def save_to_hdf5(power: AverageTFR):
if not os.path.exists(power.comment.parent):
os.makedirs(power.comment.parent)
power.comment = str(power.comment)
# replace floats in file name with integers
floats = re.findall("[-+]?\d*\.\d+", power.comment)
if floats:
for num in floats:
power.comment = power.comment.replace(num, str(int(float(num))))
file_path_with_extension = f"{power.comment}_power-tfr.h5"
logger.info(f"Saving power at {file_path_with_extension} ...")
power.save(fname=str(file_path_with_extension), overwrite=True)
logger.info("[FINISHED]")
def average_power_into_frequency_bands(power: AverageTFR,
config: dict) -> AverageTFR:
"""
Computes average power in specific frequency ranges defined in config
Parameters
----------
power
config
Returns
-------
"""
band_values = config.values()
band_power_data_arr = []
for band in band_values:
band_filter = np.logical_and(power.freqs >= float(band[0]),
power.freqs < float(band[1]))
if int(max(power.freqs)) == int(max(band_values)[1]):
band_filter[-1] = True
band_data = power.data[:, band_filter, :].mean(axis=1)
band_power_data_arr.append(band_data[np.newaxis, :])
band_power_data = np.concatenate(band_power_data_arr, axis=0)
band_power_data = np.transpose(band_power_data, (1, 0, 2))
band_freqs = np.asarray([band[0] for band in band_values])
band_power = power.copy()
band_power.data = band_power_data
band_power.freqs = band_freqs
band_power.comment = power.comment + "_band_average"
return band_power
# Roll covariance, csp and lda over time
for t, w_time in enumerate(centered_w_times):
# Center the min and max of the window
w_tmin = w_time - w_size / 2.
w_tmax = w_time + w_size / 2.
# Crop data into time-window of interest
X = epochs.copy().crop(w_tmin, w_tmax).get_data()
# Save mean scores over folds for each frequency and time window
scores[freq, t] = np.mean(cross_val_score(estimator=clf,
X=X,
y=y,
cv=cv,
n_jobs=1),
axis=0)
###############################################################################
# Plot results
# Set up time frequency object
av_tfr = AverageTFR(create_info(['freq'], sfreq), scores[np.newaxis, :],
centered_w_times, freqs[1:], 1)
chance = np.mean(y) # set chance level to white in the plot
av_tfr.plot([0],
vmin=chance,
title="Time-Frequency Decoding Scores",
cmap=plt.cm.Reds)
# Define wavelet frequencies and number of cycles
cwt_freqs = np.arange(7, 30, 2)
cwt_n_cycles = cwt_freqs / 7.
# Run the connectivity analysis using 2 parallel jobs
sfreq = raw.info['sfreq'] # the sampling frequency
con, freqs, times, _, _ = spectral_connectivity(epochs,
indices=indices,
method='wpli2_debiased',
mode='cwt_morlet',
sfreq=sfreq,
cwt_freqs=cwt_freqs,
cwt_n_cycles=cwt_n_cycles,
n_jobs=1)
# Mark the seed channel with a value of 1.0, so we can see it in the plot
con[np.where(indices[1] == seed)] = 1.0
# Show topography of connectivity from seed
title = 'WPLI2 - Visual - Seed %s' % seed_ch
layout = mne.find_layout(epochs.info, 'meg') # use full layout
tfr = AverageTFR(epochs.info, con, times, freqs, len(epochs))
tfr.plot_topo(fig_facecolor='w', font_color='k', border='k')
# %%
# References
# ----------
# .. footbibliography::
# Use 'MEG 2343' as seed
seed_ch = 'MEG 2343'
picks_ch_names = [raw.ch_names[i] for i in picks]
# Create seed-target indices for connectivity computation
seed = picks_ch_names.index(seed_ch)
targets = np.arange(len(picks))
indices = seed_target_indices(seed, targets)
# Define wavelet frequencies and number of cycles
cwt_freqs = np.arange(7, 30, 2)
cwt_n_cycles = cwt_freqs / 7.
# Run the connectivity analysis using 2 parallel jobs
sfreq = raw.info['sfreq'] # the sampling frequency
con, freqs, times, _, _ = spectral_connectivity(
epochs, indices=indices,
method='wpli2_debiased', mode='cwt_morlet', sfreq=sfreq,
cwt_freqs=cwt_freqs, cwt_n_cycles=cwt_n_cycles, n_jobs=1)
# Mark the seed channel with a value of 1.0, so we can see it in the plot
con[np.where(indices[1] == seed)] = 1.0
# Show topography of connectivity from seed
title = 'WPLI2 - Visual - Seed %s' % seed_ch
layout = mne.find_layout(epochs.info, 'meg') # use full layout
tfr = AverageTFR(epochs.info, con, times, freqs, len(epochs))
tfr.plot_topo(fig_facecolor='w', font_color='k', border='k')
epochs = Epochs(raw_filter, events, event_id, tmin - w_size, tmax + w_size,
proj=False, baseline=None, preload=True)
epochs.drop_bad()
y = le.fit_transform(epochs.events[:, 2])
# Roll covariance, csp and lda over time
for t, w_time in enumerate(centered_w_times):
# Center the min and max of the window
w_tmin = w_time - w_size / 2.
w_tmax = w_time + w_size / 2.
# Crop data into time-window of interest
X = epochs.copy().crop(w_tmin, w_tmax).get_data()
# Save mean scores over folds for each frequency and time window
tf_scores[freq, t] = np.mean(cross_val_score(estimator=clf, X=X, y=y,
scoring='roc_auc', cv=cv,
n_jobs=1), axis=0)
###############################################################################
# Plot time-frequency results
# Set up time frequency object
av_tfr = AverageTFR(create_info(['freq'], sfreq), tf_scores[np.newaxis, :],
centered_w_times, freqs[1:], 1)
chance = np.mean(y) # set chance level to white in the plot
av_tfr.plot([0], vmin=chance, title="Time-Frequency Decoding Scores",
cmap=plt.cm.Reds)
# Connectivity
for ix_c, c in enumerate([exp_sup_lon, exp_sup_sho]):
c.info['subject_info'] = 'P14'
c.info['cond'] = conds[ix_c]
calc_wpli_over_time(c)
dats = list()
for c in conds:
dat = np.load(
op.join(study_path, 'results', 'wpli', 'P14_{}_wpli.npz'.format(c)))
dats.append(dat)
info = dat['info'].item()
for ix_c, c in enumerate(conds):
tfr = AverageTFR(info, dats[ix_c]['con'][1, :, :, :], dat['times'],
dat['freqs'], 1)
tfr.plot(picks=[83], vmin=-1, vmax=1, cmap='viridis', title=c)
# TF ROI
if ref == 'avg':
rois = {
'HP_r': np.array(raw.ch_names)[picks[:4]],
'HP_l': np.array(raw.ch_names)[picks[4:]]
} # :4
elif ref == 'bip':
rois = {
'HP_l': np.array(raw.ch_names)[picks[:2]],
'HP_r': np.array(raw.ch_names)[picks[2:]]
}
roi_fig, axes = plt.subplots(len(rois), len(conds), sharey=True, sharex=True)
def load_patient_tfr(deriv_path, subject, band, task=None, verbose=True):
from mne.time_frequency import read_tfrs, AverageTFR
if task is not None:
search_str = f"*sub-{subject}_*task-{task}*.h5"
else:
search_str = f"*sub-{subject}_*.h5"
deriv_files = [f.as_posix() for f in Path(deriv_path).rglob(search_str)]
if band == "delta":
fmin, fmax = 0.5, 5
elif band == "theta":
fmin, fmax = 5, 10
elif band == "alpha":
fmin, fmax = 10, 16
elif band == "beta":
fmin, fmax = 16, 30
elif band == "gamma":
fmin, fmax = 30, 90
elif band == "highgamma":
fmin, fmax = 90, 300
else:
raise ValueError("kwarg 'band' can only be prespecified set of values")
# print(deriv_files)
patient_tfrs = []
for deriv_fpath in deriv_files:
# print(f"Loading {deriv_fpath}")
avg_tfr = read_tfrs(deriv_fpath)[0]
# print(avg_tfr.freqs)
# only obtain the Band TFR for that subject
freq_inds = np.where((avg_tfr.freqs >= fmin) & (avg_tfr.freqs < fmax))[0]
# print(freq_inds)
band_data = np.mean(avg_tfr.data[:, freq_inds, :], axis=1, keepdims=True)
band_tfr = AverageTFR(
avg_tfr.info,
data=band_data,
freqs=[fmin],
nave=1,
times=avg_tfr.times,
verbose=0,
)
json_fpath = deriv_fpath.replace(".h5", ".json")
with open(json_fpath, "r") as fin:
sidecar_json = json.load(fin)
# obtain the event IDs for the markers of interest
sz_onset_id = sidecar_json.get("sz_onset_event_id", None)
sz_offset_id = sidecar_json.get("sz_offset_event_id")
clin_onset_id = sidecar_json.get("clin_onset_event_id")
# events array
events = sidecar_json["events"]
# obtain onset/offset event markers
sz_onset_win = _get_onset_event_id(events, sz_onset_id)
sz_offset_win = _get_onset_event_id(events, sz_offset_id)
clin_onset_win = _get_onset_event_id(events, clin_onset_id)
# set those windows
sidecar_json["sz_onset_win"] = sz_onset_win
sidecar_json["sz_offset_win"] = sz_offset_win
sidecar_json["clin_onset_win"] = clin_onset_win
sidecar_json["freq_band"] = (fmin, fmax)
# create a Result object
band_tfr = Result(
Normalize.compute_fragilitymetric(band_data.squeeze(), invert=True),
info=avg_tfr.info,
metadata=sidecar_json,
)
# band_tfr.metadata.save(json_fpath)
if np.isnan(band_data).any():
print(f"Skipping {deriv_fpath} due to nans")
continue
if sz_onset_win is None:
print(f"Skipping {deriv_fpath}")
continue
patient_tfrs.append(band_tfr)
return patient_tfrs
def wpli_analysis_time(subjects, log):
conds = ['lon', 'sho']
roi_cons = {}
roi = rois['f']
spatial_con = mne.channels.read_ch_connectivity('biosemi64')
for ix_s, s in enumerate(subjects):
print('subject {} of {}' .format(ix_s+1, len(subjects)))
for ix_c, c in enumerate(conds):
filename = op.join(study_path, 'results', 'wpli', 'over_time', '{}_{}_wpli.npz' .format(s, c))
dat = np.load(filename)
info = dat['info'].item()
# Create results matrices
if ix_s == 0:
roi_cons[c] = np.empty((len(subjects), dat['con'].shape[0], dat['con'].shape[-2], dat['con'].shape[-1]))
# Get ROI connectivity
# roi_ixs = [ix for ix, ch in enumerate(info['ch_names']) if ch in roi]
# roi_con = np.mean(dat['con'][roi_ixs, :, :, :], axis=0)
roi_ixs = 33
roi_con = dat['con'][roi_ixs, :, :, :]
roi_con[roi_ixs, :, :] = 1.0
roi_cons[c][ix_s, :, :, :] = roi_con.copy()
avg_con = [np.mean(roi_cons[c], axis=0) for c in conds]
for ix_c, c in enumerate(conds):
tfr = AverageTFR(info, avg_con[ix_c], dat['times'], dat['freqs'], len(subjects))
tfr.plot_topo(fig_facecolor='w', font_color='k', border='k', vmin=0, vmax=0.5, cmap='viridis', title=c)
s = 12
for c in conds:
tfr = AverageTFR(info, roi_cons[c][s, :, :, :], dat['times'], dat['freqs'], len(subjects))
tfr.plot_topo(fig_facecolor='w', font_color='k', border='k', vmin=0, vmax=1, cmap='viridis', title=c)
# Stats
test_con = [roi_cons[c][:, 19, :, :] for c in conds]
#threshold = None
threshold = dict(start=0, step=0.2)
T_obs, clusters, cluster_p_values, H0 = \
permutation_cluster_test([test_con[0], test_con[1]],
n_permutations=1000, threshold=threshold, tail=0)
times = dat['times']
times *= 1e3
freqs = dat['freqs']
fig, ax = plt.subplots(1)
T_obs_plot = np.nan * np.ones_like(T_obs)
for c, p_val in zip(clusters, cluster_p_values):
if p_val <= 0.05:
T_obs_plot[c] = T_obs[c]
ax.imshow(T_obs,
extent=[times[0], times[-1], freqs[0], freqs[-1]],
aspect='auto', origin='lower', cmap='gray')
ax.imshow(T_obs_plot,
extent=[times[0], times[-1], freqs[0], freqs[-1]],
aspect='auto', origin='lower', cmap='RdBu_r')
plt.xlabel('Time (ms)')
plt.ylabel('Frequency (Hz)')
plt.title('ROI Connectivity')