def changeModel(inFile, prefix, sourceModel):
print '\nget ' + os.path.basename(inFile) + '..',
model = PDBModel(inFile)
model.update()
model = model.sort()
eq = model.equals(sourceModel)
if not eq[0] and eq[1]:
raise ConvertError('source and other models are not equal: ' + str(eq))
# model.validSource()
model.setSource(sourceModel.validSource())
#model.atomsChanged = 0
for k in model.atoms:
model.atoms[k, 'changed'] = N0.all(model[k] == sourceModel[k])
model.xyzChanged = (0 != N0.sum(N0.ravel(model.xyz - sourceModel.xyz)))
model.update(updateMissing=1)
if model.xyzChanged:
doper = PDBDope(model)
if 'MS' in sourceModel.atoms.keys():
doper.addSurfaceRacer(probe=1.4)
if 'density' in sourceModel.atoms.keys():
doper.addDensity()
if 'foldX' in sourceModel.info.keys():
doper.addFoldX()
if 'delphi' in sourceModel.info.keys():
doper.addDelphi()
outFile = os.path.dirname( inFile ) + '/' + prefix +\
T.stripFilename( inFile ) + '.model'
T.dump(model, outFile)
print '-> ' + os.path.basename(outFile)
def changeModel( inFile, prefix, sourceModel ):
print '\nget ' + os.path.basename( inFile ) + '..',
model = PDBModel( inFile )
model.update()
model = model.sort()
eq = model.equals( sourceModel )
if not eq[0] and eq[1]:
raise ConvertError('source and other models are not equal: ' + str(eq))
# model.validSource()
model.setSource( sourceModel.validSource() )
#model.atomsChanged = 0
for k in model.atoms:
model.atoms[k,'changed'] = N.all( model[k] == sourceModel[k] )
model.xyzChanged = ( 0 != N.sum( N.ravel( model.xyz - sourceModel.xyz)) )
model.update( updateMissing=1 )
if model.xyzChanged:
doper = PDBDope( model )
if 'MS' in sourceModel.atoms.keys():
doper.addSurfaceRacer( probe=1.4 )
if 'density' in sourceModel.atoms.keys():
doper.addDensity()
if 'foldX' in sourceModel.info.keys():
doper.addFoldX()
if 'delphi' in sourceModel.info.keys():
doper.addDelphi()
outFile = os.path.dirname( inFile ) + '/' + prefix +\
T.stripFilename( inFile ) + '.model'
T.dump( model, outFile )
print '-> ' + os.path.basename( outFile )
def calcProfiles(self, m):
"""
Calculate needed profiles.
@param m: PDBModel to calculate data for
@type m: PDBModel
"""
if self.verbose: print "Initiating PDBDope..."
d = PDBDope(m)
if not self.profileName in m.atoms.keys():
if self.profileName in ['MS', 'AS', 'curvature', 'relAS', 'relMS']:
if self.verbose: print "Adding SurfaceRacer profile...",
d.addSurfaceRacer()
if self.profileName in ['density']:
if self.verbose: print "Adding surface density...",
d.addDensity()
if not self.profileName in m.residues.keys():
if self.profileName in ['cons_abs', 'cons_max', 'cons_ent']:
if self.verbose: print "Adding conservation data...",
d.addConservation()
if self.verbose: print 'Done.'
## convert atom profiles to average residue profile
if self.profileName in m.atoms.keys():
prof = []
aProfile = m.profile(self.profileName)
resIdx = m.resIndex().tolist()
resIdx += [m.lenAtoms()]
for i in range(len(resIdx) - 1):
prof += [
N0.average(
N0.take(aProfile, range(resIdx[i], resIdx[i + 1])))
]
else:
prof = m.profile(self.profileName)
return prof
def randomSurfaces( base_folder, label, mask ):
"""
calculate surfaces for all peptides and return the
average and SD
"""
## container for results and standard deviations
MS, AS = {}, {}
MS_sd, AS_sd = {}, {}
## loop over peptide directories
for k in MOU.aaAtoms.keys():
dir = base_folder + 'GLY-%s-GLY_pcr/pcr_00'%(k)
fLst = glob.glob( dir + '/*.pdb')
msLst = []
asLst = []
## loop over pdb files for each peptide
T.flushPrint( '\nNow collecting data in %s'%dir )
for f in fLst:
## load peptide and remove waters and hydrogens
m = PDBModel( f )
m = m.compress( m.maskProtein() * m.maskHeavy() )
T.flushPrint( '.')
## add surface data
try:
d = PDBDope( m )
d.addSurfaceRacer( probe=1.4 )
## remove tailing GLY
m = m.compress( m.res2atomMask(mask) )
## collect surface data for each peptide
msLst += [ m.profile('MS') ]
asLst += [ m.profile('AS') ]
except:
print 'Failed calculating exposure for GLY-%s-GLY'%(k)
print '\t and file %s'%f
## get result dictionary for peptide
T.flushPrint('\nCollecting data ...\n')
msDic = {}
asDic = {}
msDic_sd = {}
asDic_sd = {}
j = 0
#atoms = [ a['name'] for a in m.atoms ]
for n in m['name']:
msDic[n] = N0.average(msLst)[j]
asDic[n] = N0.average(asLst)[j]
msDic_sd[n] = MAU.SD( msLst )[j]
asDic_sd[n] = MAU.SD( asLst )[j]
j += 1
MS[ k ] = msDic
AS[ k ] = asDic
MS_sd[ k ] = msDic_sd
AS_sd[ k ] = asDic_sd
return MS, AS, MS_sd, AS_sd
def randomSurfaces( base_folder, label, mask ):
"""
calculate surfaces for all peptides and return the
average and SD
"""
## container for results and standard deviations
MS, AS = {}, {}
MS_sd, AS_sd = {}, {}
## loop over peptide directories
for k in MOU.aaAtoms.keys():
dir = base_folder + 'GLY-%s-GLY_pcr/pcr_00'%(k)
fLst = glob.glob( dir + '/*.pdb')
msLst = []
asLst = []
## loop over pdb files for each peptide
T.flushPrint( '\nNow collecting data in %s'%dir )
for f in fLst:
## load peptide and remove waters and hydrogens
m = PDBModel( f )
m = m.compress( m.maskProtein() * m.maskHeavy() )
T.flushPrint( '.')
## add surface data
try:
d = PDBDope( m )
d.addSurfaceRacer( probe=1.4 )
## remove tailing GLY
m = m.compress( m.res2atomMask(mask) )
## collect surface data for each peptide
msLst += [ m.profile('MS') ]
asLst += [ m.profile('AS') ]
except:
print 'Failed calculating exposure for GLY-%s-GLY'%(k)
print '\t and file %s'%f
## get result dictionary for peptide
T.flushPrint('\nCollecting data ...\n')
msDic = {}
asDic = {}
msDic_sd = {}
asDic_sd = {}
j = 0
#atoms = [ a['name'] for a in m.atoms ]
for n in m['name']:
msDic[n] = N.average(msLst)[j]
asDic[n] = N.average(asLst)[j]
msDic_sd[n] = MAU.SD( msLst )[j]
asDic_sd[n] = MAU.SD( asLst )[j]
j += 1
MS[ k ] = msDic
AS[ k ] = asDic
MS_sd[ k ] = msDic_sd
AS_sd[ k ] = asDic_sd
return MS, AS, MS_sd, AS_sd
def prepareSource(inFile,
outFile,
wat=1,
sort=1,
foldx=1,
surf=1,
dens=1,
cons=1,
dssp=1,
delphi=0):
"""
Strip waters, add profiles and save as doped source model.
"""
source = PDBModel(inFile)
if wat:
source.remove(lambda a: a['residue_name'] in ['HOH', 'WAT', 'TIP3'])
if sort:
source = source.sort()
doper = PDBDope(source)
if surf:
## doper.addASA()
## doper.addSurfaceMask()
doper.addSurfaceRacer(probe=1.4)
if foldx:
doper.addFoldX()
if dens:
doper.addDensity()
if dssp:
doper.addSecondaryStructure()
if delphi:
doper.addDelphi()
try:
if cons:
doper.addConservation()
except:
errWriteln('\n ERROR: Conservation profile could not be added to '\
+ str(sourceOut) + '\n' )
source.saveAs(outFile)
return source
def prepareSource( inFile, outFile, wat=1, sort=1,
surf=1, dens=1, cons=1, dssp=1, delphi=0 ):
"""
Strip waters, add profiles and save as doped source model.
"""
source = PDBModel( inFile )
if wat:
source.remove( lambda a: a['residue_name'] in ['HOH','WAT','TIP3'] )
if sort:
source = source.sort()
doper = PDBDope( source )
if surf:
## doper.addASA()
## doper.addSurfaceMask()
doper.addSurfaceRacer( probe=1.4 )
## if foldx:
## doper.addFoldX()
if dens:
doper.addDensity()
if dssp:
doper.addSecondaryStructure()
if delphi:
doper.addDelphi()
try:
if cons:
doper.addConservation( )
except:
errWriteln('\n ERROR: Conservation profile could not be added to '\
+ str(sourceOut) + '\n' )
source.saveAs( outFile )
return source