def test_intensity_nonoverlap_controller_with_simulated_chems(
self, fragscan_dataset):
logger.info(
'Testing intensity non-overlap controller with simulated chemicals'
)
assert len(fragscan_dataset) == N_CHEMS
isolation_width = 1 # the isolation window in Dalton around a selected precursor ion
N = 10
rt_tol = 15
mz_tol = 10
min_roi_intensity = 50
min_roi_length = 0
ionisation_mode = POSITIVE
min_bound, max_bound = get_rt_bounds(fragscan_dataset, CENTRE_RANGE)
rt_box_size, mz_box_size = 1, 0.3
# create a simulated mass spec with noise and ROI controller
mass_spec = IndependentMassSpectrometer(ionisation_mode,
fragscan_dataset)
grid = GridEstimator(
AllOverlapGrid(min_bound, max_bound, rt_box_size, 0, 3000,
mz_box_size), IdentityDrift())
controller = IntensityNonOverlapController(
ionisation_mode,
isolation_width,
mz_tol,
MIN_MS1_INTENSITY,
min_roi_intensity,
min_roi_length,
N,
grid,
rt_tol=rt_tol,
min_roi_length_for_fragmentation=0)
# create an environment to run both the mass spec and controller
env = Environment(mass_spec,
controller,
min_bound,
max_bound,
progress_bar=True)
run_environment(env)
assert len(controller.scans[2]) > 0
# check that there is at least one non-empty MS2 scan
check_non_empty_MS2(controller)
# write simulated output to mzML file
filename = 'intensity_non_overlap_controller_simulated_chems.mzML'
check_mzML(env, OUT_DIR, filename)
def test_flexible_non_overlap_controller_with_beer_chems_and_smartROI_rules(
self):
logger.info(
'Testing flexible non-overlap controller with QC beer chemicals and '
'SmartROI rules')
isolation_width = 1 # the isolation window in Dalton around a selected precursor ion
N = 10
rt_tol = 15
mz_tol = 10
min_roi_intensity = 5000
min_roi_length = 10
ionisation_mode = POSITIVE
rt_box_size, mz_box_size = 1, 0.3
# create a simulated mass spec with noise and ROI controller
mass_spec = IndependentMassSpectrometer(ionisation_mode, BEER_CHEMS)
grid = GridEstimator(
AllOverlapGrid(BEER_MIN_BOUND, BEER_MAX_BOUND, rt_box_size, 0,
3000, mz_box_size), IdentityDrift())
controller = FlexibleNonOverlapController(
ionisation_mode,
isolation_width,
mz_tol,
MIN_MS1_INTENSITY,
min_roi_intensity,
min_roi_length,
N,
grid,
rt_tol=rt_tol,
min_roi_length_for_fragmentation=0,
roi_type=ROI_TYPE_SMART,
reset_length_seconds=1e6,
intensity_increase_factor=10,
drop_perc=0.1 / 100)
# create an environment to run both the mass spec and controller
env = Environment(mass_spec,
controller,
BEER_MIN_BOUND,
BEER_MAX_BOUND,
progress_bar=True)
run_environment(env)
# check that there is at least one non-empty MS2 scan
check_non_empty_MS2(controller)
# write simulated output to mzML file
filename = 'flexible_non_overlap_qcbeer_chems_smartroi.mzML'
check_mzML(env, OUT_DIR, filename)
def test_intensity_non_overlap_controller_with_beer_chems_and_weighteddew_rules(
self):
logger.info(
'Testing intensity non-overlap controller with QC beer chemicals '
'and WeightedDEW rules')
isolation_width = 1 # the isolation window in Dalton around a selected precursor ion
N = 10
rt_tol = 120
exclusion_t_0 = 15
mz_tol = 10
min_roi_intensity = 5000
min_roi_length = 10
ionisation_mode = POSITIVE
rt_box_size, mz_box_size = 1, 0.3
# create a simulated mass spec with noise and ROI controller
mass_spec = IndependentMassSpectrometer(ionisation_mode, BEER_CHEMS)
grid = GridEstimator(
AllOverlapGrid(BEER_MIN_BOUND, BEER_MAX_BOUND, rt_box_size, 0,
3000, mz_box_size), IdentityDrift())
controller = IntensityNonOverlapController(
ionisation_mode,
isolation_width,
mz_tol,
MIN_MS1_INTENSITY,
min_roi_intensity,
min_roi_length,
N,
grid,
rt_tol=rt_tol,
min_roi_length_for_fragmentation=0,
exclusion_method=ROI_EXCLUSION_WEIGHTED_DEW,
exclusion_t_0=exclusion_t_0)
# create an environment to run both the mass spec and controller
env = Environment(mass_spec,
controller,
BEER_MIN_BOUND,
BEER_MAX_BOUND,
progress_bar=True)
run_environment(env)
# check that there is at least one non-empty MS2 scan
check_non_empty_MS2(controller)
# write simulated output to mzML file
filename = 'intensity_non_overlap_qcbeer_chems_weighteddew.mzML'
check_mzML(env, OUT_DIR, filename)
def non_overlap_experiment(datasets,
base_chemicals,
rt_range,
isolation_width,
mz_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
N,
rt_tol,
min_roi_length_for_fragmentation,
rt_box_size,
mz_box_size,
ionisation_mode=POSITIVE):
env_list = []
grid = GridEstimator(
LocatorGrid(rt_range[0], rt_range[1], rt_box_size, 0, 3000,
mz_box_size), IdentityDrift())
for i in range(len(datasets)):
mass_spec = IndependentMassSpectrometer(ionisation_mode, datasets[i])
controller = NonOverlapController(
ionisation_mode,
isolation_width,
mz_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
N,
grid,
rt_tol=rt_tol,
min_roi_length_for_fragmentation=min_roi_length_for_fragmentation)
env = Environment(mass_spec,
controller,
rt_range[0],
rt_range[1],
progress_bar=True)
env.run()
env_list.append(env)
final_evaluation = evaluate_multiple_simulated_env(
env_list, base_chemicals=base_chemicals)
return env_list, final_evaluation
def run_vimms(no_injections, rt_box_size, mz_box_size):
rt_range = [(0, 1440)]
min_rt, max_rt = rt_range[0]
ionisation_mode, isolation_width = POSITIVE, 1
N, rt_tol, mz_tol, min_ms1_intensity = 10, 15, 10, 5000
min_roi_intensity, min_roi_length, min_roi_length_for_fragmentation = \
500, 3, 3
grid = GridEstimator(
LocatorGrid(min_rt, max_rt, rt_box_size, 0, 3000, mz_box_size),
IdentityDrift())
hmdbpath = os.path.join(os.path.abspath(os.getcwd()), "..", "..", "tests",
"fixtures", "hmdb_compounds.p")
hmdb = load_obj(hmdbpath)
df = DatabaseFormulaSampler(hmdb, min_mz=100, max_mz=1000)
cm = ChemicalMixtureCreator(df, adduct_prior_dict={POSITIVE: {"M+H": 1}})
chemicals = cm.sample(2000, 1)
boxes = []
for i in range(no_injections):
mz_noise = GaussianPeakNoise(0.1)
mass_spec = IndependentMassSpectrometer(POSITIVE, chemicals,
mz_noise=mz_noise)
controller = NonOverlapController(
ionisation_mode, isolation_width, mz_tol, min_ms1_intensity,
min_roi_intensity,
min_roi_length, N, grid, rt_tol=rt_tol,
min_roi_length_for_fragmentation=min_roi_length_for_fragmentation
)
env = Environment(mass_spec, controller, min_rt, max_rt,
progress_bar=True)
set_log_level_warning()
env.run()
boxes.append(
[r.to_box(0.01, 0.01) for r in controller.roi_builder.get_rois()])
return boxes
def case_control_non_overlap_experiment_evaluation(
datasets,
min_rt,
max_rt,
N,
isolation_window,
mz_tol,
rt_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
rt_box_size,
mz_box_size,
min_roi_length_for_fragmentation,
scoring_params=None,
base_chemicals=None,
mzmine_files=None,
rt_tolerance=100,
experiment_dir=None,
box_method='mean',
roi_type=ROI_TYPE_NORMAL,
reset_length_seconds=1e6,
intensity_increase_factor=10,
drop_perc=0.1 / 100,
exclusion_method=ROI_EXCLUSION_DEW,
exclusion_t_0=None,
progress_bar=False):
if base_chemicals is not None or mzmine_files is not None:
env_list = []
grid = CaseControlGridEstimator(AllOverlapGrid(min_rt, max_rt,
rt_box_size, 0, 3000,
mz_box_size),
IdentityDrift(),
rt_tolerance=rt_tolerance,
box_method=box_method)
mzml_files = []
source_files = ['sample_' + str(i) for i in range(len(datasets))]
for i in range(len(datasets)):
mass_spec = IndependentMassSpectrometer(POSITIVE, datasets[i])
controller = FlexibleNonOverlapController(
POSITIVE,
isolation_window,
mz_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
N,
grid,
rt_tol=rt_tol,
min_roi_length_for_fragmentation=
min_roi_length_for_fragmentation,
scoring_params=scoring_params,
roi_type=roi_type,
reset_length_seconds=reset_length_seconds,
intensity_increase_factor=intensity_increase_factor,
drop_perc=drop_perc,
exclusion_method=exclusion_method,
exclusion_t_0=exclusion_t_0)
env = Environment(mass_spec,
controller,
min_rt,
max_rt,
progress_bar=progress_bar)
env.run()
if progress_bar is False:
print('Processed dataset ' + str(i))
env_list.append(env)
if base_chemicals is None:
file_link = os.path.join(experiment_dir,
source_files[i] + '.mzml')
mzml_files.append(file_link)
env.write_mzML(experiment_dir, source_files[i] + '.mzml')
if base_chemicals is not None:
evaluation = evaluate_multiple_simulated_env(
env_list, base_chemicals=base_chemicals)
else:
roi_aligner = FrequentistRoiAligner(rt_tolerance=rt_tolerance)
for i in range(len(mzml_files)):
roi_aligner.add_picked_peaks(mzml_files[i], mzmine_files[i],
source_files[i], 'mzmine')
evaluation = evaluate_multi_peak_roi_aligner(
roi_aligner, source_files, True)
return env_list, evaluation
else:
return None, None