def main():
if len(sys.argv) < 3:
raise RuntimeError("need arguments: tdf_directory output.mzML")
analysis_dir = sys.argv[1]
output_fname = sys.argv[2]
if sys.version_info.major == 2:
analysis_dir = unicode(analysis_dir)
td = timsdata.TimsData(analysis_dir)
conn = td.conn
# Get total frame count:
q = conn.execute("SELECT COUNT(*) FROM Frames")
row = q.fetchone()
N = row[0]
print("Analysis has {0} frames.".format(N))
# Store output
if output_fname.lower().endswith("mzml"):
consumer = pyopenms.PlainMSDataWritingConsumer(output_fname)
# Compress output
try:
opt = consumer.getOptions()
cfg = pyopenms.NumpressConfig()
cfg.estimate_fixed_point = True
cfg.numpressErrorTolerance = -1.0 # skip check, faster
cfg.setCompression("linear");
cfg.linear_fp_mass_acc = -1; # set the desired RT accuracy in seconds
opt.setNumpressConfigurationMassTime(cfg)
cfg = pyopenms.NumpressConfig()
cfg = pyopenms.NumpressConfig()
cfg.estimate_fixed_point = True
cfg.numpressErrorTolerance = -1.0 # skip check, faster
cfg.setCompression("slof");
opt.setNumpressConfigurationIntensity(cfg)
opt.setCompression(True) # zlib compression
consumer.setOptions(opt)
except Exception:
pass
if output_fname.lower().endswith("sqmass"):
consumer = pyopenms.MSDataSqlConsumer(output_fname)
for frame_id in range(N):
store_frame(frame_id+1, td, conn, consumer, compressFrame=True)
def get_consumer(output_fname):
# Store output
if output_fname.lower().endswith("mzml"):
consumer = pyopenms.PlainMSDataWritingConsumer(output_fname)
# Compress output
try:
opt = consumer.getOptions()
diapysef.util.setCompressionOptions(opt)
consumer.setOptions(opt)
except Exception as e:
print(e)
print(
"Your version of pyOpenMS does not support any compression, your files may get rather large"
)
pass
elif output_fname.lower().endswith("sqmass"):
consumer = pyopenms.MSDataSqlConsumer(output_fname)
else:
raise Exception("Supported filenames: mzML and sqMass.")
return consumer
class FilteringConsumer:
"""
Consumer that forwards all calls the internal consumer (after filtering)
"""
def __init__(self, consumer):
self._internal_consumer = consumer
def setExperimentalSettings(self, s):
self._internal_consumer.setExperimentalSettings(s)
def setExpectedSize(self, a, b):
self._internal_consumer.setExpectedSize(a, b)
def consumeChromatogram(self, c):
if c.getNativeID().find(filter_string) != -1:
self._internal_consumer.consumeChromatogram(c)
def consumeSpectrum(self, s):
if s.getNativeID().find(filter_string) != -1:
self._internal_consumer.consumeSpectrum(s)
###################################
# Do the actual work
###################################
consumer = pyopenms.PlainMSDataWritingConsumer(outfile)
consumer = FilteringConsumer(consumer)
pyopenms.MzMLFile().transform(infile, consumer)
def writeCombinedMzML(peptides, outputFile, templateFile, mzDelta,
minIntensity):
import pyopenms, copy
print 'Writing combined mzML file'
# generator for template file spectra
print 'Opening template file'
spectra = MZMLtoSpectrum(templateFile)
# output_file = pyopenms.MzMLFile()
# output_experiment = pyopenms.MSExperiment()
# can't use usual writer - stores all data in memory before
# writing to mzML at once ---> too memory intensive
# use consumer to write spectra on the fly
consumer = pyopenms.PlainMSDataWritingConsumer(outputFile)
def replaceData(targetmz, mzs, ints, MZmin, MZmax, synthMZs, synthINTs,
minIntensity, peptide):
# # 1 remove synth data below threshold
# maparray = np.where(synthINTs > minIntensity)
# synthMZs = synthMZs[maparray]
# synthINTs = synthINTs[maparray]
# check some data is still present
if synthMZs.shape[0] < 1:
return mzs, ints
noise = np.empty(0)
maxSignal = 0
maxSN = 0
# section data into isotopes
for i in range(20):
i = float(i) / 2
isotopeMin = MZmin + i - 0.05
isotopeMax = MZmin + i + 0.3
if isotopeMin > MZmax: break
# get synthetic data chuncks
indexArray = np.where((synthMZs >= isotopeMin)
& (synthMZs < isotopeMax))
isotopemzs = synthMZs[indexArray]
isotopeints = synthINTs[indexArray]
if isotopemzs.shape[0] < 1:
#print 'no isotope peaks for peptide %s: mz: %s, rt: %s i-value: %s' %(peptide.index, peptide.newmz, peptide.newrt, i)
#print 'isotopeMin: %s, isotopeMax: %s' %(isotopeMin, isotopeMax)
peptide.write = False
continue
# get experimental data within isotope boundaries
minIsotope = np.min(isotopemzs)
maxIsotope = np.max(isotopemzs)
indexArray = np.where((mzs > minIsotope) & (mzs < maxIsotope))
isotopeExptlMzs = mzs[indexArray]
isotopeExptlInts = ints[indexArray]
# match exptl data to nearest synth mz
# and increment corresponding xynth int
for i in range(isotopeExptlMzs.shape[0]):
eMz = isotopeExptlMzs[i]
index = np.argmin(np.absolute(synthMZs - eMz))
synthINTs[index] += isotopeExptlInts[i]
overlap = np.where((mzs > minIsotope) & (mzs < maxIsotope))
# get overlapping ints
noise = ints[overlap]
signal = np.max(isotopeints)
if signal > maxSignal:
maxSignal = signal
mzs = np.delete(mzs, overlap)
ints = np.delete(ints, overlap)
mzs = np.concatenate((mzs, synthMZs))
ints = np.concatenate((ints, synthINTs))
sortmap = np.argsort(mzs)
mzs = mzs[sortmap]
ints = ints[sortmap]
return mzs, ints, maxSignal
print 'writing spectra'
for i, s in enumerate(spectra):
if i % 100 == 0:
print 'Processing spectrum: %s, RT: %.2f' % (i, s.rt / 60)
mzs = s.mzs
ints = s.ints
for p in peptides:
if s.rt > p.RTmax or s.rt < p.RTmin: continue
for rti, rt in enumerate(p.synthRTs):
if rt == s.rt: # be careful here
synthMZs = p.synthMZs
synthINTs1 = p.synthINTs[rti]
synthINTs2 = copy.deepcopy(synthINTs1)
# add light peak
mzs, ints, signal = replaceData(p.newmz, mzs, ints,
p.MZmin, p.MZmax, synthMZs,
synthINTs1, minIntensity,
p)
if signal > p.lightsignal:
p.lightsignal = signal
# add heavy peak
mzs, ints, signal = replaceData(p.newmz + mzDelta, mzs,
ints, p.MZmin + mzDelta,
p.MZmax + mzDelta,
synthMZs + mzDelta,
synthINTs2, minIntensity,
p)
if signal > p.heavysignal:
p.heavysignal = signal
new_spectrum = copy.deepcopy(s.original)
new_spectrum.set_peaks((mzs, ints))
# use consumer to write new spectrum
consumer.consumeSpectrum(new_spectrum)
#output_file.store(outputFile, output_experiment)
print 'done'
return peptides