ValueError: source space does not contain any vertices for 1 label: ['unknown-lh']

💬 Support & Discussions
1
  • MNE version: 1.2
  • operating system: Ubuntu 20.04.5 LTS

Dear MNE community,

I’m trying to perform this functional connectivity tutorial using my own data:

“Compute source space connectivity and visualize it using a circular graph”

https://mne.tools/mne-connectivity/stable/auto_examples/mne_inverse_label_connectivity.html

I’ve previously created the inverse operator using the fsaverage template.

However, when I run the function “spectral_connectivity_epochs”, I get the following error:


ValueError: source space does not contain any vertices for 1 label:

['unknown-lh']

Here is my code:


path_epochs_Tfr = '/home/(...)/epochs_OGT_Tfr_p004_A-epo.fif'
epochs_Tfr = mne.read_epochs(path_epochs_Tfr, preload=True)
fname_inv = '/home/(...)/p004_A-inv.fif'
inverse_operator = read_inverse_operator(fname_inv)
subjects_dir='/home/(...)/mne_data/MNE-fsaverage-data'
subj = 'p004_A'

#%%

# Compute the inverse solution for this data

# Returns the sources / source activity that I’ll use in computing connectivity. 

# I can specify particular frequencies to include in the connectivity with the fmin and fmax flags. 

# mne-python does:

# reads an epoch from the raw file
# applies SSP and baseline correction
# computes the inverse to obtain a source estimate
# averages the source estimate to obtain a time series for each label
# includes the label time series in the connectivity computation
# moves to the next epoch.


# Compute inverse solution and for each epoch. By using "return_generator=True"
# stcs will be a generator object instead of a list.
snr = 1.0  # use lower SNR for single epochs
lambda2 = 1.0 / snr ** 2
method = "dSPM"  # use dSPM method (could also be MNE or sLORETA)
stcs = apply_inverse_epochs(epochs_Tfr, inverse_operator, lambda2, method,
                            pick_ori="normal", return_generator=True)

# Get labels for FreeSurfer 'aparc' cortical parcellation with 34 labels/hemi
labels = mne.read_labels_from_annot(subj, parc='aparc',
                                    subjects_dir=subjects_dir)


label_colors = [label.color for label in labels]


# Average the source estimates within each label using sign-flips to reduce
# signal cancellations, also here we return a generator
src = inverse_operator['src']
label_ts = mne.extract_label_time_course(
    stcs, labels, src, mode='mean_flip', return_generator=True)#, allow_empty=True)


fmin = 8.
fmax = 13.
sfreq = epochs_Tfr.info['sfreq']  # the sampling frequency


con_methods = ['pli', 'wpli2_debiased', 'ciplv']

con = spectral_connectivity_epochs(
    label_ts, method=con_methods, mode='multitaper', sfreq=sfreq, fmin=fmin,
    fmax=fmax, faverage=True, mt_adaptive=True, n_jobs=1)

# con is a 3D array, get the connectivity for the first (and only) freq. band
# for each method
con_res = dict()
for method, c in zip(con_methods, con):
    con_res[method] = c.get_data(output='dense')[:, :, 0]

I already read the following topic, where it’s suggested to add `allow_missing=True:

If I do it as suggested, I receive the following error, when trying to make a connectivity plot:

ValueError: node_order has to be the same length as node_names

my node_order variable has length ‘70’ and the node_names variable was not created, but the variable ‘name’ has just the string ‘unknown-lh’ in it.

Do you have any hints, why this happens? How could I proceed with my analysis?

I’d be pleased to post any additional information and would greatly appreciate any help.

Best,

Bruno

2

Would be really great to get a feedback, I’m still struggling with this issue :slight_smile:

1 Like
3

What if you just drop the unknown label/s. Those aren’t important.

labels = mne.read_labels_from_annot(subj, parc=‘aparc’,
subjects_dir=subjects_dir)

#The code is something like the following
labels = [i for i in labels if ‘unknown’ not in i.name]

–Jeff

2 Likes
4

It worked greatly! Thanks a lot!!

5

Hi, can I ask how did you actually implement that? I can’t seem to get the line: labels = [i for i in labels if ‘unknown’ not in i.name] to work.

Any help would be greatly appreciated!

6

Can you ask this as its own, new question thread please? And include more info such as the full traceback (if you get an error), the minimal reproducible script to cause the error, what you’re trying to do, and mne.sys_info() output.