def testUpdateDataFile(self):
"""Test case - update data file
"""
self.lfh.write(
"\nStarting %s %s\n" %
(self.__class__.__name__, sys._getframe().f_code.co_name))
try:
# Create a initial data file --
#
myDataList = []
curContainer = DataContainer("myblock")
aCat = DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append([9, 2, 3, 4, 5, 6, 7])
aCat.append([10, 2, 3, 4, 5, 6, 7])
aCat.append([11, 2, 3, 4, 5, 6, 7])
aCat.append([12, 2, 3, 4, 5, 6, 7])
#self.lfh.write("Assigned data category state-----------------\n")
#aCat.dumpIt(fh=self.lfh)
curContainer.append(aCat)
myDataList.append(curContainer)
ofh = open("test-output-1.cif", "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
ofh.close()
#
#
# Read and update the data -
#
myDataList = []
ifh = open("test-output-1.cif", "r")
pRd = PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
#
myBlock = myDataList[0]
myBlock.printIt()
myCat = myBlock.getObj('pdbx_seqtool_mapping_ref')
myCat.printIt()
for iRow in xrange(0, myCat.getRowCount()):
myCat.setValue('some value', 'ref_mon_id', iRow)
myCat.setValue(100, 'ref_mon_num', iRow)
ofh = open("test-output-2.cif", "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myDataList)
ofh.close()
#
except:
traceback.print_exc(file=self.lfh)
self.fail()
def testUpdateDataFile(self):
"""Test case - update data file
"""
self.lfh.write("\nStarting %s %s\n" % (self.__class__.__name__,
sys._getframe().f_code.co_name))
try:
# Create a initial data file --
#
myDataList=[]
curContainer=DataContainer("myblock")
aCat=DataCategory("pdbx_seqtool_mapping_ref")
aCat.appendAttribute("ordinal")
aCat.appendAttribute("entity_id")
aCat.appendAttribute("auth_mon_id")
aCat.appendAttribute("auth_mon_num")
aCat.appendAttribute("pdb_chain_id")
aCat.appendAttribute("ref_mon_id")
aCat.appendAttribute("ref_mon_num")
aCat.append([9,2,3,4,5,6,7])
aCat.append([10,2,3,4,5,6,7])
aCat.append([11,2,3,4,5,6,7])
aCat.append([12,2,3,4,5,6,7])
#self.lfh.write("Assigned data category state-----------------\n")
#aCat.dumpIt(fh=self.lfh)
curContainer.append(aCat)
myDataList.append(curContainer)
ofh = open("test-output-1.cif", "w")
pdbxW=PdbxWriter(ofh)
pdbxW.write(myDataList)
ofh.close()
#
#
# Read and update the data -
#
myDataList=[]
ifh = open("test-output-1.cif", "r")
pRd=PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
#
myBlock=myDataList[0]
myBlock.printIt()
myCat=myBlock.getObj('pdbx_seqtool_mapping_ref')
myCat.printIt()
for iRow in xrange(0,myCat.getRowCount()):
myCat.setValue('some value', 'ref_mon_id',iRow)
myCat.setValue(100, 'ref_mon_num',iRow)
ofh = open("test-output-2.cif", "w")
pdbxW=PdbxWriter(ofh)
pdbxW.write(myDataList)
ofh.close()
#
except:
traceback.print_exc(file=self.lfh)
self.fail()
def testSimpleInitialization(self):
"""Test case - Simple initialization of a data category and data block
"""
self.lfh.write(
"\nStarting %s %s\n" %
(self.__class__.__name__, sys._getframe().f_code.co_name))
try:
#
fn = "test-simple.cif"
attributeNameList = [
'aOne', 'aTwo', 'aThree', 'aFour', 'aFive', 'aSix', 'aSeven',
'aEight', 'aNine', 'aTen'
]
rowList = [[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10],
[1, 2, 3, 4, 5, 6, 7, 8, 9, 10]]
nameCat = 'myCategory'
#
#
curContainer = DataContainer("myblock")
aCat = DataCategory(nameCat, attributeNameList, rowList)
aCat.printIt()
curContainer.append(aCat)
curContainer.printIt()
#
myContainerList = []
myContainerList.append(curContainer)
ofh = open(fn, "w")
pdbxW = PdbxWriter(ofh)
pdbxW.write(myContainerList)
ofh.close()
myContainerList = []
ifh = open(fn, "r")
pRd = PdbxReader(ifh)
pRd.read(myContainerList)
ifh.close()
for container in myContainerList:
for objName in container.getObjNameList():
name, aList, rList = container.getObj(objName).get()
self.lfh.write("Recovered data category %s\n" % name)
self.lfh.write("Attribute list %r\n" %
repr(aList))
self.lfh.write("Row list %r\n" %
repr(rList))
except:
traceback.print_exc(file=self.lfh)
self.fail()
def testReadSFDataFile(self):
"""Test case - read PDB structure factor data file and compute statistics on f/sig(f).
"""
self.lfh.write(
"\nStarting %s %s\n" %
(self.__class__.__name__, sys._getframe().f_code.co_name))
try:
#
myContainerList = []
ifh = open(self.pathSFDataFile, "r")
pRd = PdbxReader(ifh)
pRd.read(myContainerList)
c0 = myContainerList[0]
#
catObj = c0.getObj("refln")
if catObj is None:
return false
nRows = catObj.getRowCount()
#
# Get column name index.
#
itDict = {}
itNameList = catObj.getItemNameList()
for idxIt, itName in enumerate(itNameList):
itDict[str(itName).lower()] = idxIt
#
idf = itDict['_refln.f_meas_au']
idsigf = itDict['_refln.f_meas_sigma_au']
minR = 100
maxR = -1
sumR = 0
icount = 0
for row in catObj.getRowList():
try:
f = float(row[idf])
sigf = float(row[idsigf])
ratio = sigf / f
#self.lfh.write(" %f %f %f\n" % (f,sigf,ratio))
maxR = max(maxR, ratio)
minR = min(minR, ratio)
sumR += ratio
icount += 1
except:
continue
ifh.close()
self.lfh.write("f/sig(f) min %f max %f avg %f count %d\n" %
(minR, maxR, sumR / icount, icount))
except:
traceback.print_exc(file=sys.stderr)
self.fail()
def testReadSFDataFile(self):
"""Test case - read PDB structure factor data file and compute statistics on f/sig(f).
"""
self.lfh.write("\nStarting %s %s\n" % (self.__class__.__name__,
sys._getframe().f_code.co_name))
try:
#
myContainerList=[]
ifh = open(self.pathSFDataFile, "r")
pRd=PdbxReader(ifh)
pRd.read(myContainerList)
c0=myContainerList[0]
#
catObj=c0.getObj("refln")
if catObj is None:
return false
nRows=catObj.getRowCount()
#
# Get column name index.
#
itDict={}
itNameList=catObj.getItemNameList()
for idxIt,itName in enumerate(itNameList):
itDict[str(itName).lower()]=idxIt
#
idf=itDict['_refln.f_meas_au']
idsigf=itDict['_refln.f_meas_sigma_au']
minR=100
maxR=-1
sumR=0
icount=0
for row in catObj.getRowList():
try:
f=float(row[idf])
sigf=float(row[idsigf])
ratio=sigf/f
#self.lfh.write(" %f %f %f\n" % (f,sigf,ratio))
maxR=max(maxR,ratio)
minR=min(minR,ratio)
sumR+=ratio
icount+=1
except:
continue
ifh.close()
self.lfh.write("f/sig(f) min %f max %f avg %f count %d\n" % (minR, maxR, sumR/icount,icount))
except:
traceback.print_exc(file=sys.stderr)
self.fail()
def testSimpleInitialization(self):
"""Test case - Simple initialization of a data category and data block
"""
self.lfh.write("\nStarting %s %s\n" % (self.__class__.__name__,
sys._getframe().f_code.co_name))
try:
#
fn="test-simple.cif"
attributeNameList=['aOne','aTwo','aThree','aFour','aFive','aSix','aSeven','aEight','aNine','aTen']
rowList=[[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10],
[1,2,3,4,5,6,7,8,9,10]
]
nameCat='myCategory'
#
#
curContainer=DataContainer("myblock")
aCat=DataCategory(nameCat,attributeNameList,rowList)
aCat.printIt()
curContainer.append(aCat)
curContainer.printIt()
#
myContainerList=[]
myContainerList.append(curContainer)
ofh = open(fn, "w")
pdbxW=PdbxWriter(ofh)
pdbxW.write(myContainerList)
ofh.close()
myContainerList=[]
ifh = open(fn, "r")
pRd=PdbxReader(ifh)
pRd.read(myContainerList)
ifh.close()
for container in myContainerList:
for objName in container.getObjNameList():
name,aList,rList=container.getObj(objName).get()
self.lfh.write("Recovered data category %s\n" % name)
self.lfh.write("Attribute list %r\n" % repr(aList))
self.lfh.write("Row list %r\n" % repr(rList))
except:
traceback.print_exc(file=self.lfh)
self.fail()
def testReadBigDataFile(self):
"""Test case - read data file
"""
self.lfh.write("\nStarting %s %s\n" % (self.__class__.__name__,
sys._getframe().f_code.co_name))
try:
#
myDataList=[]
ifh = open(self.pathBigPdbxDataFile, "r")
pRd=PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
except:
traceback.print_exc(file=sys.stderr)
self.fail()
def testReadDataFile(self):
"""Test case - read data file
"""
self.lfh.write("\nStarting %s %s\n" % (self.__class__.__name__,
sys._getframe().f_code.co_name))
try:
#
myDataList=[]
ifh = open(self.pathPdbxDataFile, "r")
pRd=PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
except:
traceback.print_exc(file=sys.stderr)
self.fail()
def testReadWriteDataFile(self):
"""Test case - data file read write test
"""
self.lfh.write("\nStarting %s %s\n" % (self.__class__.__name__,
sys._getframe().f_code.co_name))
try:
myDataList=[]
ifh = open(self.pathPdbxDataFile, "r")
pRd=PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
ofh = open(self.pathOutputFile, "w")
pWr=PdbxWriter(ofh)
pWr.write(myDataList)
ofh.close()
except:
traceback.print_exc(file=self.lfh)
self.fail()
def testReadWriteDataFile(self):
"""Test case - data file read write test
"""
self.lfh.write(
"\nStarting %s %s\n" %
(self.__class__.__name__, sys._getframe().f_code.co_name))
try:
myDataList = []
ifh = open(self.pathPdbxDataFile, "r")
pRd = PdbxReader(ifh)
pRd.read(myDataList)
ifh.close()
ofh = open(self.pathOutputFile, "w")
pWr = PdbxWriter(ofh)
pWr.write(myDataList)
ofh.close()
except:
traceback.print_exc(file=self.lfh)
self.fail()
def __init__(self, file):
"""Load a PDBx/mmCIF file.
The atom positions and Topology can be retrieved by calling
getPositions() and getTopology().
Parameters
----------
file : string
the name of the file to load. Alternatively you can pass an open
file object.
"""
top = Topology()
## The Topology read from the PDBx/mmCIF file
self.topology = top
self._positions = []
# Load the file.
inputFile = file
if isinstance(file, str):
inputFile = open(file)
reader = PdbxReader(inputFile)
data = []
reader.read(data)
block = data[0]
# Build the topology.
atomData = block.getObj('atom_site')
atomNameCol = atomData.getAttributeIndex('auth_atom_id')
atomIdCol = atomData.getAttributeIndex('id')
resNameCol = atomData.getAttributeIndex('auth_comp_id')
resNumCol = atomData.getAttributeIndex('auth_seq_id')
resInsertionCol = atomData.getAttributeIndex('pdbx_PDB_ins_code')
chainIdCol = atomData.getAttributeIndex('auth_asym_id')
elementCol = atomData.getAttributeIndex('type_symbol')
altIdCol = atomData.getAttributeIndex('label_alt_id')
modelCol = atomData.getAttributeIndex('pdbx_PDB_model_num')
xCol = atomData.getAttributeIndex('Cartn_x')
yCol = atomData.getAttributeIndex('Cartn_y')
zCol = atomData.getAttributeIndex('Cartn_z')
lastChainId = None
lastResId = None
atomTable = {}
atomsInResidue = set()
models = []
for row in atomData.getRowList():
atomKey = ((row[resNumCol], row[chainIdCol], row[atomNameCol]))
model = ('1' if modelCol == -1 else row[modelCol])
if model not in models:
models.append(model)
self._positions.append([])
modelIndex = models.index(model)
if row[altIdCol] != '.' and atomKey in atomTable and len(self._positions[modelIndex]) > atomTable[atomKey].index:
# This row is an alternate position for an existing atom, so ignore it.
continue
if modelIndex == 0:
# This row defines a new atom.
if lastChainId != row[chainIdCol]:
# The start of a new chain.
chain = top.addChain(row[chainIdCol])
lastChainId = row[chainIdCol]
lastResId = None
if lastResId != row[resNumCol] or lastChainId != row[chainIdCol] or (lastResId == '.' and row[atomNameCol] in atomsInResidue):
# The start of a new residue.
resId = (None if resNumCol == -1 else row[resNumCol])
resIC = ('' if resInsertionCol == -1 else row[resInsertionCol])
res = top.addResidue(row[resNameCol], chain, resId, resIC)
lastResId = row[resNumCol]
atomsInResidue.clear()
element = None
try:
element = elem.get_by_symbol(row[elementCol])
except KeyError:
pass
atom = top.addAtom(row[atomNameCol], element, res, row[atomIdCol])
atomTable[atomKey] = atom
atomsInResidue.add(row[atomNameCol])
else:
# This row defines coordinates for an existing atom in one of the later models.
try:
atom = atomTable[atomKey]
except KeyError:
raise ValueError('Unknown atom %s in residue %s %s for model %s' % (row[atomNameCol], row[resNameCol], row[resNumCol], model))
if atom.index != len(self._positions[modelIndex]):
raise ValueError('Atom %s for model %s does not match the order of atoms for model %s' % (row[atomIdCol], model, models[0]))
self._positions[modelIndex].append(Vec3(float(row[xCol]), float(row[yCol]), float(row[zCol]))*0.1)
for i in range(len(self._positions)):
self._positions[i] = self._positions[i]*nanometers
## The atom positions read from the PDBx/mmCIF file. If the file contains multiple frames, these are the positions in the first frame.
self.positions = self._positions[0]
self.topology.createStandardBonds()
self._numpyPositions = None
# Record unit cell information, if present.
cell = block.getObj('cell')
if cell is not None and cell.getRowCount() > 0:
row = cell.getRow(0)
(a, b, c) = [float(row[cell.getAttributeIndex(attribute)])*0.1 for attribute in ('length_a', 'length_b', 'length_c')]
(alpha, beta, gamma) = [float(row[cell.getAttributeIndex(attribute)])*math.pi/180.0 for attribute in ('angle_alpha', 'angle_beta', 'angle_gamma')]
self.topology.setPeriodicBoxVectors(computePeriodicBoxVectors(a, b, c, alpha, beta, gamma))
# Add bonds based on struct_conn records.
connectData = block.getObj('struct_conn')
if connectData is not None:
res1Col = connectData.getAttributeIndex('ptnr1_label_seq_id')
res2Col = connectData.getAttributeIndex('ptnr2_label_seq_id')
atom1Col = connectData.getAttributeIndex('ptnr1_label_atom_id')
atom2Col = connectData.getAttributeIndex('ptnr2_label_atom_id')
asym1Col = connectData.getAttributeIndex('ptnr1_label_asym_id')
asym2Col = connectData.getAttributeIndex('ptnr2_label_asym_id')
typeCol = connectData.getAttributeIndex('conn_type_id')
connectBonds = []
for row in connectData.getRowList():
type = row[typeCol][:6]
if type in ('covale', 'disulf', 'modres'):
key1 = (row[res1Col], row[asym1Col], row[atom1Col])
key2 = (row[res2Col], row[asym2Col], row[atom2Col])
if key1 in atomTable and key2 in atomTable:
connectBonds.append((atomTable[key1], atomTable[key2]))
if len(connectBonds) > 0:
# Only add bonds that don't already exist.
existingBonds = set(top.bonds())
for bond in connectBonds:
if bond not in existingBonds and (bond[1], bond[0]) not in existingBonds:
top.addBond(bond[0], bond[1])
existingBonds.add(bond)
def __init__(self, file):
"""Load a PDBx/mmCIF file.
The atom positions and Topology can be retrieved by calling getPositions() and getTopology().
Parameters:
- file (string) the name of the file to load. Alternatively you can pass an open file object.
"""
top = Topology()
## The Topology read from the PDBx/mmCIF file
self.topology = top
self._positions = []
# Load the file.
inputFile = file
if isinstance(file, str):
inputFile = open(file)
reader = PdbxReader(inputFile)
data = []
reader.read(data)
block = data[0]
# Build the topology.
atomData = block.getObj('atom_site')
atomNameCol = atomData.getAttributeIndex('label_atom_id')
atomIdCol = atomData.getAttributeIndex('id')
resNameCol = atomData.getAttributeIndex('label_comp_id')
resIdCol = atomData.getAttributeIndex('label_seq_id')
resNumCol = atomData.getAttributeIndex('auth_seq_id')
asymIdCol = atomData.getAttributeIndex('label_asym_id')
chainIdCol = atomData.getAttributeIndex('label_entity_id')
elementCol = atomData.getAttributeIndex('type_symbol')
altIdCol = atomData.getAttributeIndex('label_alt_id')
modelCol = atomData.getAttributeIndex('pdbx_PDB_model_num')
xCol = atomData.getAttributeIndex('Cartn_x')
yCol = atomData.getAttributeIndex('Cartn_y')
zCol = atomData.getAttributeIndex('Cartn_z')
lastChainId = None
lastResId = None
lastAsymId = None
atomTable = {}
atomsInResidue = set()
models = []
for row in atomData.getRowList():
atomKey = ((row[resIdCol], row[asymIdCol], row[atomNameCol]))
model = ('1' if modelCol == -1 else row[modelCol])
if model not in models:
models.append(model)
self._positions.append([])
modelIndex = models.index(model)
if row[altIdCol] != '.' and atomKey in atomTable and len(
self._positions[modelIndex]) > atomTable[atomKey].index:
# This row is an alternate position for an existing atom, so ignore it.
continue
if modelIndex == 0:
# This row defines a new atom.
if lastChainId != row[chainIdCol]:
# The start of a new chain.
chain = top.addChain(row[asymIdCol])
lastChainId = row[chainIdCol]
lastResId = None
lastAsymId = None
if lastResId != row[resIdCol] or lastAsymId != row[
asymIdCol] or (lastResId == '.'
and row[atomNameCol] in atomsInResidue):
# The start of a new residue.
res = top.addResidue(
row[resNameCol], chain,
None if resNumCol == -1 else row[resNumCol])
lastResId = row[resIdCol]
lastAsymId = row[asymIdCol]
atomsInResidue.clear()
element = None
try:
element = elem.get_by_symbol(row[elementCol])
except KeyError:
pass
atom = top.addAtom(row[atomNameCol], element, res,
row[atomIdCol])
atomTable[atomKey] = atom
atomsInResidue.add(row[atomNameCol])
else:
# This row defines coordinates for an existing atom in one of the later models.
try:
atom = atomTable[atomKey]
except KeyError:
raise ValueError(
'Unknown atom %s in residue %s %s for model %s' %
(row[atomNameCol], row[resNameCol], row[resIdCol],
model))
if atom.index != len(self._positions[modelIndex]):
raise ValueError(
'Atom %s for model %s does not match the order of atoms for model %s'
% (row[atomIdCol], model, models[0]))
self._positions[modelIndex].append(
Vec3(float(row[xCol]), float(row[yCol]), float(row[zCol])) *
0.1)
for i in range(len(self._positions)):
self._positions[i] = self._positions[i] * nanometers
## The atom positions read from the PDBx/mmCIF file. If the file contains multiple frames, these are the positions in the first frame.
self.positions = self._positions[0]
self.topology.createStandardBonds()
self._numpyPositions = None
# Record unit cell information, if present.
cell = block.getObj('cell')
if cell is not None and cell.getRowCount() > 0:
row = cell.getRow(0)
(a, b, c) = [
float(row[cell.getAttributeIndex(attribute)]) * 0.1
for attribute in ('length_a', 'length_b', 'length_c')
]
(alpha, beta, gamma) = [
float(row[cell.getAttributeIndex(attribute)]) * math.pi / 180.0
for attribute in ('angle_alpha', 'angle_beta', 'angle_gamma')
]
self.topology.setPeriodicBoxVectors(
computePeriodicBoxVectors(a, b, c, alpha, beta, gamma))
# Add bonds based on struct_conn records.
connectData = block.getObj('struct_conn')
if connectData is not None:
res1Col = connectData.getAttributeIndex('ptnr1_label_seq_id')
res2Col = connectData.getAttributeIndex('ptnr2_label_seq_id')
atom1Col = connectData.getAttributeIndex('ptnr1_label_atom_id')
atom2Col = connectData.getAttributeIndex('ptnr2_label_atom_id')
asym1Col = connectData.getAttributeIndex('ptnr1_label_asym_id')
asym2Col = connectData.getAttributeIndex('ptnr2_label_asym_id')
typeCol = connectData.getAttributeIndex('conn_type_id')
connectBonds = []
for row in connectData.getRowList():
type = row[typeCol][:6]
if type in ('covale', 'disulf', 'modres'):
key1 = (row[res1Col], row[asym1Col], row[atom1Col])
key2 = (row[res2Col], row[asym2Col], row[atom2Col])
if key1 in atomTable and key2 in atomTable:
connectBonds.append((atomTable[key1], atomTable[key2]))
if len(connectBonds) > 0:
# Only add bonds that don't already exist.
existingBonds = set(top.bonds())
for bond in connectBonds:
if bond not in existingBonds and (
bond[1], bond[0]) not in existingBonds:
top.addBond(bond[0], bond[1])
existingBonds.add(bond)
