def MZMLtoSpectrum(filename, rtOnly=False):
spectra = []
import pyopenms
# need to use OnDiscMSExperiment to sequentially read
# spectra without loading whole file into memory
experiment = pyopenms.OnDiscMSExperiment()
pyopenms.IndexedMzMLFileLoader().load(filename, experiment)
for i in range(experiment.getNrSpectra()):
spectrum = experiment.getSpectrum(i)
try:
time = spectrum.getRT()
except KeyError, e:
time_prev = time
if delta_time > 0:
time += delta_time
else:
time += 1.0
if rtOnly:
mzData, intData, spectrum = None, None, None
else:
(mzData, intData) = spectrum.get_peaks()
if mzData.shape[0] == 0:
mzData = np.empty(0, dtype='float32')
intData = np.empty(0, dtype='float32')
yield Spectrum(time, mzData, intData, spectrum)
def reader(path: str, on_disc: bool = True):
"""
Load `path` file into an OnDiskExperiment. If the file is not indexed, load
the file.
Parameters
----------
path : str
path to read mzML file from.
on_disc : bool
if True doesn't load the whole file on memory.
Returns
-------
pyopenms.OnDiskMSExperiment or pyopenms.MSExperiment
"""
if on_disc:
try:
exp_reader = pyopenms.OnDiscMSExperiment()
exp_reader.openFile(path)
except RuntimeError:
msg = "{} is not an indexed mzML file, switching to MSExperiment"
print(msg.format(path))
exp_reader = pyopenms.MSExperiment()
pyopenms.MzMLFile().load(path, exp_reader)
else:
exp_reader = pyopenms.MSExperiment()
pyopenms.MzMLFile().load(path, exp_reader)
return exp_reader
def test_readfile_content(self):
e = pyopenms.OnDiscMSExperiment()
pyopenms.IndexedMzMLFileLoader().load(self.filename, e)
self.assertEqual(e.getNrSpectra(), 2)
self.assertEqual(e.getNrChromatograms(), 1)
s = e.getSpectrum(0)
data_mz, data_int = s.get_peaks()
self.assertEqual(len(data_mz), 19914)
self.assertEqual(len(data_int), 19914)
self.assertEqual(len(e.getSpectrum(1).get_peaks()[0]), 19800)
self.assertEqual(len(e.getSpectrum(1).get_peaks()[1]), 19800)
self.assertEqual(len(e.getChromatogram(0).get_peaks()[0]), 48)
self.assertEqual(len(e.getChromatogram(0).get_peaks()[1]), 48)
if False:
# Currently we don't deal with exceptions properly
raised = False
try:
e.getChromatogram(2).get_peaks()
except Exception as e:
raised = True
self.assertTrue(raised)
def test_readfile_content(self):
e = pyopenms.OnDiscMSExperiment()
pyopenms.IndexedMzMLFileLoader().load(self.filename, e)
self.assertEqual(e.getNrSpectra(), 2)
self.assertEqual(e.getNrChromatograms(), 1)
s = e.getSpectrum(0)
data = s.get_peaks()
self.assertEqual(len(data), 19914)
self.assertEqual(len(e.getSpectrum(1).get_peaks()), 19800)
self.assertEqual(len(e.getChromatogram(0).get_peaks()), 48)
raised = False
try:
e.getChromatogram(2).get_peaks()
except Exception, e:
raised = True
def test_readfile(self):
e = pyopenms.OnDiscMSExperiment()
success = pyopenms.IndexedMzMLFileLoader().load(self.filename, e)
self.assertTrue(success)
if not os.path.isfile(args.in_):
print('Error:', args.in_, 'is not a file')
exit(1)
if not os.path.isdir(args.dir):
print('Error:', args.dir, 'is not an existing directory')
exit(1)
if not args.out.endswith('.featureXML') or args.out.endswith(
'.mzML'): # TODO: implement mzML support
print('Error:', args.out, 'must be a featureXML or mzML file')
exit(1)
if args.bench:
time_out = open(args.dir + '/benchmark.txt', 'w')
start_t = time.time()
exp = ms.OnDiscMSExperiment()
print('Loading mzML input file.', end=' ', flush=True)
if not exp.openFile(args.in_):
print('Error:', args.in_, 'is not an indexed mzML file')
exit(1)
print('Done', flush=True)
if args.bench:
total_t = time.time() - start_t
time_out.write(f'mzml load: {total_t}s\n')
time_out.close()
ff = FeatureFinderIonMobility()
features = ff.run(exp, args.num_bins, args.pp_type, args.peak_radius,
args.window_radius, args.pp_mode, args.ff_type, args.dir,
args.filter, args.debug, args.bench)
def readMS(self, file_path):
# Later: process other types of file
exp = pyopenms.OnDiscMSExperiment()
exp.openFile(file_path)
return exp
