def create(self, fileName, sizes, srcDataset=None, title=""):
Dataset.createDataset(fileName, "", sizes, True)
dataset = self.open(fileName, True)
if srcDataset:
nDim = len(sizes)
nDimSrc = srcDataset.getNDim()
for iDim in range(min(nDim, nDimSrc)):
srcDataset.copyHeader(dataset, iDim)
dataset.writeHeader()
return dataset
def genRNASecStrPeaks(self,
dataset,
listName="",
condition="sim",
scheme=""):
self.setWidths([self.widthH, self.widthH])
peakList = self.getPeakList(dataset, listName)
if scheme == "" and dataset != None:
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
self.setLabelScheme(dataset.getProperty("labelScheme"))
scheme = dataset.getProperty("editScheme")
if scheme == "":
scheme = "aa"
#print(self.vienna)
ss = SSGen(self.mol, self.vienna)
ss.genRNAResidues()
ss.pairTo()
ss.secondaryStructGen()
peakList.setSampleConditionLabel(condition)
self.addRNASecStrPeaks(peakList, scheme)
return peakList
def genTOCSYPeaks(self,
dataset,
listName="",
condition="sim",
transfers=2):
self.setWidths([self.widthH, self.widthH])
if dataset != None and dataset != "":
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
labelScheme = dataset.getProperty("labelScheme")
self.setLabelScheme(labelScheme)
peakList = self.getPeakList(dataset, listName)
peakList.setSampleConditionLabel(condition)
entities = self.getResiduesAndCompounds(self.mol)
for entity in entities:
cList = CouplingList()
cList.generateCouplings(entity, 3, 2, transfers, 2)
tLinks = cList.getTocsyLinks()
for link in tLinks:
a0 = link.getAtom(0)
a1 = link.getAtom(1)
shell = link.getShell()
self.addPeak(peakList, [a0, a1])
return peakList
def createSub(self, fileName, nDim, srcDataset, title=""):
if isinstance(srcDataset, basestring):
srcDataset = nd.open(srcDataset, False)
sizes = []
nDimSrc = srcDataset.getNDim()
if nDim > nDimSrc:
raise Exception("New dataset has more dimensions than source")
for iDim in range(nDim):
sizes.append(srcDataset.getSize(iDim))
Dataset.createDataset(fileName, "", sizes, True)
dataset = self.open(fileName, True)
for iDim in range(nDim):
srcDataset.copyHeader(dataset, iDim)
dataset.writeHeader()
return dataset
def processEndSpectrum(line):
global spectrumData
global projectDir
if 'pathname' in spectrumData:
fileName = spectrumData['pathname'][0]
fileName = os.path.join(projectDir, fileName)
print 'path is', fileName
if os.path.exists(fileName):
dataset = Dataset(fileName, "", False, False)
def getListName(self, dataset, tail="_gen"):
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
dataName = dataset.getName()
index = dataName.find(".")
if index != -1:
dataName = dataName[0:index]
listName = dataName + tail
return listName
def getPeakList(self, dataset, listName):
if (dataset == None or dataset == "") and listName != "":
peakList = PeakList.get(listName)
else:
if listName == "":
listName = self.getListName(dataset)
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
peakList = peakgen.makePeakListFromDataset(listName, dataset)
return peakList
def setDims(self, dataset=None, dims=None):
if dataset != None:
if dims != None:
dataObj = Dataset.getDataset(dataset)
iDims = []
for dim in dims:
if isinstance(dim, int):
iDims.append(dim)
else:
iDim = dataObj.getDim(dim)
iDims.append(iDim)
self.cmd.setDims(dataset, iDims)
def compareFiles(dName1, dName2, threshold):
dataset1 = Dataset(dName1, 'data1.nv', False)
dataset2 = Dataset(dName2, 'data2.nv', False)
iDim = 0
sum = 0.0
for (vec1, vec2) in izip(dataset1.vectors(iDim), dataset2.vectors(iDim)):
vec1.softThreshold(threshold)
vec2.softThreshold(threshold)
vec3 = vec1 - vec2
vec3.abs()
sum += vec3.sum().getReal()
return sum
def loadSaveFile(fileName, listName, aDataset=None):
global peakListName
global dataset
global peakList
global pMap
peakList = None
dataset = aDataset
print pMap
print 'dd1', dataset
if dataset != None:
print 'dd2', dataset
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
print 'dd3', dataset
print fileName, listName
peakListName = listName
with open(fileName, 'r') as f1:
for line in f1:
line = line.strip()
processLine(line)
def genHSQCPeaks(self, pType, dataset, listName="", condition="sim"):
self.setWidths([self.widthH * 2, self.widthC])
if dataset != None and dataset != "":
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
labelScheme = dataset.getProperty("labelScheme")
self.setLabelScheme(labelScheme)
peakList = self.getPeakList(dataset, listName)
peakList.setSampleConditionLabel(condition)
atoms = self.mol.getAtoms("*.H*")
for atom in atoms:
sym = atom.getElementName()
if sym == "H":
if atom.isMethyl() and not atom.isFirstInMethyl():
continue
parent = atom.getParent()
if parent != None:
pSym = parent.getElementName()
if pType == pSym:
self.addPeak(peakList, [atom, parent])
return peakList
def genDistancePeaks(self,
dataset,
listName="",
condition="sim",
scheme="",
tol=5.0):
self.setWidths([self.widthH, self.widthH])
if dataset != None and dataset != "":
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
labelScheme = dataset.getProperty("labelScheme")
self.setLabelScheme(labelScheme)
if scheme == "":
scheme = dataset.getProperty("editScheme")
if scheme == "":
scheme = "aa"
(d1Edited, d2Edited) = editingModes[scheme]
peakList = self.getPeakList(dataset, listName)
self.mol.selectAtoms("*.H*")
protonPairs = self.mol.getDistancePairs(tol, False)
for protonPair in protonPairs:
dis = protonPair.getDistance()
if dis > 4.0:
volume = 0.2
elif dis > 3.0:
volume = 0.5
else:
volume = 1.0
intensity = volume
self.addProtonPairPeak(peakList, protonPair.getAtom1(),
protonPair.getAtom2(), intensity, d1Edited,
d2Edited)
self.addProtonPairPeak(peakList, protonPair.getAtom2(),
protonPair.getAtom1(), intensity, d1Edited,
d2Edited)
return peakList
def get(self, datasetName):
dataset = Dataset.getDataset(datasetName)
return dataset
def datasets(self):
return Dataset.datasets()
def names(self):
return Dataset.names()
def open(self, fileName, writable=False):
dataset = Dataset(fileName, "", writable, False)
return dataset
def doSim(datasetName,
bmrbFileName=None,
nSigs=20,
sigRng=None,
noise=0.00005,
dataPars=None,
delay=None,
fp=0.5,
frMul=None,
offset=0.0,
ph=0.0,
pep=False,
decayFile=None):
if dataPars == None:
dimSizes = [2048, 512]
sfH = 600.0
sf = [sfH, sfH * 0.101329118]
sw = [3000.0, 3000.0]
dLabels = ['H', 'N']
dataPars = DatasetPars(dimSizes, sf, sw, dLabels)
nDims = len(dataPars.dimSizes)
dataPars.pseudoDims = [2]
if bmrbFileName != None:
shiftSets = loadBMRBShifts(bmrbFileName)
signals = makeSignals(shiftSets)
else:
signals = makeEmptySignals(nSigs)
if sigRng == None:
sigRng = SignalRng(frMin=[6.0, 90.0],
frMax=[11.0, 125.0],
amp=0.01,
ampSd=0.002,
lw=[20, 18],
lwSd=[5, 4])
randomizeSignals(signals, sigRng)
center = findCenter(2, signals)
center = [0.0, 0.0]
fcenter = list(center)
fcenter.append(1)
#dataPars.setRef(fcenter)
convertToHz(dataPars, signals, center)
if not decayFile:
createDataset(dataPars, datasetName)
dataset = Dataset(datasetName, 'hsqc.nv', True)
for plane in range(dataPars.dimSizes[2]):
newSignals = decaySignal(signals, plane * 100, 0.003)
addDatasetSignals(dataset, dataPars, datasetName, newSignals,
plane, noise, delay, fp, frMul, offset, ph, pep)
values = [1.0] * dataPars.dimSizes[2]
else:
headValues, decayValues, planes = readDecayFile(decayFile)
if planes != dataPars.dimSizes[2]:
print "Changing size of dim 3 from", dataPars.dimSizes[
2], "to", planes
dataPars.dimSizes = dataPars.dimSizes[0:2] + [planes]
createDataset(dataPars, datasetName)
dataset = Dataset(datasetName, 'hsqc.nv', True)
for plane in range(dataPars.dimSizes[2]):
newSignals = applyDecay(signals, decayValues, plane)
addDatasetSignals(dataset, dataPars, datasetName, newSignals,
plane, noise, delay, fp, frMul, offset, ph, pep)
if headValues != []:
values = headValues
if "CPMG" in decayFile:
values.insert(0, 0.0)
else:
values = [1.0] * dataPars.dimSizes[2]
dataset.setValues(2, values)
saveRefPars(dataset, dataPars)
def createDataset(dataPars, datasetName):
dimSizeArray = jarray.array(dataPars.dimSizes, 'i')
dataset = Dataset.createDataset(datasetName, 'hsqc.nv', dimSizeArray)
dataset = Dataset(datasetName, 'hsqc.nv', True)
for dim in range(dataPars.nDims):
dataset.setSf(dim, dataPars.sf[dim])
dataset.setSw(dim, dataPars.sw[dim])
dataset.setLabel(dim, dataPars.dimLabels[dim])
dataset.setRefValue(dim, dataPars.ref[dim])
dataset.setRefPt(dim, dataPars.dimSizes[dim] / 2)
dataset.syncPars(dim)
if dataPars.nDims > 2:
dataset.setComplex(2, False)
dataset.setFreqDomain(2, False)
dataset.close
def genProteinPeaks(self, expType, dataset, listName="", condition="sim"):
expType = expType.lower()
anames = {}
anames['hsqc'] = [['H', 0, 'N', 0]]
anames['hncaco'] = [['H', 0, 'N', 0, 'C', 0],
['H', 0, 'N', 0, 'C', -1]]
anames['hnco'] = [['H', 0, 'N', 0, 'C', -1]]
anames['hncoca'] = [['H', 0, 'N', 0, 'CA', -1]]
anames['hncocacb'] = [['H', 0, 'N', 0, 'CB', -1],
['H', 0, 'N', 0, 'CA', -1]]
anames['hnca'] = [['H', 0, 'N', 0, 'CA', 0],
['H', 0, 'N', 0, 'CA', -1]]
anames['hncacb'] = [['H', 0, 'N', 0, 'CB', 0],
['H', 0, 'N', 0, 'CB', -1],
['H', 0, 'N', 0, 'CA', 0],
['H', 0, 'N', 0, 'CA', -1]]
self.refMode = False
expValues = anames[expType]
if not isinstance(dataset, Dataset):
dataset = Dataset.getDataset(dataset)
peakList = self.getPeakList(dataset, listName)
peakList.setSampleConditionLabel(condition)
nDim = peakList.getNDim()
nucNames = []
for i in range(nDim):
if dataset == None:
nucName = peakList.getSpectralDim(i).getDimName()[0]
else:
nucName = dataset.getNucleus(i).getName()
nucNames.append(nucName)
polymers = self.mol.getPolymers()
for polymer in polymers:
residues = polymer.getResidues()
for iRes, aResidue in enumerate(residues):
for expSet in expValues:
nAtoms = len(expSet) / 2
atoms = [None] * nAtoms
ok = True
for aName, dRes in zip(expSet[0::2], expSet[1::2]):
if not aName[0] in nucNames:
ok = False
break
if dRes == 0:
res = aResidue
elif dRes == -1:
res = aResidue.getPrevious()
elif dRes == 1:
res = aResidue.getNext()
else:
raise ValueError("dRes (" + str(dRes) +
" is invalid")
if res == None:
ok = False
break
atom = res.getAtom(aName)
if atom == None:
ok = False
break
index = nucNames.index(aName[0])
atoms[index] = atom
if ok:
self.addPeak(peakList, atoms)
return peakList