def loadPeakMap(path=None):
""" loads mzXML, mzML and mzData files
If *path* is missing, a dialog for file selection is opened
instead.
"""
# local import in order to keep namespaces clean
import os.path
import sys
from pyopenms import MSExperiment, FileHandler
from ..core.data_types import PeakMap
path = _prepare_path(path, extensions=["mzML", "mzXML", "mzData"])
if path is None:
return None
# open-ms returns empty peakmap if file not exists, so we
# check ourselves:
if not os.path.exists(path):
raise Exception("file %s does not exist" % path)
if not os.path.isfile(path):
raise Exception("path %s is not a file" % path)
experiment = MSExperiment()
fh = FileHandler()
if sys.platform == "win32":
path = path.replace("/", "\\") # needed for network shares
fh.loadExperiment(path, experiment)
return PeakMap.fromMSExperiment(experiment)
def loadPeakMap(path=None):
""" loads mzXML, mzML and mzData files
If *path* is missing, a dialog for file selection is opened
instead.
"""
# local import in order to keep namespaces clean
import ms
import os.path
import sys
from pyopenms import MSExperiment, FileHandler
from libms.DataStructures import PeakMap
if isinstance(path, unicode):
path = path.encode(sys.getfilesystemencoding())
elif path is None:
path = ms.askForSingleFile(extensions="mzML mzXML mzData".split())
if path is None:
return None
# open-ms returns empty peakmap if file not exists, so we
# check ourselves:
if not os.path.exists(path):
raise Exception("file %s does not exist" % path)
if not os.path.isfile(path):
raise Exception("path %s is not a file" % path)
experiment = MSExperiment()
fh = FileHandler()
if sys.platform == "win32":
path = path.replace("/","\\") # needed for network shares
fh.loadExperiment(path, experiment)
return PeakMap.fromMSExperiment(experiment)
def storePeakMap(pm, path=None):
""" Stores peakmap *pm* in mzXML, mzML or mzData format.
The used format depends on the file extension given
in *path*. If no *path* is given, a dialog for
choosing an output file name is opened.
"""
# local import in order to keep namespaces clean
import ms
import sys
from pyopenms import FileHandler
if isinstance(path, unicode):
path = path.encode(sys.getfilesystemencoding())
elif path is None:
path = ms.askForSave(extensions="mzML mzXML mzData".split())
if path is None:
return None
if sys.platform == "win32":
path = path.replace("/","\\") # needed for network shares
experiment = pm.toMSExperiment()
fh = FileHandler()
fh.storeExperiment(path, experiment)
def storePeakMap(pm, path=None):
""" Stores peakmap *pm* in mzXML, mzML or mzData format.
The used format depends on the file extension given
in *path*. If no *path* is given, a dialog for
choosing an output file name is opened.
"""
# local import in order to keep namespaces clean
import ms
import sys
from pyopenms import FileHandler
if isinstance(path, unicode):
path = path.encode(sys.getfilesystemencoding())
elif path is None:
path = ms.askForSave(extensions="mzML mzXML mzData".split())
if path is None:
return None
if sys.platform == "win32":
path = path.replace("/", "\\") # needed for network shares
experiment = pm.toMSExperiment()
fh = FileHandler()
fh.storeExperiment(path, experiment)
def read_ms1_experiment(filepath):
source_experiment = MSExperiment()
file_handler = FileHandler()
# bytes is required by `loadExperiment()` called below
typed_fp = filepath if isinstance(filepath, bytes) else filepath.encode()
file_handler.loadExperiment(typed_fp, source_experiment)
ms1_experiment = MSExperiment()
for spectrum in source_experiment:
if spectrum.getMSLevel() == 1:
ms1_experiment.addSpectrum(spectrum)
return ms1_experiment
def read_ms1_experiment(filepath):
# pylint: disable=import-outside-toplevel,no-name-in-module,import-error
from pyopenms import FileHandler, MSExperiment
source_experiment = MSExperiment()
file_handler = FileHandler()
# bytes is required by `loadExperiment()` called below
typed_fp = filepath if isinstance(filepath,
bytes) else filepath.encode()
file_handler.loadExperiment(typed_fp, source_experiment)
ms1_experiment = MSExperiment()
for spectrum in source_experiment:
if spectrum.getMSLevel() == 1:
ms1_experiment.addSpectrum(spectrum)
return ms1_experiment
def storePeakMap(pm, path=None):
""" Stores peakmap *pm* in mzXML, mzML or mzData format.
The used format depends on the file extension given
in *path*. If no *path* is given, a dialog for
choosing an output file name is opened.
"""
# local import in order to keep namespaces clean
import sys
from pyopenms import FileHandler
path = _prepare_path(path, extensions=["mzML", "mzXML", "mzData"], store=False)
if path is None:
return None
if sys.platform == "win32":
path = path.replace("/", "\\") # needed for network shares
experiment = pm.toMSExperiment()
fh = FileHandler()
fh.storeExperiment(path, experiment)
def test001(self):
exp = MSExperiment()
basename = "SHORT_MS2_FILE.mzData"
here = os.path.dirname(os.path.abspath(__file__))
FileHandler().loadExperiment(os.path.join(here, "data", basename), exp)
assert exp.size() > 0
pc = Precursor()
pc.setMZ(1.0)
pc.setIntensity(100)
s0 = exp[0]
s0.setPrecursors([pc])
s0.setMSLevel(2)
spec = Spectrum.fromMSSpectrum(s0)
settings = InstrumentSettings()
settings.setPolarity(IonSource.Polarity.POSITIVE)
s0.setInstrumentSettings(settings)
self.compare_specs(spec, s0)
specneu = Spectrum.fromMSSpectrum(spec.toMSSpectrum())
self.compare_specs(specneu, s0)
pm = PeakMap.fromMSExperiment(exp)
assert os.path.basename(pm.meta["source"]) == basename
rtmin, rtmax = pm.rtRange(None)
ms1s = pm.msNPeaks(1, rtmin, rtmax)
assert ms1s.shape == (1797, 2), ms1s.shape
ms1s2 = pm.msNPeaks(1, rtmax=rtmax)
assert np.all(ms1s == ms1s2)
ms1s3 = pm.msNPeaks(1, rtmin=0)
assert np.all(ms1s == ms1s3)
spec = pm.spectra[0]
assert len(list(spec)) == len(spec) # calls iter
allrts = pm.allRts()
assert allrts[0] <= pm.rtRange()[0]
assert allrts[-1] >= pm.rtRange()[1]
assert len(allrts) == 41, len(allrts)
level1 = pm.levelNSpecs(1, 1)
level2 = pm.levelNSpecs(2, 2)
level12 = pm.levelNSpecs(1, 2)
assert len(level1) > 0
assert len(level2) > 0
assert len(level1) + len(level2) == len(level12) == len(pm)
assert level1[0].msLevel == 1
assert level2[0].msLevel == 2
# use default arg: nmax = nmin if not provided:
level1 = pm.levelNSpecs(1)
level2 = pm.levelNSpecs(2)
assert len(level1) > 0
assert len(level2) > 0
assert len(level1) + len(level2) == len(level12) == len(pm)
assert level1[0].msLevel == 1
assert level2[0].msLevel == 2
lone = pm.levelOneRts()
assert len(lone) == len(level1)
self.compare_exp(pm, exp, basename)
pm2 = PeakMap.fromMSExperiment(pm.toMSExperiment())
self.compare_exp(pm2, exp, basename)
pm2 = pm.extract(rtmin=rtmin + .000001)
assert len(pm2) == len(pm) - 1
pm2 = pm2.extract(rtmax=rtmax - 0.000001)
assert len(pm2) == len(pm) - 2
mzmin, mzmax = pm.mzRange(2)
assert mzmin < 250
assert mzmax > 860
mzmin, mzmax = pm.mzRange(1)
assert mzmin >= 700
assert mzmax <= 1050
pm2 = pm.extract(rtmin + 0.00001, mzmin=300)
mzmin2, mzmax2 = pm2.mzRange()
assert mzmin2 >= 300
assert mzmax2 == mzmax
pm2 = pm.extract(rtmin=rtmin + 0.000001, mzmin=300, mzmax=1000)
mzmin2, mzmax2 = pm2.mzRange()
assert mzmin2 >= 300
assert mzmax2 <= 1000
with pytest.raises(Exception):
pm.spectra[0].peaksInRange()
pp1 = pm.spectra[0].peaksInRange(mzmax=10000)
pp2 = pm.spectra[0].peaksInRange(mzmin=0)
assert np.all(pp1 == pp2)
specs0 = list(pm.spectra)
specs1 = pm.specsInRange(0, 99999)
specs2 = pm.specsInRange(0, specs0[0].rt)
specs3 = pm.specsInRange(specs0[-1].rt, 999999)
assert specs0 == specs1
assert specs2 == [specs0[0]]
assert specs3 == [specs0[-1]]
pm.spectra[0].polarity = "+"
pm.spectra[1].polarity = "-"
pm = PeakMap(pm.spectra)
mz = pm.representingMzPeak(0, 99999, 0, 99999)
assert abs(mz - 831.86538) < 0.0001
def process(self, peakMap):
assert isinstance(peakMap, PeakMap)
if len(peakMap) == 0:
raise Exception("empty peakmap")
temp_peakmap = _get_temp_peakmap(self.config.get("msLevel"), peakMap)
minRt = peakMap.spectra[0].rt
# matched filter does not like rt <= 0, so we shift that rt starts
# with 1.0: we have to undo this shift later when parsing the output of
# xcms
shift = minRt - 1.0
peakMap.shiftRt(-shift)
td = tempfile.mkdtemp(prefix="emzed_r_script_matched_filter_")
temp_input = os.path.join(td, "input.mzData")
temp_output = os.path.join(td, "output.csv")
# needed for network shares:
if sys.platform == "win32":
temp_input = temp_input.replace("/", "\\")
FileHandler().storeExperiment(temp_input, peakMap.toMSExperiment())
dd = self.config.copy()
dd["temp_input"] = temp_input
dd["temp_output"] = temp_output
dd["index"] = str(dd["index"]).upper()
script = """
library(xcms)
xs <- xcmsSet(%(temp_input)r, method="matchedFilter",
fwhm = %(fwhm)f, sigma = %(sigma)f,
max = %(max_)d,
snthresh = %(snthresh)f,
step = %(step)f, steps=%(steps)d,
mzdiff = %(mzdiff)f,
index = %(index)s,
sleep=0
)
peaks <- data.frame(xs@peaks)
""" % dd
del dd["temp_input"]
del dd["temp_output"]
table = execute(script).get_df_as_table("peaks",
types=dict(mz=float,
mzmin=float,
mzmax=float,
rt=float,
rtmin=float,
ftmax=float,
into=float,
intf=float,
maxo=float,
maxf=float,
i=int,
sn=float,
sample=int),
formats=dict(
mz="%10.5f",
mzmin="%10.5f",
mzmax="%10.5f",
rt=formatSeconds,
rtmin=formatSeconds,
rtmax=formatSeconds,
into="%.2e",
intf="%.2e",
maxo="%.2e",
maxf="%.2e",
sn="%.2e"))
# parse csv and
table.addConstantColumn("matchedfilter_config", dd, dict, None)
table.meta["generator"] = "xcms.matchedfilter"
decorate(table, temp_peakmap)
# undo shiftRt done above:
table.rtmin += shift
table.rtmax += shift
table.rt += shift
return table
def process(self, peakMap):
assert isinstance(peakMap, PeakMap)
if len(peakMap) == 0:
raise Exception("empty peakmap")
temp_peakmap = _get_temp_peakmap(self.config.get("msLevel"), peakMap)
td = tempfile.mkdtemp(prefix="emzed_r_script_centwave_")
temp_input = os.path.join(td, "input.mzData")
temp_output = os.path.join(td, "output.csv")
# needed for network shares:
if sys.platform == "win32":
temp_input = temp_input.replace("/", "\\")
FileHandler().storeExperiment(temp_input,
temp_peakmap.toMSExperiment())
dd = self.config.copy()
dd["temp_input"] = temp_input
dd["temp_output"] = temp_output
dd["fitgauss"] = str(dd["fitgauss"]).upper()
dd["verbose_columns"] = str(dd["verbose_columns"]).upper()
script = """
library(xcms)
xs <- xcmsSet(%(temp_input)r, method="centWave",
ppm=%(ppm)d,
peakwidth=c%(peakwidth)r,
prefilter=c%(prefilter)r,
snthresh = %(snthresh)f,
integrate= %(integrate)d,
mzdiff = %(mzdiff)f,
noise = %(noise)f,
fitgauss = %(fitgauss)s,
verbose.columns = %(verbose_columns)s,
mzCenterFun = %(mzCenterFun)r
)
print(xs@peaks)
peaks <- data.frame(xs@peaks)
""" % dd
del dd["temp_input"]
del dd["temp_output"]
table = execute(script).get_df_as_table("peaks",
types=dict(mz=float,
mzmin=float,
mzmax=float,
rt=float,
rtmin=float,
ftmax=float,
into=float,
intb=float,
maxo=float,
sn=float,
sample=int),
formats=dict(
mz="%10.5f",
mzmin="%10.5f",
mzmax="%10.5f",
rt=formatSeconds,
rtmin=formatSeconds,
rtmax=formatSeconds,
into="%.2e",
intb="%.2e",
maxo="%.2e",
sn="%.2e",
peakmap="%s"))
table.addConstantColumn("centwave_config", dd, dict, None)
table.meta["generator"] = "xcms.centwave"
decorate(table, temp_peakmap)
return table