def run_ff(self,
exp: ms.MSExperiment,
type: str = 'centroided') -> ms.FeatureMap:
"""Runs an existing OpenMS feature finder on an experiment.
Keyword arguments:
exp: the experiment to run the existing feature finder on
type: the name of the existing feature finder to run
Returns: the features in the experiment.
"""
if type == 'multiplex':
return self.run_ffm(exp)
ff = ms.FeatureFinder()
ff.setLogType(ms.LogType.NONE)
features, seeds = ms.FeatureMap(), ms.FeatureMap()
params = ms.FeatureFinder().getParameters(
type) # default (Leon's) (modified)
params.__setitem__(b'mass_trace:min_spectra', 7) # 10 (5) (7)
params.__setitem__(b'mass_trace:max_missing', 1) # 1 (2) (1)
params.__setitem__(b'seed:min_score', 0.65) # 0.8 (0.5) (0.65)
params.__setitem__(b'feature:min_score', 0.6) # 0.7 (0.5) (0.6)
exp.updateRanges()
ff.run(type, exp, features, params, seeds)
features.setUniqueIds()
return features
def load_feature_maps(self, **kwargs):
self.reference = oms.FeatureMap()
self.toAlign = oms.FeatureMap()
self.xml_file = oms.FeatureXMLFile()
self.xml_file.load(self.input_fm_1, self.reference)
self.xml_file.load(self.input_fm_2, self.toAlign)
def main(self):
#after path_parsing method we have self.src_full_name_list
for f in get_list_full_names(self.src):
print("Map Alignment implementation")
print("Source file:", f)
# to prepare(init) empty list and entity;
self.init_entity(**self.kw)
self.reference_map = oms.FeatureMap()
self.toAlign_map = oms.FeatureMap()
oms.FeatureXMLFile().load(self.reference_file, self.reference_map)
oms.FeatureXMLFile().load(f, self.toAlign_map)
#Set reference_map file
self.ma.entity.setReference(self.reference_map)
#3rd step create object for the computed transformation
transformation = oms.TransformationDescription()
# the 4rd step:
self.ma.entity.align(self.toAlign_map, transformation)
# the 5th step: is store result into file;
self.dst_full_file_name = os.path.join(self.dst,\
convert_src_to_dst_file_name(f,
self.dst,
self.suffix_dst_files,
self.ext_dst_files) )
#print("dst=",dst_full_file_name)
oms.FeatureXMLFile().store(self.dst_full_file_name, self.toAlign_map)
oms.FeatureXMLFile().store(self.dst_full_file_name, self.reference_map)
print("Aligned data stored into:", self.dst_full_file_name)
def run_feature_finder_centroided_on_experiment(input_map):
"""Function that runs FeatureFinderCentroided on the given input map.
Args:
input_map (MSExperiment): An OpenMS MSExperiment object.
Returns:
FeatureMap: A FeatureMap containing the found features from the given
experiment.
"""
# Load data
input_map.updateRanges()
ff = ms.FeatureFinder()
ff.setLogType(ms.LogType.CMD)
# Run the feature finder
name = 'centroided'
features = ms.FeatureMap()
seeds = ms.FeatureMap()
params = ms.FeatureFinder().getParameters(name)
params.__setitem__(b'mass_trace:min_spectra', 5)
params.__setitem__(b'mass_trace:max_missing', 2)
params.__setitem__(b'seed:min_score', 0.5)
params.__setitem__(b'feature:min_score', 0.5)
ff.run(name, input_map, features, params, seeds)
features.setUniqueIds()
return features
def setup_feature_finder(self):
# setting up the FeatureFinder
self.seeds = oms.FeatureMap()
self.ff = oms.FeatureFinder()
self.features = oms.FeatureMap()
self.ff.setLogType(oms.LogType.CMD)
def detect_peaks_gcms_centroid(ms_experiment, parameters, debug=False):
"""
Applicable to centroided experiments, also see https://abibuilder.informatik.uni-tuebingen.de/archive/openms/Documentation/nightly/html/a16103.html
:param ms_experiment:
:param parameters:
:return:
"""
print(f"Detecting peaks with {GCMSPeakDetectionMethod.CENTROIDED}")
ff = oms.FeatureFinder()
if not debug:
ff.setLogType(oms.LogType.NONE)
else:
ff.setLogType(oms.LogType.CMD)
# Run the feature finder
name = "centroided"
pdm_name = GCMSPeakDetectionMethod.CENTROIDED.name
parameters['detection_mode'] = name
parameters['pdm_name'] = pdm_name
# name = parameters['detection_mode']
features = oms.FeatureMap()
seeds = oms.FeatureMap()
ff_params = oms.FeatureFinder().getParameters(name)
ff.run(name, ms_experiment, features, ff_params, seeds)
# features.setUniqueIds()
features.ensureUniqueId()
fh = oms.FeatureXMLFile()
feature_storage_path = f"{parameters['input_filename']}_output.featureXML"
fh.store(feature_storage_path, features)
parameters['feature_storage'] = feature_storage_path
print("Found", features.size(), "features")
return parameters
def detect_peaks_gcms_peak_picker_wavelet(ms_experiment, parameters):
"""
Use isotop wavelet to process raw data - can perform poorly on centroided data
https://abibuilder.informatik.uni-tuebingen.de/archive/openms/Documentation/nightly/html/a16159.html
TODO use "corrected" intensity_type
:param ms_experiment:
:param parameters:
:return:
"""
#set estimate_peak_width to true
# Run the feature finder
name = "peak_picker_wavelet"
parameters['detection_mode'] = name
# outdated code in https://github.com/OpenMS/pyopenms-extra/blob/master/src/examples/peakpicker_scipyFFT.py
# pick spectrum
ff = oms.FeatureFinder()
ff.setLogType(oms.LogType.NONE)
algo_name = oms.FeatureFinderAlgorithmIsotopeWavelet().getProductName()
# picker = oms.FeatureFinderAlgorithmIsotopeWavelet()
feature_map = oms.FeatureMap()
seeds = oms.FeatureMap()
# seeds = FeatureMap()
algo_params = ff.getParameters(algo_name)
ff.run(algo_name, ms_experiment, feature_map, algo_params, seeds)
feature_map.setUniqueIds()
parameters = default_store_feature_xml(feature_map, parameters)
return parameters
def algorithm(exp, targeted, picker):
output = pyopenms.FeatureMap()
chrom_map = {}
pepmap = {}
trmap = {}
for i, chrom in enumerate(exp.getChromatograms()):
chrom_map[chrom.getNativeID()] = i
for i, pep in enumerate(targeted.getPeptides()):
pepmap[pep.id] = i
for i, tr in enumerate(targeted.getTransitions()):
tmp = trmap.get(tr.getPeptideRef(), [])
tmp.append(i)
trmap[tr.getPeptideRef()] = tmp
for key, value in trmap.iteritems():
print key, value
transition_group = getTransitionGroup(exp, targeted, key, value,
chrom_map)
picker.pickTransitionGroup(transition_group)
for mrmfeature in transition_group.getFeatures():
features = mrmfeature.getFeatures()
for f in features:
# TODO
# f.getConvexHulls().clear()
f.ensureUniqueId()
mrmfeature.setSubordinates(
features) # add all the subfeatures as subordinates
output.push_back(mrmfeature)
return output
def test_run_mrmfeaturefinder(self):
# load chromatograms
chromatograms = pyopenms.MSExperiment()
fh = pyopenms.FileHandler()
fh.loadExperiment(self.chromatograms, chromatograms)
# load TraML file
targeted = pyopenms.TargetedExperiment()
tramlfile = pyopenms.TraMLFile()
tramlfile.load(self.tramlfile, targeted)
# Create empty files as input and finally as output
empty_swath = pyopenms.MSExperiment()
trafo = pyopenms.TransformationDescription()
output = pyopenms.FeatureMap()
# set up featurefinder and run
featurefinder = pyopenms.MRMFeatureFinderScoring()
featurefinder.pickExperiment(chromatograms, output, targeted, trafo,
empty_swath)
self.assertAlmostEqual(output.size(), 3)
self.assertAlmostEqual(output[0].getRT(), 3119.092041015, eps)
self.assertAlmostEqual(output[0].getIntensity(), 3614.99755859375, eps)
self.assertAlmostEqual(
output[0].getMetaValue(b"var_xcorr_shape_weighted"),
0.997577965259552, eps)
self.assertAlmostEqual(output[0].getMetaValue(b"sn_ratio"),
86.00413513183594, eps)
def main(self):
#after path_parsing method we have self.src_full_name_list
print("FeatureFindingMetabo implementation")
for f in get_list_full_names(self.src):
print("Source file:", f)
# to prepare(init) empty list and entity;
self.init_entity(**self.kw)
input_map = oms.PeakMap() # the 1st step: load map;
fm = oms.FeatureMap()
oms.MzMLFile().load(f, input_map)
# the 2nd step: apply_ffm;
self.mtd.entity.run(input_map, self.output_mt)
self.epd.entity.detectPeaks(self.output_mt, self.splitted_mt)
self.ffm.entity.run(self.splitted_mt, fm, self.filtered_mt)
# the 3d step: is store result into file;
dst_full_file_name = os.path.join(self.dst,\
convert_src_to_dst_file_name(f,
self.dst,
self.suffix_dst_files,
self.ext_dst_files) )
oms.FeatureXMLFile().store(dst_full_file_name, fm)
print("Centroided data stored into:", dst_full_file_name)
def algorithm(chromatograms, targeted):
# Create empty files as input and finally as output
empty_swath = pyopenms.MSExperiment()
trafo = pyopenms.TransformationDescription()
output = pyopenms.FeatureMap()
# set up featurefinder and run
featurefinder = pyopenms.MRMFeatureFinderScoring()
# set the correct rt use values
scoring_params = pyopenms.MRMFeatureFinderScoring().getDefaults()
scoring_params.setValue("Scores:use_rt_score", 'false', '')
featurefinder.setParameters(scoring_params)
featurefinder.pickExperiment(chromatograms, output, targeted, trafo,
empty_swath)
# get the pairs
pairs = []
simple_find_best_feature(output, pairs, targeted)
pairs_corrected = pyopenms.MRMRTNormalizer().rm_outliers(pairs, 0.95, 0.6)
pairs_corrected = [list(p) for p in pairs_corrected]
# // store transformation, using a linear model as default
trafo_out = pyopenms.TransformationDescription()
trafo_out.setDataPoints(pairs_corrected)
model_params = pyopenms.Param()
model_params.setValue("symmetric_regression", 'false', '')
model_type = "linear"
trafo_out.fitModel(model_type, model_params)
return trafo_out
def algorithm(exp, targeted, picker, scorer, trafo):
output = pyopenms.FeatureMap()
scorer.prepareProteinPeptideMaps_(targeted)
chrom_map = {}
pepmap = {}
trmap = {}
for i, chrom in enumerate(exp.getChromatograms()):
chrom_map[ chrom.getNativeID() ] = i
for i, pep in enumerate(targeted.getCompounds() ):
pepmap[ pep.id ] = i
for i, tr in enumerate(targeted.getTransitions() ):
tmp = trmap.get( tr.getPeptideRef() , [])
tmp.append( i )
trmap[ tr.getPeptideRef() ] = tmp
swath_maps_dummy = []
for key, value in trmap.iteritems():
try:
transition_group = getTransitionGroup(exp, targeted, key, value, chrom_map)
except Exception:
print "Skip ", key, value
continue
picker.pickTransitionGroup(transition_group);
scorer.scorePeakgroups(transition_group, trafo, swath_maps_dummy, output, False);
return output
def detect_peaks_gcms_isotopewavelet(ms_experiment, parameters, debug=False):
"""
Use isotop wavelet to process raw data - can perform poorly on centroided data
- also see https://abibuilder.informatik.uni-tuebingen.de/archive/openms/Documentation/nightly/html/a16105.html
TODO use "corrected" intensity_type
:param ms_experiment:
:param parameters:
:return:
"""
print(f"Detecting peaks with {GCMSPeakDetectionMethod.ISOTOPEWAVELET}")
ff = oms.FeatureFinder()
if not debug:
ff.setLogType(oms.LogType.NONE)
else:
ff.setLogType(oms.LogType.CMD)
# Run the feature finder
name = "isotope_wavelet"
pdm_name = GCMSPeakDetectionMethod.ISOTOPEWAVELET.name
parameters['detection_mode'] = name
parameters['pdm_name'] = pdm_name
# name = parameters['detection_mode']
features = oms.FeatureMap()
seeds = oms.FeatureMap()
ff_params = ff.getParameters(name)
# complains about "the extremal length of the wavelet is larger (47661) than the number of data points"
# wavelet_length is defined by mz_cutoff / min_spacing
# hr_data must be true if high-resolution data (orbitrap, FTICR)
# hr_data parameter for isotopewavelet function
is_hr_data = parameters.get("hr_data", False)
if is_hr_data:
hr_key = b"hr_data" # hr_data takes extremely long - >= 2h per measurement of (!)32MB - there are way larger spectra...
ff_params.setValue(hr_key, b"true")
ff.run(name, ms_experiment, features, ff_params, seeds)
features.setUniqueIds()
fh = oms.FeatureXMLFile()
feature_storage_path = f"{parameters['input_filename']}_output.featureXML"
fh.store(feature_storage_path, features)
parameters['feature_storage'] = feature_storage_path
print("Found", features.size(), "features")
return parameters
def run_featurefinder_centroided(input_map, params, seeds, out_path):
ff = pms.FeatureFinder()
ff.setLogType(pms.LogType.CMD)
features = pms.FeatureMap()
name = pms.FeatureFinderAlgorithmPicked.getProductName()
ff.run(name, input_map, features, params, seeds)
fh = pms.FeatureXMLFile()
fh.store(out_path, features)
def fit(self, filenames, max_peaks_per_file=1000):
try:
feature_map = oms.FeatureMap()
except:
pass
n_files = len(filenames)
for i, fn in enumerate(filenames):
self.progress = 100*(i+1)/n_files
feature_map += oms_ffmetabo_single_file(
fn, max_peaks_per_file=max_peaks_per_file
)
self._feature_map = feature_map
def testFeatureFinder():
"""
@tests:
FeatureFinder.__init__
FeatureFinder.endProgress
FeatureFinder.getLogType
FeatureFinder.getParameters
FeatureFinder.run
FeatureFinder.setLogType
FeatureFinder.setProgress
FeatureFinder.startProgress
"""
ff = pyopenms.FeatureFinder()
name = pyopenms.FeatureFinderAlgorithmPicked.getProductName()
ff.run(name, pyopenms.MSExperiment(), pyopenms.FeatureMap(),
pyopenms.Param(), pyopenms.FeatureMap())
_testProgressLogger(ff)
p = ff.getParameters(name)
_testParam(p)
def collect_convex_hulls(self):
self.convex_hulls = []
# opening featureXML
xml_file = oms.FeatureXMLFile()
self.fmap = oms.FeatureMap()
xml_file.load(self.feature_xml_fname, self.fmap)
feature_mzs = []
for i, fe in enumerate(self.fmap):
feature_mzs.append([i, fe.getMZ()])
feature_mzs = np.array(feature_mzs)
feature_mzs = feature_mzs[feature_mzs[:, 1].argsort(), :]
# looking up the example features in the featureXML
self.examples_oms_features = {}
for ex, (mz_t, mz_m) in self.peaks_peaks.items():
i_mz_fe = lookup.find(feature_mzs[:, 1], mz_m, t=10)
if i_mz_fe:
self.examples_oms_features[feature_mzs[i_mz_fe, 0]] = ex
# collecting convex hulls
for ife, fe in enumerate(self.fmap):
if ife in self.examples_oms_features:
hull_list = fe.getConvexHulls()
self.extend_hulls(hull_list, ife, 0)
subord_feature = fe.getSubordinates()
if subord_feature:
for subfe in subord_feature:
hull_list = subfe.getConvexHulls()
self.extend_hulls(hull_list, ife, 1)
# columns: rt, mz, feature index, hull index, is sub-feature
self.convex_hulls = np.vstack(self.convex_hulls)
self.oms_feature_mzs = feature_mzs[feature_mzs[:, 0].argsort(), :]
def id_mapper(in_file, id_file, out_file, params, use_centroid_rt,
use_centroid_mz, use_subelements):
in_type = pms.FileHandler.getType(in_file)
protein_ids = []
peptide_ids = []
pms.IdXMLFile().load(id_file, protein_ids, peptide_ids)
mapper = pms.IDMapper()
mapper.setParameters(params)
if in_type == pms.Type.CONSENSUSXML:
file_ = pms.ConsensusXMLFile()
map_ = pms.ConsensusMap()
file_.load(in_file, map_)
mapper.annotate(map_, peptide_ids, protein_ids, use_subelements)
addDataProcessing(
map_, params,
pms.DataProcessing.ProcessingAction.IDENTIFICATION_MAPPING)
file_.store(out_file, map_)
elif in_type == pms.Type.FEATUREXML:
file_ = pms.FeatureXMLFile()
map_ = pms.FeatureMap()
file_.load(in_file, map_)
mapper.annotate(map_, peptide_ids, protein_ids, use_centroid_rt,
use_centroid_mz)
addDataProcessing(
map_, params,
pms.DataProcessing.ProcessingAction.IDENTIFICATION_MAPPING)
file_.store(out_file, map_)
elif in_type == pms.Type.MZQ:
file_ = pms.MzQuantMLFile()
msq = pms.MSQuantifications()
file_.load(in_file, msq)
maps = msq.getConsensusMaps()
for map_ in maps:
mapper.annotate(map_, peptide_ids, protein_ids, use_subelements)
addDataProcessing(
map_, params,
pms.DataProcessing.ProcessingAction.IDENTIFICATION_MAPPING)
msq.setConsensusMaps(maps)
file_.store(out_file, msq)
else:
raise Exception("invalid input file format")
def mgf(self):
"""
Opens an mzML file.
"""
options = oms.PeakFileOptions()
options.setMSLevels([2])
self.mzml = oms.MzMLFile()
self.mzml.setOptions(options)
self.exp = oms.MSExperiment()
self.mzml.load(self.fname, self.exp)
self.feature_finder = oms.FeatureFinder()
self.ffname = 'centroided'
self.features = oms.FeatureMap()
self.seeds = oms.FeatureMap()
self.params = oms.FeatureFinder().getParameters(self.ffname)
self.feature_finder.run(
self.ffname,
self.exp,
self.features,
self.params,
self.seeds,
)
self.features.setUniqueIds()
def match_features_internal(self,
features: ms.FeatureMap) -> ms.FeatureMap:
"""Matches features in a single bin; intended to correct satellite features.
The feature in each feature set with the largest convex hull becomes the 'representative'
feature of that set and the rest are discarded.
Keyword arguments:
features: the features of a single bin for intra-bin matching
Returns: a matched set of features.
"""
features.sortByRT()
matched = ms.FeatureMap()
for i in range(features.size()):
feature1 = features[i]
max_area = util.polygon_area(
feature1.getConvexHull().getHullPoints())
max_feature = feature1
similar = []
first_idx = util.binary_search_left_rt(
features,
feature1.getRT() - self.RT_THRESHOLD)
for j in range(first_idx, features.size()):
if i == j:
continue
feature2 = features[j]
if feature2.getRT() > feature1.getRT() + self.RT_THRESHOLD:
break
if util.similar_features(feature1, feature2, self.RT_THRESHOLD,
self.MZ_THRESHOLD):
similar.append(feature2)
for feature2 in similar:
area = util.polygon_area(
feature2.getConvexHull().getHullPoints())
if area > max_area:
max_area = area
max_feature = feature2
if max_feature not in matched:
matched.push_back(max_feature)
return matched
def oms_ffmetabo_single_file(filename, max_peaks_per_file=5000):
feature_map = oms.FeatureMap()
mass_traces = []
mass_traces_split = []
mass_traces_filtered = []
exp = oms.MSExperiment()
peak_map = oms.PeakMap()
options = oms.PeakFileOptions()
options.setMSLevels([1])
if filename.lower().endswith('.mzxml'):
fh = oms.MzXMLFile()
elif filename.lower().endswith('.mzml'):
fh = oms.MzMLFile()
else:
assert False, filename
fh.setOptions(options)
# Peak map
fh.load(filename, exp)
#for chrom in exp.getChromatograms():
# peak_map.addChrom(chrom)
for spec in exp.getSpectra():
peak_map.addSpectrum(spec)
mass_trace_detect = oms.MassTraceDetection()
mass_trace_detect.run(peak_map, mass_traces, max_peaks_per_file)
elution_peak_detection = oms.ElutionPeakDetection()
elution_peak_detection.detectPeaks(mass_traces, mass_traces_split)
feature_finding_metabo = oms.FeatureFindingMetabo()
feature_finding_metabo.run(
mass_traces_split,
feature_map,
mass_traces_filtered)
feature_map.sortByOverallQuality()
return feature_map
def testFeatureXMLFile():
"""
@tests:
FeatureXMLFile.__init__
FeatureXMLFile.load
FeatureXMLFile.store
FileHandler.__init__
FileHandler.loadFeatures
"""
fm = pyopenms.FeatureMap()
fm.setUniqueIds()
fh = pyopenms.FeatureXMLFile()
fh.store("test.featureXML", fm)
fh.load("test.featureXML", fm)
fh = pyopenms.FileHandler()
fh.loadFeatures("test.featureXML", fm)
def test_run_mrmrtnormalizer(self):
# load chromatograms
chromatograms = pyopenms.MSExperiment()
fh = pyopenms.FileHandler()
fh.loadExperiment(self.chromatograms, chromatograms)
# load TraML file
targeted = pyopenms.TargetedExperiment()
tramlfile = pyopenms.TraMLFile()
tramlfile.load(self.tramlfile, targeted)
# Create empty files as input and finally as output
empty_swath = pyopenms.MSExperiment()
trafo = pyopenms.TransformationDescription()
output = pyopenms.FeatureMap()
# set up featurefinder and run
featurefinder = pyopenms.MRMFeatureFinderScoring()
# set the correct rt use values
scoring_params = pyopenms.MRMFeatureFinderScoring().getDefaults()
scoring_params.setValue("Scores:use_rt_score".encode(),
'false'.encode(), ''.encode())
featurefinder.setParameters(scoring_params)
featurefinder.pickExperiment(chromatograms, output, targeted, trafo,
empty_swath)
# get the pairs
pairs = []
simple_find_best_feature(output, pairs, targeted)
pairs_corrected = pyopenms.MRMRTNormalizer().removeOutliersIterative(
pairs, 0.95, 0.6, True, "iter_jackknife")
pairs_corrected = [list(p) for p in pairs_corrected]
expected = [(1497.56884765625, 1881.0), (2045.9776611328125, 2409.0),
(2151.4814453125, 2509.0), (1924.0750732421875, 2291.0),
(612.9832153320312, 990.0), (1086.2474365234375, 1470.0),
(1133.89404296875, 1519.0), (799.5291137695312, 1188.0),
(1397.1541748046875, 1765.0)]
for exp, res in zip(expected, pairs_corrected):
self.assertAlmostEqual(exp[0], res[0], eps)
self.assertAlmostEqual(exp[1], res[1], eps)
def parse_featureXML_GT(feature_file):
featuremap = pyopenms.FeatureMap()
featurexml = pyopenms.FeatureXMLFile()
featurexml.load(feature_file, featuremap)
hulls = pd.DataFrame(
columns=['rt_min', 'rt_max', 'mz_min', 'mz_max', 'detected', 'pic_id'])
for i in range(featuremap.size()):
feature = featuremap[i]
chs = feature.getConvexHulls()
for j in range(len(chs)):
pts = chs[j].getHullPoints()
hulls.loc[len(hulls)] = [
pts.min(0)[0],
pts.max(0)[0],
pts.min(0)[1],
pts.max(0)[1], False, -1
]
return hulls
def FeatureFindingMetabo(mzfile, noise_threshold_int, snr):
finder = 'C:/Program Files/OpenMS/bin/FeatureFinderMetabo.exe'
feature_file = 'tmp.featureXML'
noise_threshold_int = noise_threshold_int / snr
subprocess.call([finder, '-in', mzfile, '-out', feature_file,
'-algorithm:common:noise_threshold_int', f'{noise_threshold_int}',
'-algorithm:common:chrom_peak_snr', f'{snr}',
'-algorithm:common:chrom_fwhm', '10',
'-algorithm:mtd:mass_error_ppm', '20',
'-algorithm:mtd:reestimate_mt_sd', 'true',
'-algorithm:mtd:min_sample_rate', '0',
'-algorithm:mtd:min_trace_length', '2',
'-algorithm:epd:width_filtering', 'off',
'-algorithm:ffm:charge_lower_bound', '1',
'-algorithm:ffm:charge_lower_bound', '5'])
featuremap = pyopenms.FeatureMap()
featurexml = pyopenms.FeatureXMLFile()
featurexml.load(feature_file, featuremap)
os.remove(feature_file)
return featuremap
def FeatureFindingMetabo1(mzfile):
exp = pyopenms.MSExperiment()
pyopenms.MzMLFile().load(mzfile, exp)
mtd_params = pyopenms.MassTraceDetection().getDefaults()
mtd = pyopenms.MassTraceDetection()
mtd.setParameters(mtd_params)
mass_traces = []
mtd.run(exp, mass_traces)
epdet_params = pyopenms.ElutionPeakDetection().getDefaults()
epdet = pyopenms.ElutionPeakDetection()
epdet.setParameters(epdet_params)
splitted_mass_traces = []
epdet.detectPeaks(mass_traces, splitted_mass_traces)
ffm_params = pyopenms.FeatureFindingMetabo().getDefaults()
ffm = pyopenms.FeatureFindingMetabo()
ffm.setParameters(ffm_params)
feature_map = pyopenms.FeatureMap()
ffm.run(splitted_mass_traces, feature_map)
return feature_map
def run_featurefinder_centroided(input_path, params, seeds, out_path):
fh = pms.MzMLFile()
options = pms.PeakFileOptions()
options.setMSLevels([1,1])
fh.setOptions(options)
input_map = pms.MSExperiment()
fh.load(input_path, input_map)
input_map.updateRanges()
ff = pms.FeatureFinder()
ff.setLogType(pms.LogType.CMD)
features = pms.FeatureMap()
name = pms.FeatureFinderAlgorithmPicked.getProductName()
ff.run(name, input_map, features, params, seeds)
features.setUniqueIds()
addDataProcessing(features, params, pms.ProcessingAction.QUANTITATION)
fh = pms.FeatureXMLFile()
fh.store(out_path, features)
def extract_data(in_file_name, mz_search):
xml_file = oms.FeatureXMLFile()
fmap = oms.FeatureMap()
xml_file.load(in_file_name, fmap)
delta = 0.01
#print( "FeatureMap size=", fmap.size() )
for n in fmap:
#if mz_search == n.getMZ():
if abs(mz_search - n.getMZ()) < delta:
"""
print( "mz=", n.getMZ(),
"rt=", n.getRT(),
" intensity=", n. getIntensity(),
"width=", n.getWidth(),
"charge=", n. getCharge() )
"""
hull_list = n.getConvexHulls() #getConvexHull() return ConvexHull2D;
for hull in hull_list:
hull_points = hull.getHullPoints() # hull_points is numpy.ndarray;
#print( "hull_points.size=", hull_points.size )
for p in hull_points:
print( p[0], p[1] )
subord_feature = n.getSubordinates()
if subord_feature:
#print("getSubordinates:")
for f in subord_feature:
hull_list = f.getConvexHulls()
for hull in hull_list:
hull_points = hull.getHullPoints() # hull_points is numpy.ndarray;
#print( "hull_points.size=", hull_points.size )
for p in hull_points:
print( p[0], p[1] )
else:
continue
def main(options):
# load featureXML
features = pyopenms.FeatureMap()
fh = pyopenms.FileHandler()
fh.loadFeatures(options.infile, features)
# load TraML file
targeted = pyopenms.TargetedExperiment();
tramlfile = pyopenms.TraMLFile();
tramlfile.load(options.traml_in, targeted);
# write TSV file
filename = options.infile.split(".")[0]
fh = open(options.outfile, "w")
wr = csv.writer(fh, delimiter='\t')
header = get_header(features)
wr.writerow(header)
for f in features:
keys = []
f.getKeys(keys)
row = convert_to_row(f,targeted,filename,keys,filename)
wr.writerow(row)
import pyopenms
"""
Producing the test data for TOPP_FeatureLinkerUnlabeledQT_5 and TOPP_FeatureLinkerUnlabeledQT_6
"""
fmaps = [pyopenms.FeatureMap() for i in range(3)]
pepids = []
pepseq = ["PEPTIDEA", "PEPTIDEK", "PEPTIDER"]
for s in pepseq:
pepid = pyopenms.PeptideIdentification()
hit = pyopenms.PeptideHit()
hit.setSequence(pyopenms.AASequence.fromString(s, True))
pepid.insertHit(hit)
pepid.setIdentifier("Protein0")
pepids.append(pepid)
protid = pyopenms.ProteinIdentification()
protid.setIdentifier("Protein0")
for i, fmap in enumerate(fmaps):
fmap.setProteinIdentifications([protid])
# add 3 features to each map, but with a twist (adding different peptide ids to different maps)
for k in range(3):
f = pyopenms.Feature()
f.setRT(300 + k * 100 + i * 10)
f.setMZ(500 + k * 0.001 + i * 0.01)
f.setIntensity(500 + i * 100)
f.setMetaValue("sequence",
pepseq[(i + k) % 3]) # easier viewing in TOPPView
f.setPeptideIdentifications([pepids[(i + k) % 3]])
fmap.push_back(f)
pyopenms.FeatureXMLFile().store("output_%s.featureXML" % i, fmap)
