def convert(self, input: str):
"""
converts the given input and stores it at the defined postgres database location
:param input:
:return:
"""
finder = MSMSFinder()
def callback(msms: PySpectrum, file_name: str):
with db.atomic() as transaction:
if msms is None:
self.extract_record(file_name)
else:
# 3. load sample object
record = MZMLSampleRecord.get(
MZMLSampleRecord.file_name == file_name)
# 3. associated msms spectra to it
try:
# 4. commit transaction
highest = msms.highest_peaks(1)[0]
spectra = msms.convert(msms, mode="raw").spectra
precurosr = msms.selected_precursors[0] if len(
msms.selected_precursors) > 0 else {}
scannumber = msms.index
splash = Splash().splash(
Spectrum(spectra, SpectrumType.MS))
spectra = MZMLMSMSSpectraRecord.create(
sample=record,
msms=spectra,
rt=msms.scan_time[0],
splash=splash,
scan_number=scannumber,
level=msms.ms_level,
base_peak=highest[0],
base_peak_intensity=highest[1],
precursor=precurosr['mz']
if 'mz' in precurosr else 0,
precursor_intensity=precurosr['i']
if 'i' in precurosr else 0,
precursor_charge=precurosr['charge']
if 'charge' in precurosr else 0,
ion_count=len(msms.peaks("centroided")))
except IndexError as e:
# not able to find highest peak
pass
finder.locate(msmsSource=input,
callback=callback,
filters=[MSMinLevelFilter(2)])
def test_encode_msms(source):
finder = MSMSFinder()
encoder = DualEncoder(intensity_max=1000, min_mz=0, max_mz=2000, directory="data/encoded")
data = []
def callback(msms: Spectrum, file_name: str):
nonlocal data
if msms is not None:
data.append(msms.convert(msms))
finder.locate(msmsSource=source, callback=callback, filters=[MSMinLevelFilter(2)])
from joblib import Parallel, delayed
Parallel(n_jobs=multiprocessing.cpu_count())(delayed(encoder.encode)(x) for x in data)
def test_locate_with_msms_filter(source):
"""
:return:
"""
finder = MSMSFinder()
count = 0
def callback(msms: Spectrum, file_name: str):
nonlocal count
count = count + 1
assert msms.ms_level > 1
finder.locate(msmsSource=source,
callback=callback,
filters=[MSMinLevelFilter(2)])
assert count > 0
def test_convert(source):
"""
:return:
"""
finder = MSMSFinder()
count = 0
def callback(msms: Spectrum, file_name: str):
nonlocal count
count = count + 1
converted = msms.convert(msms)
finder.locate(msmsSource=source,
callback=callback,
filters=[MSMinLevelFilter(2)])
assert count > 0
def convert(self, input: str, output: str):
"""
converts the given file to a csv file containing all msms information
:param name:
:return:
"""
finder = MSMSFinder()
with open(output, "w+") as out:
out.write("Level;Basepeak;Basepeak Intensity;Time;Splash;MSMS\n")
def callback(msms: PySpectrum, file_name: str):
if msms is not None:
highest = msms.highest_peaks(1)[0]
spectra = msms.convert(msms).spectra
splash = Splash().splash(Spectrum(spectra,
SpectrumType.MS))
out.write("{};{};{};{};{};{}\n".format(
msms.ms_level, highest[0], highest[1],
msms.scan_time[0], splash, spectra))
finder.locate(msmsSource=input,
callback=callback,
filters=[MSMinLevelFilter(2)])
min_mz=args.min_mz,
max_mz=args.max_mz,
directory="{}".format(args.destination))
data = []
def callback(msms: Spectrum, file_name: str):
"""
builds our data list
:param msms:
:param file_name:
:return:
"""
data.append(msms.convert(msms))
try:
finder.locate(msmsSource=file,
callback=callback,
filters=[MSMinLevelFilter(2)])
if len(data) > 0:
from joblib import Parallel, delayed
Parallel(n_jobs=multiprocessing.cpu_count())(
delayed(encoder.encode)(spec=x, store_string=args.txt)
for x in data)
except ParseError:
print("ignoring file: {}, due to format errors!".format(file))
print("processing was finished for {} files".format(counter))