Loading File

Comments

Please sign in to contribute to the discussion.
By effigies@gmail.com - almost 3 years ago
"PhaseEncodingDirection" is missing from task-facerecognition_bold.json. This makes fieldmap correction impossible. Is this information recoverable?
By rik.henson@mrc-cbu.cam.ac.uk - almost 3 years ago
Sorry for the delay - I have updated the "task-facerecognition_bold.json" in the root directory, which now includes the "PhaseEncodingDirection" field (which should be inherited by all subjects and runs).

Note there are also many more fields (from dcm2bids), including "SliceTiming" and "EffectiveEchoSpacing" that others have asked for (I'm not sure why BIDS Validator still complains that they are not present, unless Validator does not implement this BIDS inheritance?)

By krzysztof.gorgolewski@gmail.com - over 3 years ago
This dataset seems to include an empty file 'derivatives/meg_derivatives/ct_sparse.fif' - was that intentional?
By rik.henson@mrc-cbu.cam.ac.uk - over 3 years ago
Not intentional - an error, sorry - I will upload correct file, plus several other empty placeholder files, eg stimulus files (left over from github filespace limits) - once the S3 version of OpenNeuro is ready, so I can upload them separately, because the dataset is so large that a single upload takes >10 hours and is more prone to interruption.
By rik.henson@mrc-cbu.cam.ac.uk - over 3 years ago
ct_sparse.fif and *.bmp stimulus files now uploaded in v4
By krzysztof.gorgolewski@gmail.com - almost 3 years ago
The S3 version of OpenNeuro is ready. You can update your dataset in bulk using a command line client. More details at https://github.com/OpenNeuroOrg/openneuro/tree/master/packages/openneuro-cli
By dtborek@gmail.com - over 2 years ago
Why version 1.0.3 is 460 gb but 1.0.2 weights 30 gb? Anyway when I use dtatalad it starting downloading dataset that is around 160 gb :(
By doyle.andrew@gmail.com - about 2 years ago
That's probably because datalad reuses the overlap from the 460 gb between the versions, but I can't say for sure!
By doyle.andrew@gmail.com - about 2 years ago
This is a great dataset, thanks for sharing!
By psoulos@gmail.com - over 1 year ago
Hello, I have a couple of questions about using FreeSurfer with version 1.0.3 compared to version 0.1.1 on OpenfMRI. Version 0.1.1 uses the standard FreeSurfer directory structure, and I understand how to use FreeSurfer with this format. I am unsure of how to use FreeSurfer with version 1.0.3.

In Version 0.1.1, I can run `recon-all` on sub001/anatomy/highres001.nii.gz
In Version 1.0.3, should I run `recon-all` on sub-01/ses-mri/anat/sub-01_ses-mri_acq-mprage_T1w.nii.gz?

Is there a recommended way to run `preproc-sess`, `mkanalysis-sess`, and other session level commands using the version 1.0.3 directory structure?

I also noticed that the fMRI stimuli are different between the two versions. The identity of the faces is the same between the two versions, but the actual image of each identity is different. Which image was actually shown to participants?

Thank you,
Paul
By andrewjanke@gmail.com - over 1 year ago
Hi there. I just downloaded version 1.0.3 of this dataset using the aws CLI downloader mechanism. There are couple files with weird-looking names beginning with dots and ending in an extra extension that looks like a random sequence of characters.

[~/Downloads/BIDS/ds000117-download]
$ find . -type f -name '.*.*' -exec ls -lh {} \;
-rw-r--r-- 1 janke staff 512K Aug 19 2018 ./derivatives/freesurfer/sub-01/ses-mri/anat/label/.lh.BA.thresh.annot.f3h5wZ
-rw-r--r-- 1 janke staff 239M Nov 26 2018 ./sub-02/ses-meg/meg/.sub-02_ses-meg_task-facerecognition_run-05_meg.fif.Seznkq

They correspond to not-starting-with-a-dot, not-ending-in-a-random-suffix files of similar names.

$ ls -l derivatives/freesurfer/sub-01/ses-mri/anat/label/lh.BA.thresh.annot
-rw-r--r-- 1 janke staff 1097187 Aug 19 2018 derivatives/freesurfer/sub-01/ses-mri/anat/label/lh.BA.thresh.annot
$ ls -l sub-02/ses-meg/meg/sub-02_ses-meg_task-facerecognition_run-05_meg*
-rw-r--r-- 1 janke staff 865455174 Nov 26 2018 sub-02/ses-meg/meg/sub-02_ses-meg_task-facerecognition_run-05_meg.fif

Are these temporary files of some sort that got accidentally included in the dataset? From what I've read of the BIDS specification, it doesn't define any starting-with-dots files like this.
By psoulos@gmail.com - over 1 year ago
Those look like temporary files from downloading the dataset. I would redownload those specific files just to make sure that you don't end up with a corrupted version.

By andrewjanke@gmail.com - over 1 year ago
Hi there. I just downloaded version 1.0.3 of this dataset using the aws CLI downloader mechanism. There are couple files with weird-looking names beginning with dots and ending in an extra extension that looks like a random sequence of characters.

[~/Downloads/BIDS/ds000117-download]
$ find . -type f -name '.*.*' -exec ls -lh {} \;
-rw-r--r-- 1 janke staff 512K Aug 19 2018 ./derivatives/freesurfer/sub-01/ses-mri/anat/label/.lh.BA.thresh.annot.f3h5wZ
-rw-r--r-- 1 janke staff 239M Nov 26 2018 ./sub-02/ses-meg/meg/.sub-02_ses-meg_task-facerecognition_run-05_meg.fif.Seznkq

They correspond to not-starting-with-a-dot, not-ending-in-a-random-suffix files of similar names.

$ ls -l derivatives/freesurfer/sub-01/ses-mri/anat/label/lh.BA.thresh.annot
-rw-r--r-- 1 janke staff 1097187 Aug 19 2018 derivatives/freesurfer/sub-01/ses-mri/anat/label/lh.BA.thresh.annot
$ ls -l sub-02/ses-meg/meg/sub-02_ses-meg_task-facerecognition_run-05_meg*
-rw-r--r-- 1 janke staff 865455174 Nov 26 2018 sub-02/ses-meg/meg/sub-02_ses-meg_task-facerecognition_run-05_meg.fif

Are these temporary files of some sort that got accidentally included in the dataset? From what I've read of the BIDS specification, it doesn't define any starting-with-dots files like this.
By neurolang@g.ncu.edu.tw - about 1 year ago
Hi, thank you very much for sharing this dataset!

I have a question about the MEG data (*.fif files). Are the FIF files (including *fif files of each participant and *fif files of empty room data) processed with MaxFilter? Thanks!

All the best,
Kevin Hsu
By neurolang@g.ncu.edu.tw - about 1 year ago
I am very sorry for posting the same message several times. I did not see the post of my comment so I sent this comment for several times... Now I know that the latest message is posted at the bottom of the page. I am very sorry for my mistake!

Kevin Hsu
By rik.henson@mrc-cbu.cam.ac.uk - about 1 year ago
Hi Kevin. The MEG data in the main folders are raw, ie have not been maxfiltered. Maxfiltered versions are in the "derivatives" sub-folder. Best wishes, Rik.
By neurolang@g.ncu.edu.tw - about 1 year ago
Hi Rik, Thank you for explaining the files. I will go for the data in the derivatives sub-folder. All the best, Kevin

By fatemeh.391@gmail.com - about 1 year ago
Many thanks for sharing the dataset. Is it possible please to tell me is there EEG data in this dataset too?
By rik.henson@mrc-cbu.cam.ac.uk - about 1 year ago
Yes, 70 channels of nose-referenced EEG data, recorded simultaneously with the MEG, are part of the channels in the FIF file. Various freeware (eg FieldTrip) can extract them from the FIF files into, eg, Matlab.
By moritz.maschke@merkur.uni-freiburg.de - 3 months ago
I just got a reply from Prof Henson that I'll share: "EEG061 and EEG062 are H+V EOG. EEG063 is ECG. EEG064 is floating noise, ie nothing."
I hope that this is helpful to other users. The noise channel was not mentioned in the paper, which I recommend as a good read. :)
By agathe.senellart@gmail.com - 3 months ago
Hi!
I tried downloading a single file
sub-01_ses-meg_meg_sub-01_ses-meg_task-facerecognition_run-01_meg.fif
But it doesn't seem to work today although I did it a month ago and it worked fine. The file only contains an error message :
{"error": "file not found"}
Do you know what might be happening here ?
Thank you for your help !