def annotate_ms2scans(self):
# Checks each peak against list of potential fragments
t0 = timer.time()
self.ms2_annotation = {}
xcorr = XCorr()
for peptide in self.ms2_spectra.keys():
self.ms2_annotation[peptide] = {}
for z in self.ms2_spectra[peptide].keys():
self.ms2_annotation[peptide][z]={}
for activ in self.ms2_spectra[peptide][z].keys():
fragments = mspy.mod_proteo.fragment(mspy.sequence(peptide),"by")
fragments_loss = mspy.mod_proteo.fragmentlosses(fragments,["H20","NH3"])
fragment_mzs = []
fragment_loss_mzs = []
for z2 in range(1,int(z)+1):
for fragment in fragments:
fragment_mzs.append([
fragment.mz(charge=z2)[1],
fragment.format('f ') +' '+ str(z2) + '+'
])
for fragment in fragments_loss:
fragment_loss_mzs.append([
fragment.mz(charge=z2)[1],
fragment.format('f ') +' '+ str(z2) + '+'
])
self.ms2_annotation[peptide][z][activ] = {}
self.ms2_annotation[peptide][z][activ] = fragment_mzs
self.ms2_annotation[peptide][z][activ] += fragment_loss_mzs
self.ms2_annotation[peptide][z][activ].append((
mspy.mod_proteo.fragment(mspy.sequence(peptide),"M")[0].mz(charge=int(z))[1],
mspy.mod_proteo.fragment(mspy.sequence(peptide),"M")[0].format('f '+' '+str(z)+'+')
))
for scan_obj in self.ms2_spectra[peptide][z][activ]:
xcorr_score = xcorr.get_xcorr(scan_obj['peaklist'],scan_obj['scan_info']['precursorMZ'],int(z),map(lambda x: x[0],fragment_mzs),map(lambda x: x[0],fragment_loss_mzs))
scan_obj['scan_info']['Xcorr'] = xcorr_score
t1 = timer.time() - t0
print 'Calculated annotations for MS2 scans in %s ' % t1
def generate_pattern_objects(peptides, charge_min, charge_max):
# Calculates theoretical patterns
SequenceObjects = {}
PatternObjects = {}
for peptide in peptides:
SequenceObjects[peptide] = mspy.sequence(peptide)
PatternObjects[peptide] = {}
for z in range(charge_min, charge_max + 1):
PatternObjects[peptide][z] = mspy.pattern(SequenceObjects[peptide].formula(), charge=z, real=False)
return [SequenceObjects, PatternObjects]
def generate_peptide_list(self):
"""Generates list of all possible peptides and their profiles"""
t0 = timer.time()
seq_obj = mspy.sequence(self.sequence)
peptide_objects = mspy.mod_proteo.digest(seq_obj,
'Non-Specific',miscleavage=self.max_length)
self.peptide_indexed_iso_maxpeak = {}
self.peptide_indexed_iso_dist = {}
for peptide in peptide_objects:
compound = mspy.obj_compound.compound(peptide.formula())
self.peptide_indexed_iso_dist[peptide.format()] = {}
self.peptide_indexed_iso_maxpeak[peptide.format()] = {}
for z in range(self.charge_min,self.charge_max+1):
pattern = compound.pattern(charge=z,real=False)
if pattern[0][0] > self.mz_min and pattern[-1][0] < self.mz_max:
highest_intensity_peak = max(pattern, key=lambda p: p[1])
self.peptide_indexed_iso_maxpeak[peptide.format()][str(z)] = highest_intensity_peak[0]
self.peptide_indexed_iso_dist[peptide.format()][str(z)] = pattern
t1 = timer.time() - t0
print 'Produced peptide isotopic distributions in %s ' % t1
def pepFrag(seq, X, Y, plotCanvas, annotation=None, plotSpec=True):
# w = MPL_Widget(enableAutoScale = False, doublePlot = True, enableEdit = True)
ax1 = plotCanvas.ax
ax2 = plotCanvas.ax2
pep = seq
mz = X
yVals = Y
sortInd = X.argsort()
X = X[sortInd]
Y = Y[sortInd]
curSeq = mspy.sequence(pep)
ans, series = doFrag(curSeq)
fragRange = N.arange(len(series))
fragType = []
fragMZ = []
# print "Series: ", series
# for val in ans:
# print val
for i, item in enumerate(ans):
fragType.append(ans[i][0])
fragMZ.append(ans[i][4])
# print item
fragMZ = N.array(fragMZ)
fragType = N.array(fragType)
fragMZInd = fragMZ.argsort()
fragMZ = fragMZ[fragMZInd]
fragType = fragType[fragMZInd]
fragYVals = N.zeros_like(fragMZ)
fragYVals += 1 #
absTol = 3000 #ppm
ppmErrs = []
ppmErrType = []
for i, frag in enumerate(fragMZ):
foundInd = X.searchsorted(frag)
if foundInd == 0:
prevInd = 0
else:
prevInd = foundInd - 1
if foundInd >= len(X):
foundInd += -1
prevInd += -1
if foundInd < 0:
foundInd = 0
if prevInd < 0:
prevInd = 0
# print len(mz), foundInd, prevInd
foundMZ = X[foundInd]
prevMZ = X[prevInd]
foundDiff = foundMZ - frag
prevDiff = prevMZ - frag
foundDiffOk = N.abs(foundDiff) < (foundMZ * absTol * 1E-6)
prevDiffOk = N.abs(prevDiff) < (prevMZ * absTol * 1E-6)
if foundDiffOk and prevDiffOk:
if N.abs(foundDiff) < N.abs(prevDiff):
ppmErrs.append([frag, foundDiff / frag * 1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[foundInd]
else:
ppmErrs.append([frag, prevDiff / frag * 1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[prevInd]
elif foundDiffOk:
ppmErrs.append([frag, foundDiff / frag * 1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[foundInd]
elif prevDiffOk:
ppmErrs.append([frag, prevDiff / frag * 1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[prevInd]
#plot the lines for each spectrum before being matched
if plotSpec:
ax1.vlines(X, 0, Y, colors='k', alpha=1.0) #0.6)
m = 1
ppmErrType = N.array(ppmErrType)
ppmErrs = N.array(ppmErrs)
errs = ppmErrs[:, 1]
errY = N.arange(len(errs))
errY += 1
# ax2.plot(errs, errY, 'go', ms = 5, alpha = 0.6)
tempColors = ['r', 'b', 'g', 'y', 'm', 'b', 'k']
tempMarkers = ['o', 's', 'd', '^', 'h', 'p', 'o', 'd', 's']
n = 1
for m, frag in enumerate(fragRange):
fragInd = N.where(fragType == frag)[0]
tempFrag = fragMZ[fragInd]
errInd = N.where(ppmErrType == frag)[0]
tempErrs = errs[errInd]
tempErrY = N.arange(n, len(tempErrs) + n)
n += len(tempErrs)
ax2.plot(tempErrs,
tempErrY,
linestyle='None',
marker=tempMarkers[frag],
color=tempColors[frag],
ms=5,
alpha=0.6)
# print frag, tempFrag
tempInt = fragYVals[fragInd]
ax1.vlines(tempFrag,
0,
tempInt,
colors=tempColors[frag],
linestyles='solid',
alpha=0.8) #
# ax1.legend()#legend is broken in mpl 0.98.5
# ax2.axvline(ymax = errY.max(), color = 'k', ls = '--')
# ax2.set_xlim(xmin = N.abs(errs.max())*-1.1, xman = N.abs(errs.max())*1.1)
errMax = n - len(tempErrs) + 1
ax2.axvline(ymax=errMax, color='k', ls='--')
ax2.set_xlim(xmin=N.abs(errs).max() * -1.1, xman=N.abs(errs).max() * 1.1)
ax2.set_ylim(ymin=0, yman=errMax)
if annotation != None:
textTag = seq + annotation
else:
textTag = seq
ax1.text(0.03, 0.95, textTag, fontsize=10,\
bbox=dict(facecolor='yellow', alpha=0.1),\
transform=ax1.transAxes, va='top')
matchedInd = N.where(fragYVals > 1)[0]
labelPeaks(fragMZ[matchedInd], fragYVals[matchedInd], ax1, yCutoff=10)
plotCanvas.format_labels()
plotCanvas.draw()
def pepFrag(seq, X, Y, plotCanvas, annotation = None, plotSpec = True):
# w = MPL_Widget(enableAutoScale = False, doublePlot = True, enableEdit = True)
ax1 = plotCanvas.ax
ax2 = plotCanvas.ax2
pep = seq
mz = X
yVals = Y
sortInd = X.argsort()
X = X[sortInd]
Y = Y[sortInd]
curSeq = mspy.sequence(pep)
ans, series = doFrag(curSeq)
fragRange = N.arange(len(series))
fragType = []
fragMZ = []
# print "Series: ", series
# for val in ans:
# print val
for i, item in enumerate(ans):
fragType.append(ans[i][0])
fragMZ.append(ans[i][4])
# print item
fragMZ = N.array(fragMZ)
fragType = N.array(fragType)
fragMZInd = fragMZ.argsort()
fragMZ = fragMZ[fragMZInd]
fragType = fragType[fragMZInd]
fragYVals = N.zeros_like(fragMZ)
fragYVals+=1#
absTol = 3000#ppm
ppmErrs = []
ppmErrType = []
for i,frag in enumerate(fragMZ):
foundInd = X.searchsorted(frag)
if foundInd == 0:
prevInd = 0
else:
prevInd = foundInd-1
if foundInd >= len(X):
foundInd+=-1
prevInd+=-1
if foundInd < 0:
foundInd = 0
if prevInd < 0:
prevInd = 0
# print len(mz), foundInd, prevInd
foundMZ = X[foundInd]
prevMZ = X[prevInd]
foundDiff = foundMZ-frag
prevDiff = prevMZ-frag
foundDiffOk = N.abs(foundDiff) < (foundMZ*absTol*1E-6)
prevDiffOk = N.abs(prevDiff) < (prevMZ*absTol*1E-6)
if foundDiffOk and prevDiffOk:
if N.abs(foundDiff) < N.abs(prevDiff):
ppmErrs.append([frag, foundDiff/frag*1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[foundInd]
else:
ppmErrs.append([frag, prevDiff/frag*1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[prevInd]
elif foundDiffOk:
ppmErrs.append([frag, foundDiff/frag*1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[foundInd]
elif prevDiffOk:
ppmErrs.append([frag, prevDiff/frag*1E6])
ppmErrType.append(fragType[i])
fragYVals[i] = Y[prevInd]
#plot the lines for each spectrum before being matched
if plotSpec:
ax1.vlines(X, 0, Y, colors = 'k', alpha = 1.0)#0.6)
m = 1
ppmErrType = N.array(ppmErrType)
ppmErrs = N.array(ppmErrs)
errs = ppmErrs[:,1]
errY = N.arange(len(errs))
errY+=1
# ax2.plot(errs, errY, 'go', ms = 5, alpha = 0.6)
tempColors = ['r', 'b', 'g', 'y', 'm', 'b', 'k']
tempMarkers = ['o','s','d','^', 'h', 'p', 'o', 'd', 's']
n = 1
for m,frag in enumerate(fragRange):
fragInd = N.where(fragType == frag)[0]
tempFrag = fragMZ[fragInd]
errInd = N.where(ppmErrType == frag)[0]
tempErrs = errs[errInd]
tempErrY = N.arange(n,len(tempErrs)+n)
n+=len(tempErrs)
ax2.plot(tempErrs, tempErrY, linestyle = 'None', marker = tempMarkers[frag], color = tempColors[frag], ms = 5, alpha = 0.6)
# print frag, tempFrag
tempInt = fragYVals[fragInd]
ax1.vlines(tempFrag, 0, tempInt, colors = tempColors[frag], linestyles = 'solid', alpha = 0.8)#
# ax1.legend()#legend is broken in mpl 0.98.5
# ax2.axvline(ymax = errY.max(), color = 'k', ls = '--')
# ax2.set_xlim(xmin = N.abs(errs.max())*-1.1, xman = N.abs(errs.max())*1.1)
errMax = n-len(tempErrs)+1
ax2.axvline(ymax = errMax, color = 'k', ls = '--')
ax2.set_xlim(xmin = N.abs(errs).max()*-1.1, xman = N.abs(errs).max()*1.1)
ax2.set_ylim(ymin = 0, yman = errMax)
if annotation != None:
textTag = seq+annotation
else:
textTag = seq
ax1.text(0.03, 0.95, textTag, fontsize=10,\
bbox=dict(facecolor='yellow', alpha=0.1),\
transform=ax1.transAxes, va='top')
matchedInd = N.where(fragYVals>1)[0]
labelPeaks(fragMZ[matchedInd], fragYVals[matchedInd], ax1, yCutoff = 10)
plotCanvas.format_labels()
plotCanvas.draw()
