def top_n_roi_experiment_evaluation(datasets,
min_rt,
max_rt,
N,
isolation_window,
mz_tol,
rt_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
base_chemicals=None,
mzmine_files=None,
rt_tolerance=100,
experiment_dir=None,
progress_bar=False):
if base_chemicals is not None or mzmine_files is not None:
env_list = []
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 = TopN_RoiController(POSITIVE,
isolation_window,
mz_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
N=N,
rt_tol=rt_tol)
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 = RoiAligner(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)
return env_list, evaluation
else:
return None, None
def _update_grid(self):
self.grid.boxes = self.grid.init_boxes(self.grid.rtboxes,
self.grid.mzboxes)
roi_aligner = RoiAligner(rt_tolerance=self.rt_tolerance)
for inj_num, inj in enumerate(self.observed_rois):
fn = self.drift_models[inj_num].get_estimator(inj_num)
rt_shifts = [-fn(roi, inj_num)[0] for roi in inj]
roi_aligner.add_sample(self.observed_rois,
self.grid.sample_number,
rt_shifts=rt_shifts)
boxes = roi_aligner.get_boxes(
method=self.box_method) # TODO might need to add intensity here
for box in boxes:
self.grid.register_box(box)
def top_n_box_experiment(datasets,
base_chemicals,
rt_range,
boxes_params,
dataset_group_list,
isolation_width,
mz_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
N,
rt_tol,
ionisation_mode=POSITIVE):
env_list = []
aligner = RoiAligner()
boxes = None
boxes_intensity = None
for i in range(len(datasets)):
mass_spec = IndependentMassSpectrometer(ionisation_mode, datasets[i])
controller = TopNBoxRoiController(ionisation_mode,
isolation_width,
mz_tol,
min_ms1_intensity,
min_roi_intensity,
min_roi_length,
boxes_params=boxes_params,
boxes=boxes,
boxes_intensity=boxes_intensity,
N=N,
rt_tol=rt_tol)
env = Environment(mass_spec,
controller,
rt_range[0],
rt_range[1],
progress_bar=True)
env.run()
env_list.append(env)
rois = env.controller.live_roi + env.controller.dead_roi
aligner.add_sample(rois, 'sample_' + str(i), dataset_group_list[i])
boxes = aligner.get_boxes()
boxes_intensity = aligner.get_max_frag_intensities()
final_evaluation = evaluate_multiple_simulated_env(
env_list, base_chemicals=base_chemicals)
return env_list, final_evaluation
def dsda_experiment_evaluation(datasets,
base_dir,
min_rt,
max_rt,
N,
isolation_window,
mz_tol,
rt_tol,
min_ms1_intensity,
mzmine_files=None,
rt_tolerance=100,
progress_bar=False):
data_dir = os.path.join(base_dir, 'Data')
schedule_dir = os.path.join(base_dir, 'settings')
mass_spec = IndependentMassSpectrometer(
POSITIVE, datasets[0]) # necessary to get timings for schedule
create_dsda_schedule(mass_spec, N, min_rt, max_rt, base_dir)
print('Please open and run R script now')
time.sleep(1)
template_file = os.path.join(base_dir, 'DsDA_Timing_schedule.csv')
env_list = []
mzml_files = []
source_files = ['sample_' + "%03d" % i for i in range(len(datasets))]
for i in range(len(datasets)):
mass_spec = IndependentMassSpectrometer(POSITIVE, datasets[i])
if i == 0:
controller = TopNController(POSITIVE,
N,
isolation_window,
mz_tol,
rt_tol,
min_ms1_intensity,
ms1_shift=0,
initial_exclusion_list=None,
force_N=False)
else:
print('Looking for next schedule')
new_schedule = get_schedule(i, schedule_dir)
print('Found next schedule')
time.sleep(1)
schedule_param_list = dsda_get_scan_params(new_schedule,
template_file,
isolation_window,
mz_tol, rt_tol)
controller = FixedScansController(schedule=schedule_param_list)
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)
file_link = os.path.join(data_dir, source_files[i] + '.mzml')
mzml_files.append(file_link)
print("Processed ", i + 1, " files")
env.write_mzML(data_dir, source_files[i] + '.mzml')
print("Waiting for R to process .mzML files")
if mzmine_files is None:
evaluation = evaluate_multiple_simulated_env(env_list)
else:
roi_aligner = RoiAligner(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)
return env_list, evaluation
else:
return None, None
def flexible_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,
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 = GridEstimator(
AllOverlapGrid(min_rt, max_rt, rt_box_size, 0, 3000, mz_box_size),
IdentityDrift())
mzml_files = []
source_files = ['sample_' + str(i) for i in range(len(datasets))]
if scoring_params['theta3'] != 0:
register_all_roi = True
else:
register_all_roi = False
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,
register_all_roi=register_all_roi,
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 = RoiAligner(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)
return env_list, evaluation
else:
return None, None