def initialize(self):
self.constraints = []
self.cyanaLib = CyanaLibrary(version=self.version)
self.constraintElements = 2
def initialize(self):
self.constraints = []
self.cyanaLib = CyanaLibrary(version = self.version)
self.constraintElements = 2
# There are two different types of file, the old ones have only one atom, new one has two atoms defined.
self.singleAtom = True
self.orientations = {}
def initialize(self, parent, saveFrame=None):
self.constraints = []
self.constraintElements = 4
self.cyanaLibUsed = 0
self.cyanaLib = CyanaLibrary()
self.saveFrame = saveFrame
self.parent = parent
self.version = parent.version
if self.saveFrame:
self.parseSaveFrame()
def initialize(self):
self.sequences = []
self.cyanaLib = CyanaLibrary(version=self.version)
class DyanaSequenceFile(DyanaGenericFile):
# Information on file level
def initialize(self):
self.sequences = []
self.cyanaLib = CyanaLibrary(version=self.version)
def read(self, verbose=0):
if verbose == 1:
print "Reading %s sequence file %s" % (self.format, self.name)
#
# Parse the file
#
sequenceCols = []
fin = open(self.name, 'rU')
# Read, look for first line
line = fin.readline()
while line:
cols = line.split()
if len(cols) == 0 or self.patt['hash'].search(line):
pass
else:
for col in cols:
if not self.patt['hash'].search(col):
sequenceCols.append(col)
else:
break
line = fin.readline()
fin.close()
#
# Loop through the sequence info from the file
#
# Will now remove the following (and start a new chain):
#
#
# PL: Linker from a protein amino acid residue to a linker residue
# LL: Linker residue with a virtual bond length of 1 A
# LL2: Linker residue with a virtual bond length of 2 A
# LL5: Linker residue with a virtual bond length of 5 A
# LP: Linker from a linker residue to a protein amino acid residue
#
seqCode = 0
seqColNum = 0
lineErrors = []
self.sequences.append(DyanaSequence())
chainResiduesAdded = False
while (seqColNum < len(sequenceCols)):
seqCol = sequenceCols[seqColNum]
if not self.patt['onlydigit'].search(seqCol):
isLinker = False
# Removing linker residues - not for CCPN...
if seqCol in self.linkerResidueCodes:
if chainResiduesAdded:
# Create a new chain!
chainCode = self.chainCodesString[len(self.sequences) -
1]
self.sequences.append(
DyanaSequence(chainCode=chainCode))
print " Warning: started new sequence '%s' based on linker residues." % chainCode
chainResiduesAdded = False
isLinker = True
#
# Check whether recognized...
#
if not self.cyanaLib.findResLabel(
seqCol) and not self.cyanaLib.findResLabel(seqCol[1:]):
lineErrors.append(
" Warning: %s sequence element not recognized by standard CYANA library."
% seqCol)
#
# Check if next element is a number
#
if seqColNum + 1 < len(
sequenceCols) and self.patt['onlydigit'].search(
sequenceCols[seqColNum + 1]):
seqColNum += 1
seqCode = returnInt(sequenceCols[seqColNum])
else:
seqCode += 1
#
# Set the data, if valid
#
if not isLinker:
self.sequences[-1].elements.append(
DyanaSequenceElement(seqCode, seqCol))
self.sequences[-1].elements[-1].setFormatCode(
self.sequences[-1].elements[-1].code3Letter)
chainResiduesAdded = True
else:
lineErrors.append(
" Error: %s sequence element %s doesn't fit format." %
(self.format, seqCol))
seqColNum += 1
#
# Validity check
#
fileReadOk = True
numLineErrors = len(lineErrors)
# Assume that if have a certain amount of errors, this file is not the right format...
if numLineErrors > min((seqColNum * 0.1), 5):
self.sequences[-1].elements = []
fileReadOk = False
else:
for lineError in lineErrors:
print lineError
return fileReadOk
def readFromCoordinates(self, coordinateFile, verbose=0):
if verbose == 1:
print "Extracting %s sequence from coordinate file %s" % (
self.format, coordinateFile.name)
seqCode = ""
seqInsertCode = defaultSeqInsertCode
chainCode = None
modelNums = coordinateFile.modelCoordinates.keys()
modelNums.sort()
for coordinate in coordinateFile.modelCoordinates[modelNums[0]]:
if chainCode != coordinate.refChainId:
self.sequences.append(
DyanaSequence(chainCode=coordinate.refChainId))
chainCode = coordinate.refChainId
if seqCode != coordinate.seqCode or seqInsertCode != coordinate.insertionCode:
#
# New residue/item
#
seqCode = coordinate.seqCode
seqInsertCode = coordinate.insertionCode
fullSeqCode = str(seqCode) + seqInsertCode
self.sequences[-1].elements.append(
DyanaSequenceElement(fullSeqCode, coordinate.resName))
self.sequences[-1].elements[-1].setFormatCode(
coordinate.resName)
def write(self, verbose=0):
if verbose == 1:
print "Writing %s sequence file %s" % (self.format, self.name)
if len(self.sequences) > 1:
print "Warning: multiple sequences - writing to same file."
#
# TODO TODO: have to fill in sequence gaps with linker residues!?>!
#
fout = open(self.name, 'w')
for sequence in self.sequences:
#
# Write three letter codes (one per line)
#
for residue in sequence.elements:
if residue.hasCisPeptideBond:
resText = 'c' + residue.formatCode
else:
resText = residue.formatCode
fout.write("%-4s %4d" % (resText, residue.seqCode))
fout.write(self.newline)
fout.close()
class DyanaSequenceFile(DyanaGenericFile):
# Information on file level
def initialize(self):
self.sequences = []
self.cyanaLib = CyanaLibrary(version = self.version)
def read(self,verbose = 0):
if verbose == 1:
print "Reading %s sequence file %s" % (self.format,self.name)
self.sequences.append(DyanaSequence())
sequenceCols = []
fin = open(self.name, 'rU')
# Read, look for first line
line = fin.readline()
while line:
cols = line.split()
if len(cols) == 0 or self.patt['hash'].search(line):
pass
else:
sequenceCols.extend(cols)
line = fin.readline()
fin.close()
#
# Parse info
#
seqCode = 0
seqColNum = 0
lineErrors = []
while (seqColNum < len(sequenceCols)):
seqCol = sequenceCols[seqColNum]
if not self.patt['onlydigit'].search(seqCol):
if not self.cyanaLib.findResLabel(seqCol):
lineErrors.append(" Warning: %s sequence element not recognized by standard CYANA library." % seqCol)
#
# Check if next element is a number
#
if seqColNum + 1 < len(sequenceCols) and self.patt['onlydigit'].search(sequenceCols[seqColNum + 1]):
seqColNum += 1
seqCode = returnInt(sequenceCols[seqColNum])
else:
seqCode += 1
#
# Set the data, if valid
#
self.sequences[-1].elements.append(DyanaSequenceElement(seqCode,seqCol))
self.sequences[-1].elements[-1].setFormatCode(seqCol)
else:
lineErrors.append(" Error: %s sequence element %s doesn't fit format." % (self.format,seqCol))
seqColNum += 1
#
# Validity check
#
fileReadOk = True
numLineErrors = len(lineErrors)
# Assume that if have a certain amount of errors, this file is not the right format...
if numLineErrors > min((seqColNum * 0.1),5):
self.sequences[-1].elements = []
fileReadOk = False
else:
for lineError in lineErrors:
print lineError
return fileReadOk
def readFromCoordinates(self,coordinateFile, verbose = 0):
if verbose == 1:
print "Extracting %s sequence from coordinate file %s" % (self.format,coordinateFile.name)
self.sequences.append(DyanaSequence())
seqCode = ""
seqInsertCode = defaultSeqInsertCode
modelNums = coordinateFile.modelCoordinates.keys()
modelNums.sort()
for coordinate in coordinateFile.modelCoordinates[modelNums[0]]:
if seqCode != coordinate.seqCode or seqInsertCode != coordinate.insertionCode:
#
# New residue/item
#
seqCode = coordinate.seqCode
seqInsertCode = coordinate.insertionCode
fullSeqCode = str(seqCode) + seqInsertCode
self.sequences[-1].elements.append(DyanaSequenceElement(fullSeqCode,coordinate.resName))
self.sequences[-1].elements[-1].setFormatCode(coordinate.resName)
def write(self,verbose = 0):
if verbose == 1:
print "Writing %s sequence file %s" % (self.format,self.name)
if len(self.sequences) > 1:
print "Warning: multiple sequences - writing to same file."
#
# TODO TODO: have to fill in sequence gaps with linker residues!?>!
#
fout = open(self.name,'w')
for sequence in self.sequences:
#
# Write three letter codes (one per line)
#
for residue in sequence.elements:
if residue.hasCisPeptideBond:
resText = 'c' + residue.origCode3Letter
else:
resText = residue.origCode3Letter
fout.write("%-4s %4d" % (resText,residue.seqCode))
fout.write(self.newline)
fout.close()