarray == 1 or asarray == 0:
return list(prof)
raise ProfileError, "Cannot create array from given list. %r"\
% T.lastError()
def array_or_list( self, prof, asarray ):
"""
Convert to array or list depending on asarray option
@param prof: profile
@type prof: list OR array
@param asarray: 1.. autodetect type, 0.. force list, 2.. force array
@type asarray: 2|1|0
@return: profile
@rtype: list OR array
@raise ProfileError:
"""
try:
## autodetect type
if asarray == 1:
if isinstance( prof, N.ndarray ):
return self.__picklesave_array( prof )
if type( prof ) is str: # tolerate strings as profiles
return list( prof )
p = self.__picklesave_array( N.array( prof ) )
if p.dtype.char not in ['O','c','S']: ## no char or object arrays!
return p
return list( prof )
## force list
if asarray == 0:
if isinstance( prof, N.ndarray ):
return prof.tolist()
return list( prof )
## force array
if asarray == 2:
if isinstance( prof, N.ndarray ):
return self.__picklesave_array( prof )
return self.__picklesave_array( N.array( prof ) )
except TypeError, why:
## Numeric bug: N.array(['','','']) raises TypeError
if asarray == 1 or asarray == 0:
return list( prof )
raise ProfileError, "Cannot create array from given list. %r"\
% T.lastError()
def add(self, str):
"""
Add String str and line break to file.
@param str: string to add to pml file
@type str: str
"""
try:
self.fgenerate.write(str + '\n')
except (IOError):
T.errWriteln(
"PymolInput.add(): Error adding string to pymol script file.")
T.errWriteln( T.lastError() )
def addResProperty( self, values, key='temperature_factor'):
"""
Does the same thing as L{addProperty} but on the residue level,
i.e adds extra value to each residue in Structure.
(The same value is added to all atoms of a residue.)
These values can then be used to display properties in PyMol
via commands like 'color_b' and 'color_q'.
@param values: list of numbers, len( values ) == number of residues
@type values: [float]
@param key: key for Atom.properties dictionary
('occupancy' OR 'temperature_factor')
@type key: occupancy|temperature_factor
"""
try:
if self.struct == None:
self.struct = PDBModel( self.fname )
self.temporary = 1
self.struct[ key ] = self.struct.res2atomProfile( values )
except:
print T.lastError()
def add(self, str):
"""
Add String str and line break to file.
@param str: string to add to pml file
@type str: str
"""
try:
self.fgenerate.write(str + '\n')
except (IOError):
T.errWriteln(
"PymolInput.add(): Error adding string to pymol script file.")
T.errWriteln(T.lastError())
def addResProperty(self, values, key='temperature_factor'):
"""
Does the same thing as L{addProperty} but on the residue level,
i.e adds extra value to each residue in Structure.
(The same value is added to all atoms of a residue.)
These values can then be used to display properties in PyMol
via commands like 'color_b' and 'color_q'.
@param values: list of numbers, len( values ) == number of residues
@type values: [float]
@param key: key for Atom.properties dictionary
('occupancy' OR 'temperature_factor')
@type key: occupancy|temperature_factor
"""
try:
if self.struct is None:
self.struct = PDBModel(self.fname)
self.temporary = 1
self.struct[key] = self.struct.res2atomProfile(values)
except:
print T.lastError()
def _completeResidues(self, chain):
"""
Look for missing or unknown atom names, add missing atoms,
report unknown atoms.
@param chain: Scientific.IO.PDB.PeptideChain object
@type chain: object
@return: Scientific.IO.PDB.PeptideChain object
@rtype: chain object
"""
chain.deleteHydrogens() ## delete all hydrogens
i = 0
self.log.add("Checking atoms of chain " + chain.segment_id)
for res in chain:
try:
if i < len(chain) - 1: # normal residue
alowed = self.aminoAcidDict[res.name]
else: # c-terminal residue
alowed = self._res2Terminal(self.aminoAcidDict[res.name])
name_list = []
for atom in res.atom_list: # check for unknown atom names
# store for missing atom check
name_list = name_list + [atom.name]
if not (atom.name in alowed):
self.log.add('\tunknown atom: ' + atom.name + ' : '+\
res.name+ str(res.number))
for name in alowed: # check for missing atoms
if not (name in name_list):
# add missing atom with 0 xyz
self._addMissing(res, name)
self.log.add('\tadded missing atom -> '+ name+ ' : '+\
res.name+ str(res.number))
except:
s = "\ncompleteResidues(): Error while checking atoms.\n"
s = s + "residue " + str(i) + " :" + str(res) + "\n"
s = s + T.lastError()
T.errWriteln(
"Error while completing residues, check log for details.")
self.log.add(s)
i = i + 1
return chain
def _completeResidues(self, chain):
"""
Look for missing or unknown atom names, add missing atoms,
report unknown atoms.
@param chain: Scientific.IO.PDB.PeptideChain object
@type chain: object
@return: Scientific.IO.PDB.PeptideChain object
@rtype: chain object
"""
chain.deleteHydrogens() ## delete all hydrogens
i = 0
self.log.add("Checking atoms of chain "+chain.segment_id)
for res in chain:
try:
if i < len(chain)-1: # normal residue
alowed = self.aminoAcidDict[res.name]
else: # c-terminal residue
alowed = self._res2Terminal(self.aminoAcidDict[res.name])
name_list = []
for atom in res.atom_list: # check for unknown atom names
# store for missing atom check
name_list = name_list + [atom.name]
if not (atom.name in alowed):
self.log.add('\tunknown atom: ' + atom.name + ' : '+\
res.name+ str(res.number))
for name in alowed: # check for missing atoms
if not (name in name_list):
# add missing atom with 0 xyz
self._addMissing(res, name)
self.log.add('\tadded missing atom -> '+ name+ ' : '+\
res.name+ str(res.number))
except:
s = "\ncompleteResidues(): Error while checking atoms.\n"
s = s + "residue " + str(i)+ " :"+ str(res) + "\n"
s = s + T.lastError()
T.errWriteln(
"Error while completing residues, check log for details.")
self.log.add(s)
i = i+1
return chain
def removeTER(self, fname):
"""
Remove TER record from PDB.
@param fname: name of existing file.
@type fname: string
"""
try:
#command = 'egrep -v "^TER " ' + fname + '> temp.pdb'
path = os.path.dirname(fname)
command = 'egrep -v "^TER " %s > %s/temp.pdb' % (fname, path)
commands.getstatusoutput(command)
os.rename('%s/temp.pdb' % path, fname)
except (OSError):
T.errWriteln("Error removing 'TER' statement from %s: ")
T.errWriteln(T.lastError())
def removeTER(self, fname):
"""
Remove TER record from PDB.
@param fname: name of existing file.
@type fname: string
"""
try:
#command = 'egrep -v "^TER " ' + fname + '> temp.pdb'
path = os.path.dirname( fname )
command = 'egrep -v "^TER " %s > %s/temp.pdb'%( fname, path )
commands.getstatusoutput(command)
os.rename('%s/temp.pdb'%path, fname)
except (OSError):
T.errWriteln("Error removing 'TER' statement from %s: ")
T.errWriteln( T.lastError() )
def writeChain(self, chain):
"""
Write single chain as PDB. File name will be segid + '_seg.pdb'.
@param chain: Scientific.IO.PDB.PeptideChain object
@type chain: chain object
"""
if (chain <> None):
try:
fname = self.path + '/' + chain.segment_id + "_seg.PDB"
file = self._startPDB(chain, fname) # include comments
chain.writeToFile(file) # create PDB
file.close() # make sure file is complete!
self.removeTER(file.file.file.name) #remove TER record from PDB
return 1 # write a chain
except (IOError):
T.errWriteln("Error writing chain to file %s:" % fname)
T.errWriteln( T.lastError() )
return 0 # false, no more chains to write
def setResValues(self,
model,
values,
key='temperature_factor',
lastOnly=0):
"""
Add numeric value per residue to all atoms of all Structures
or the last Structure in a model. The values will be written to
either the B- (temperature_factor) or Q-factor 'occupancy' column
in the temporary pdb-file.
These values can then be used to display properties in PyMol
via commands like 'color_b' and 'color_q'. See also
L{setAtomValues}.
@param model: model name
@type model: str
@param values: list of numbers, len( values ) == number of residues
@type values: [float]
@param key: key for Atom.properties dictionary
(default: temperature_factor)
@type key: occupancy|temperature_factor
@param lastOnly: 0 - add to all in model OR
1 - add only to last Structure (default: 0)
@type lastOnly: 1|0
"""
if lastOnly:
self.dic[model][-1].addResProperty(values, key)
else:
for m in self.dic[model]:
try:
m.addResProperty(values, key)
except:
T.errWriteln("Warning: error while adding properties.")
T.errWriteln("Key: " + str(key) + " values: " +
str(values))
T.errWriteln(T.lastError())
def update( self, model, source, skipRes=None, updateMissing=0, force=0,
headPatterns=[]):
"""
Update empty or missing fields of model from the source. The
model will be connected to the source via model.source.
Profiles that are derived from the source are labeled 'changed'=0.
The same holds for coordinates (xyzChanged=0).
However, existing profiles or coordinates or fields remain untouched.
@param model: existing model
@type model: PDBModel
@param source: source PDB file
@type source: str
@param skipRes: list residue names that should not be parsed
@type skipRes: [ str ]
@param updateMissing: ignored
@type updateMissing: 1|0
@param headPatterns: [(putIntoKey, regex)] extract given REMARKS
@type headPatterns: [(str, str)]
@raise PDBParserError - if something is wrong with the source file
"""
try:
# atoms and/or coordinates need to be updated from PDB
if force or self.needsUpdate( model ):
atoms, xyz, info = self.__collectAll( source, skipRes,
headPatterns )
# print atoms["serial_number"]
# for atom in atoms:
# print atom
keys = MU.union( atoms.keys(), self.DEFAULTS.keys() )
for k in keys:
if model.atoms.get( k, default=0, update=False ) in \
(0,None):
dflt = self.DEFAULTS.get( k, None )
model.atoms.set(k, atoms.get(k, dflt), changed=0 )
if model.xyz is None:
model.xyz = xyz
model.xyzChanged = 0
model._resIndex =None
model._chainIndex=None
model.fileName = model.fileName or source
model.pdbCode = model.pdbCode or info.get('pdb_code', None) or \
self.idFromName( model.fileName)
model.info.update( info )
except:
msg = self._PDBParseFile__xplorAtomIndicesTest( source ) or ' '
raise PDBParserError('Cannot read ' + str(source) + ' as PQR\n'\
'\ERROR: ' + T.lastError() + msg)
model.setSource( source )
def __collectAll( self, fname, skipRes=None, headPatterns=[] ):
"""
Parse ATOM/HETATM lines from PDB. Collect coordinates plus
dictionaries with the other pdb records of each atom.
REMARK, HEADER, etc. lines are ignored.
Some changes are made to the dictionary from PDBFile.readline()::
- the 'position' entry (with the coordinates) is removed
- leading and trailing spaces are removed from 'name' ..
- .. but a 'name_original' entry keeps the old name with spaces
- a 'type' entry is added. Its value is 'ATOM' or 'HETATM'
- a 'after_ter' entry is added. Its value is 1, if atom is
preceeded by a 'TER' line, otherwise 0
- empty 'element' entries are filled with the first non-number
letter from the atom 'name'
@param fname: name of pdb file
@type fname: str
@param skipRes: list with residue names that should be skipped
@type skipRes: list of str
@return: tuple of (1) dictionary of profiles
and (2) xyz array N x 3
@rtype: ( list, array )
"""
xyz = []
aProfs = {}
info = {}
in_header = True
headPatterns = headPatterns or self.RE_REMARKS
patterns = [ (key, re.compile(ex)) for key,ex in headPatterns ]
for k in bi.PDBModel.PDB_KEYS:
aProfs[k] = list()
f = open( fname ,'r' )
try:
line, i = ('',''), 0
while line[0] <> 'END' and line[0] <> 'ENDMDL':
i += 1
try:
line = f.readline().split()
except ValueError, what:
self.log.add('Warning: Error parsing line %i of %s' %
(i, T.stripFilename( fname )) )
self.log.add('\tError: '+str(what) )
continue
if not line: break
## header handling
if in_header and line[0] == 'HEADER':
info.update( self._PDBParseFile__parseHeader( line ) )
if in_header and line[0] == 'REMARK':
info.update( self._PDBParseFile__parseRemark( line, patterns ) )
## preserve position of TER records
# print line
newChain = line[0] == 'TER'
if newChain:
line = f.readline().split()
if (line[0] in ['ATOM','HETATM'] ):
if in_header: in_header = False ## switch off HEADER parsing
if len(line) == 11: # contains chain name
a = {'serial_number': int(line[1]),
'name': line[2],
'alternate': '',
'residue_name': string.strip(line[3]),
'chain_id': string.strip(line[4]),
'residue_number': int(line[5]),
'insertion_code': '',
'position': map(float,line[6:9]),
'occupancy': 1.0,
'temperature_factor': 0.0,
'segment_id': '',
'element': '',
'charge': line[9]}
else: # without chain name
a = {'serial_number': int(line[1]),
'name': line[2],
'alternate': '',
'residue_name': string.strip(line[3]),
'chain_id': '',
'residue_number': int(line[4]),
'insertion_code': '',
'position': map(float,line[5:8]),
'occupancy': 1.0,
'temperature_factor': 0.0,
'segment_id': '',
'element': '',
'charge': line[8]}
# print line
# print a
if skipRes and a['residue_name'] in skipRes:
continue
a['name_original'] = a['name']
a['name'] = a['name'].strip()
a['type'] = line[0]
if newChain:
a['after_ter'] = 1
else:
a['after_ter'] = 0
if a['element'] == '':
a['element'] = self._PDBParseFile__firstLetter( a['name'] )
# print a
xyz.append( a['position'] )
del( a['position'])
for k, v in a.items():
aProfs[k].append( v )
except:
raise PDBParserError("Error parsing file "+fname+": " \
+ T.lastError())
try:
f.close()
except:
pass
if len( xyz ) == 0:
raise PDBParserError("Error parsing file "+fname+": "+
"Couldn't find any atoms.")
return aProfs, npy.array( xyz, npy.float64 ), info