def test_find_matching_sidecar(return_bids_test_dir):
"""Test finding a sidecar file from a BIDS dir."""
bids_root = return_bids_test_dir
bids_fpath = bids_path.copy().update(root=bids_root)
# Now find a sidecar
sidecar_fname = _find_matching_sidecar(bids_fpath,
suffix='coordsystem',
extension='.json')
expected_file = op.join('sub-01', 'ses-01', 'meg',
'sub-01_ses-01_coordsystem.json')
assert sidecar_fname.endswith(expected_file)
# Find multiple sidecars, tied in score, triggering an error
with pytest.raises(RuntimeError, match='Expected to find a single'):
open(sidecar_fname.replace('coordsystem.json', '2coordsystem.json'),
'w').close()
print_dir_tree(bids_root)
_find_matching_sidecar(bids_fpath,
suffix='coordsystem',
extension='.json')
# Find nothing and raise.
with pytest.raises(RuntimeError, match='Did not find any'):
fname = _find_matching_sidecar(bids_fpath,
suffix='foo',
extension='.bogus')
# Find nothing and receive None and a warning.
on_error = 'warn'
with pytest.warns(RuntimeWarning, match='Did not find any'):
fname = _find_matching_sidecar(bids_fpath,
suffix='foo',
extension='.bogus',
on_error=on_error)
assert fname is None
# Find nothing and receive None.
on_error = 'ignore'
fname = _find_matching_sidecar(bids_fpath,
suffix='foo',
extension='.bogus',
on_error=on_error)
assert fname is None
# Invalid on_error.
on_error = 'hello'
with pytest.raises(ValueError, match='Acceptable values for on_error are'):
_find_matching_sidecar(bids_fpath,
suffix='coordsystem',
extension='.json',
on_error=on_error)
"447",
"448",
"449",
"451",
]
event_id = {
"lexical/low": 11,
"lexical/high": 21,
"nonlex/low": 31,
"nonlex/high": 41,
"target/low": 16,
"target/high": 26,
}
print_dir_tree(bids_root, max_depth=3)
datatype = "meg"
bids_path = BIDSPath(root=bids_root, datatype=datatype)
task = "lexicaldecision"
suffix = "meg"
for subject in subjects:
bids_path = BIDSPath(
subject=subject,
task=task,
suffix=suffix,
datatype=datatype,
root=bids_root,
)
# %% # Download the data # ----------------- # # First, we need some data to work with. We will use the # `Finger Tapping Dataset <https://github.com/rob-luke/BIDS-NIRS-Tapping>`_ # available on GitHub. # We will use the MNE-NIRS package which includes convenient functions to # download openly available datasets. data_dir = mne_nirs.datasets.fnirs_motor_group.data_path() # Let's see whether the data has been downloaded using a quick visualization # of the directory tree. print_dir_tree(data_dir) # %% # The data are already in BIDS format. However, we will just use one of the # SNIRF files and demonstrate how this could be used to generate a new BIDS # compliant dataset from this single file. # Specify file to use as input to BIDS generation process file_path = data_dir / "sub-01" / "nirs" / "sub-01_task-tapping_nirs.snirf" # %% # Convert to BIDS # --------------- # # Let's start with loading the data and updating the annotations. # We are reading the data using MNE-Python's ``io`` module and the
subject_id = '01'
task = 'somato'
datatype = 'meg'
bids_path = BIDSPath(subject=subject_id,
task=task,
datatype=datatype,
suffix=datatype,
root=bids_root)
# bids basename is nicely formatted
print(bids_path)
###############################################################################
# Print the directory tree
print_dir_tree(bids_root)
###############################################################################
# Step 2: Read a BIDS dataset
# ---------------------------
#
# Let's read in the dataset and show off a few features of the
# loading function `read_raw_bids`. Note, this is just one line of code.
raw = read_raw_bids(bids_path=bids_path, verbose=True)
###############################################################################
# `raw.info` has the basic subject metadata
print(raw.info['subject_info'])
# `raw.info` has the PowerLineFrequency loaded in, which should be 50 Hz here
print(raw.info['line_freq'])
def test_match(return_bids_test_dir):
"""Test retrieval of matching basenames."""
bids_root = Path(return_bids_test_dir)
bids_path_01 = BIDSPath(root=bids_root)
paths = bids_path_01.match()
assert len(paths) == 9
assert all('sub-01_ses-01' in p.basename for p in paths)
assert all([p.root == bids_root for p in paths])
bids_path_01 = BIDSPath(root=bids_root, run='01')
paths = bids_path_01.match()
assert len(paths) == 3
assert paths[0].basename == ('sub-01_ses-01_task-testing_run-01_'
'channels.tsv')
bids_path_01 = BIDSPath(root=bids_root, subject='unknown')
paths = bids_path_01.match()
assert len(paths) == 0
bids_path_01 = bids_path.copy().update(root=None)
with pytest.raises(RuntimeError, match='Cannot match'):
bids_path_01.match()
bids_path_01.update(datatype='meg', root=bids_root)
same_paths = bids_path_01.match()
assert len(same_paths) == 3
# Check handling of `extension`, part 1: no extension specified.
bids_path_01 = BIDSPath(root=bids_root, run='01')
paths = bids_path_01.match()
assert [p.extension for p in paths] == ['.tsv', '.tsv', '.fif']
# Check handling of `extension`, part 2: extension specified.
bids_path_01 = BIDSPath(root=bids_root,
run='01',
extension='.fif',
datatype='meg')
paths = bids_path_01.match()
assert len(paths) == 1
assert paths[0].extension == '.fif'
# Check handling of `extension` and `suffix`, part 1: no suffix
bids_path_01 = BIDSPath(root=bids_root,
run='01',
extension='.tsv',
datatype='meg')
paths = bids_path_01.match()
assert len(paths) == 2
assert paths[0].extension == '.tsv'
# Check handling of `extension` and `suffix`, part 1: suffix passed
bids_path_01 = BIDSPath(root=bids_root,
run='01',
suffix='channels',
extension='.tsv',
datatype='meg')
paths = bids_path_01.match()
assert len(paths) == 1
assert paths[0].extension == '.tsv'
assert paths[0].suffix == 'channels'
# Check handling of `datatype` when explicitly passed in
print_dir_tree(bids_root)
bids_path_01 = BIDSPath(root=bids_root,
run='01',
suffix='channels',
extension='.tsv',
datatype='meg')
paths = bids_path_01.match()
print(paths)
assert len(paths) == 1
assert paths[0].extension == '.tsv'
assert paths[0].suffix == 'channels'
assert Path(paths[0]).parent.name == 'meg'
# Check handling of `datatype`, no datatype passed in
# should be exactly the same if there is only one datatype
# present in the dataset
bids_path_01 = BIDSPath(root=bids_root,
run='01',
suffix='channels',
extension='.tsv')
paths = bids_path_01.match()
assert len(paths) == 1
assert paths[0].extension == '.tsv'
assert paths[0].suffix == 'channels'
assert Path(paths[0]).parent.name == 'meg'
def test_print_dir_tree(capsys):
"""Test printing a dir tree."""
with pytest.raises(ValueError, match='Directory does not exist'):
print_dir_tree('i_dont_exist')
# We check the testing directory
test_dir = op.dirname(__file__)
with pytest.raises(ValueError, match='must be a positive integer'):
print_dir_tree(test_dir, max_depth=-1)
with pytest.raises(ValueError, match='must be a positive integer'):
print_dir_tree(test_dir, max_depth='bad')
# Do not limit depth
print_dir_tree(test_dir)
captured = capsys.readouterr()
assert '|--- test_utils.py' in captured.out.split('\n')
assert '|--- __pycache__{}'.format(os.sep) in captured.out.split('\n')
assert '.pyc' in captured.out
# Now limit depth ... we should not descend into pycache
print_dir_tree(test_dir, max_depth=1)
captured = capsys.readouterr()
assert '|--- test_utils.py' in captured.out.split('\n')
assert '|--- __pycache__{}'.format(os.sep) in captured.out.split('\n')
assert '.pyc' not in captured.out
# Limit depth even more
print_dir_tree(test_dir, max_depth=0)
captured = capsys.readouterr()
assert captured.out == '|tests{}\n'.format(os.sep)
# test if pathlib.Path object
print_dir_tree(Path(test_dir))
raw = mne.io.read_raw_fif(raw_fname)
raw.info['line_freq'] = 60 # specify power line frequency as required by BIDS
sub = '01'
ses = '01'
task = 'audiovisual'
run = '01'
bids_path = BIDSPath(subject=sub, session=ses, task=task,
run=run, root=output_path)
write_raw_bids(raw, bids_path, events_data=events_data,
event_id=event_id, overwrite=True)
###############################################################################
# Print the directory tree
print_dir_tree(output_path)
###############################################################################
# Now let's assume that we have also collected some T1 weighted MRI data for
# our subject. And furthermore, that we have already aligned our coordinate
# frames (using e.g., the `coregistration GUI`_) and obtained a transformation
# matrix :code:`trans`.
# Get the path to our MRI scan
t1_mgh_fname = op.join(data_path, 'subjects', 'sample', 'mri', 'T1.mgz')
# Load the transformation matrix and show what it looks like
trans_fname = op.join(data_path, 'MEG', 'sample',
'sample_audvis_raw-trans.fif')
trans = mne.read_trans(trans_fname)
print(trans)
def test_find_matching_sidecar(return_bids_test_dir, tmp_path):
"""Test finding a sidecar file from a BIDS dir."""
bids_root = return_bids_test_dir
bids_path = _bids_path.copy().update(root=bids_root)
# Now find a sidecar
sidecar_fname = _find_matching_sidecar(bids_path,
suffix='coordsystem',
extension='.json')
expected_file = op.join('sub-01', 'ses-01', 'meg',
'sub-01_ses-01_coordsystem.json')
assert sidecar_fname.endswith(expected_file)
# Find multiple sidecars, tied in score, triggering an error
with pytest.raises(RuntimeError, match='Expected to find a single'):
open(sidecar_fname.replace('coordsystem.json',
'2coordsystem.json'), 'w').close()
print_dir_tree(bids_root)
_find_matching_sidecar(bids_path,
suffix='coordsystem', extension='.json')
# Find nothing and raise.
with pytest.raises(RuntimeError, match='Did not find any'):
fname = _find_matching_sidecar(bids_path, suffix='foo',
extension='.bogus')
# Find nothing and receive None and a warning.
on_error = 'warn'
with pytest.warns(RuntimeWarning, match='Did not find any'):
fname = _find_matching_sidecar(bids_path, suffix='foo',
extension='.bogus', on_error=on_error)
assert fname is None
# Find nothing and receive None.
on_error = 'ignore'
fname = _find_matching_sidecar(bids_path, suffix='foo',
extension='.bogus', on_error=on_error)
assert fname is None
# Invalid on_error.
on_error = 'hello'
with pytest.raises(ValueError, match='Acceptable values for on_error are'):
_find_matching_sidecar(bids_path, suffix='coordsystem',
extension='.json', on_error=on_error)
# Test behavior of suffix and extension params when suffix and extension
# are also (not) present in the passed BIDSPath
bids_path = BIDSPath(
subject='test', task='task', datatype='eeg', root=tmp_path
)
bids_path.mkdir()
for suffix, extension in zip(
['eeg', 'eeg', 'events', 'events'],
['.fif', '.json', '.tsv', '.json']
):
bids_path.suffix = suffix
bids_path.extension = extension
bids_path.fpath.touch()
# suffix parameter should always override BIDSPath.suffix
bids_path.extension = '.json'
for bp_suffix in (None, 'eeg'):
bids_path.suffix = bp_suffix
s = _find_matching_sidecar(bids_path=bids_path, suffix='events')
assert Path(s).name == 'sub-test_task-task_events.json'
# extension parameter should always override BIDSPath.extension
bids_path.suffix = 'events'
for bp_extension in (None, '.json'):
bids_path.extension = bp_extension
s = _find_matching_sidecar(bids_path=bids_path, extension='.tsv')
assert Path(s).name == 'sub-test_task-task_events.tsv'
# If suffix and extension parameters are not passed, use BIDSPath
# attributes
bids_path.suffix = 'events'
bids_path.extension = '.tsv'
s = _find_matching_sidecar(bids_path=bids_path)
assert Path(s).name == 'sub-test_task-task_events.tsv'
# preproced meg and mri
sublist = ['{0:0>2d}'.format(i) for i in np.arange(1, 12)]
for sub in sublist:
bids_preproc_ses_dir = os.path.join(bids_preproc_dir,
'sub-{0}'.format(sub), 'ses-movie')
if os.path.exists(bids_preproc_ses_dir) is False:
os.makedirs(bids_preproc_ses_dir)
subprocess.call('mv {0}/sub-{1}/ses-movie/anat {2}/anat'.format(
preproc_mri, sub, bids_preproc_ses_dir),
shell=True)
subprocess.call('mv {0}/sub-{1}/ses-movie/meg {2}/meg'.format(
preproc_meg, sub, bids_preproc_ses_dir),
shell=True)
ses_ass_f = glob.glob(
os.path.join(preproc_meg, 'sub-{0}'.format(sub), 'ses-movie',
ses_ass_files[0]))[0]
subprocess.call('mv {0} {1}'.format(ses_ass_f, bids_preproc_ses_dir),
shell=True)
# copy study associated files
for f in mri_study_ass_files:
subprocess.call('cp {0}/{1} {2}'.format(preproc_mri, f, bids_preproc_dir),
shell=True)
# %%
tree = mne_bids.print_dir_tree(bids_dir, return_str=True)
with open('/nfs/s2/userhome/liuxingyu/workingdir/temp/gump_bids', 'w') as f:
f.write(tree)
write_anat(image=t1w_path,
bids_path=bids_path,
landmarks=mri_landmarks,
verbose=False)
# %%
# Basic anonymization
# -------------------
# Now we're ready to anonymize the dataset!
anonymize_dataset(bids_root_in=bids_root, bids_root_out=bids_root_anon)
# %%
# That's it! Let's have a look at directory structure of the anonymized
# dataset.
print_dir_tree(bids_root_anon)
# %%
# You can see that the subject ID was changed to a number (in this case, the
# digit ``1```), and the recording dates have been shifted backward in time (as
# indicated by the ``emptyroom`` session name). Anonymized IDs are zero-padded
# numbers ranging from 1 to :math:`N`, where :math:`N` is the total number of
# participants (excluding the ``emptyroom`` pseudo-subject).
#
# Limiting to specific data types
# -------------------------------
# By default, :func:`mne_bids.anonymize_dataset` will anonymize
# electrophysiological data and anatomical MR scans (T1-weighted and FLASH).
# You can limit which data types to convert using the ``datatypes`` keyword
# argument. The parameter can be a string (e.g., ``'meg'``, ``'eeg'``,
# ``'anat'``) or a list of such strings.
# First, we need some data to work with. We will use the test dataset
# available with the repository under ``data/`` directory.
#
# root of BIDs dataset
root = Path('../data/')
# BIDS entities
subject = '01'
session = 'interictalsleep'
run = '01'
datatype = 'ieeg'
###############################################################################
# show the contents of the BIDS dataset
print_dir_tree(root)
# Let's summarize the dataset.
print(make_report(root, verbose=False))
###############################################################################
# Load the dataset.
bids_path = BIDSPath(subject=subject,
session=session,
run=run,
datatype=datatype,
root=root,
suffix='ieeg',
extension='.vhdr')
raw = read_raw_bids(bids_path)
