search

neurospin / meg_study_dynamic_scales

On the relationship between 1/f and periodic brain dynamics (alpha-band oscillations)
1 stars 0 forks source link

ENH: populate repository #1

Open dengemann opened 8 years ago

dengemann commented 8 years ago

I will clone this repo in /neurospin/meg/meg_tmp and start adding scripts for preprocessing, inverse solution and all we need. I'll then add to the library some functions for 1/f and alpha.

dengemann commented 8 years ago

Started linking files programmatically from DYNACOMP to avoid pain. MNE has certain assumptions about directory layoud, all my code has. Also it is cleaner like this I think. Left a few comments on my commit: 1b2e81013a8a62186ce0ee8f57187993a151d5d7

Another questions is where do I find the freesurfer files for these guys?

cc @philouc @virvw @darialarocca

virvw commented 8 years ago

ok cool / we do need folders on meg_tmp to be consistent and give pointers to:

so better if you stick to dynacomp

yousra made a nice clean folder check yousra/subjects/

On Thu, Nov 26, 2015 at 5:58 PM, Denis A. Engemann <notifications@github.com

wrote:

Started linking files programmatically from DYNACOMP to avoid pain. MNE has certain assumptions about directory layoud, all my code has. Also it is cleaner like this I think. Left a few comments on my commit: 1b2e810 https://github.com/neurospin/meg_study_dynamic_scales/commit/1b2e81013a8a62186ce0ee8f57187993a151d5d7

Another questions is where do I find the freesurfer files for these guys?

cc @philouc https://github.com/philouc @virvw https://github.com/virvw @darialarocca https://github.com/darialarocca

— Reply to this email directly or view it on GitHub https://github.com/neurospin/meg_study_dynamic_scales/issues/1#issuecomment-159960018 .

Virginie van Wassenhove Virginie.van.Wassenhove@gmail.com work: +33(0)1.69.08.1667 skype: virginie_vw https://sites.google.com/site/virginievanwassenhove/

dengemann commented 8 years ago

This is incompatible with git-tracking your code. But I could create a dummy folder that respects the local conventions and clone this repo inside there. For the rest linking is far better. The files in dyacomp have a very nested layout and that folder is already too big. We need a flat 1-level depth subjects folder, same for MEG.

On 26 Nov 2015, at 18:25, Virginie van Wassenhove notifications@github.com wrote:

ok cool / we do need folders on meg_tmp to be consistent and give pointers to:

  • name of the researcher
  • date of project
  • project name

so better if you stick to dynacomp

yousra made a nice clean folder check yousra/subjects/

On Thu, Nov 26, 2015 at 5:58 PM, Denis A. Engemann <notifications@github.com

wrote:

Started linking files programmatically from DYNACOMP to avoid pain. MNE has certain assumptions about directory layoud, all my code has. Also it is cleaner like this I think. Left a few comments on my commit: 1b2e810 https://github.com/neurospin/meg_study_dynamic_scales/commit/1b2e81013a8a62186ce0ee8f57187993a151d5d7

Another questions is where do I find the freesurfer files for these guys?

cc @philouc https://github.com/philouc @virvw https://github.com/virvw @darialarocca https://github.com/darialarocca

— Reply to this email directly or view it on GitHub https://github.com/neurospin/meg_study_dynamic_scales/issues/1#issuecomment-159960018 .

Virginie van Wassenhove Virginie.van.Wassenhove@gmail.com work: +33(0)1.69.08.1667 skype: virginie_vw https://sites.google.com/site/virginievanwassenhove/ — Reply to this email directly or view it on GitHub.

dengemann commented 8 years ago

Sorry for my late reply @dengemann @virvw I have the summary sheet for subject rg_110386 where Nicolas has written for the empty_raw recording "-> Dossier no_name" ; Mayber check this folder out ? in the subject directory???

for rg_110386 we would expect 130606 as far as I understand it, 130606 is not in the empty room directory. Maybe I can sort it out with one of you tomorrow at Neurosping. Now time for a pleasant hack in the Navette.

virvw commented 8 years ago

you were @nspn! were did you hide?

re. organization, I trust you master the whole thing, only pointing out to organizational needs / traceability (one of Sébastien's duty now)

re. subfolders in the subject folder, 130606 was likely the 6th of June 2013, aka the date at which acquisitions were made. (meg_acq databasing --> 1 subject may have many dates)

hopefully a pleasant navette trip - tomorrow's traffic will likely be a nightmare.

On Thu, Nov 26, 2015 at 6:41 PM, Denis A. Engemann <notifications@github.com

wrote:

The other question is still remaining. For one subject we don't seem to have an empty room recording.

for rg_110386 we would expect 130606 as far as I understand it, 130606 is not in the empty room directory. Maybe I can sort it out with one of you tomorrow at Neurosping. Now time for a pleasant hack in the Navette.

— Reply to this email directly or view it on GitHub https://github.com/neurospin/meg_study_dynamic_scales/issues/1#issuecomment-159966581 .

Virginie van Wassenhove Virginie.van.Wassenhove@gmail.com work: +33(0)1.69.08.1667 skype: virginie_vw https://sites.google.com/site/virginievanwassenhove/

virvw commented 8 years ago

yes and so it is probable that one out of 36 participants have a missing empty-room, not sure how Nicolas managed it for MF

On Thu, Nov 26, 2015 at 7:34 PM, Virginie van Wassenhove < virginie.van.wassenhove@gmail.com> wrote:

you were @nspn! were did you hide?

re. organization, I trust you master the whole thing, only pointing out to organizational needs / traceability (one of Sébastien's duty now)

re. subfolders in the subject folder, 130606 was likely the 6th of June 2013, aka the date at which acquisitions were made. (meg_acq databasing --> 1 subject may have many dates)

hopefully a pleasant navette trip - tomorrow's traffic will likely be a nightmare.

On Thu, Nov 26, 2015 at 6:41 PM, Denis A. Engemann < notifications@github.com> wrote:

The other question is still remaining. For one subject we don't seem to have an empty room recording.

for rg_110386 we would expect 130606 as far as I understand it, 130606 is not in the empty room directory. Maybe I can sort it out with one of you tomorrow at Neurosping. Now time for a pleasant hack in the Navette.

— Reply to this email directly or view it on GitHub https://github.com/neurospin/meg_study_dynamic_scales/issues/1#issuecomment-159966581 .

Virginie van Wassenhove Virginie.van.Wassenhove@gmail.com work: +33(0)1.69.08.1667 skype: virginie_vw https://sites.google.com/site/virginievanwassenhove/

Virginie van Wassenhove Virginie.van.Wassenhove@gmail.com work: +33(0)1.69.08.1667 skype: virginie_vw https://sites.google.com/site/virginievanwassenhove/

dengemann commented 8 years ago

Alright, finished the linking script and created a subfolder for my git-tracked code inside the Dynacomp. There we now have a clear mapping of things that support writing flat and boring (good!) code.

de246814@is210297:/neurospin/meg/meg_tmp/Dynacomp_Ciuciu_2011$ ls 2015_Denis/meg_study_dynamic_scales/data/MEG
aa_100234-110531  bl_110396-111025  da_110453-111124  gc_100388-110309  jh_110224-110630  ks_110142-110609  mj_130216-130529  na_110353-111006  rg_110386-130606
ak_130184-130619  cd_100449-120403  fb_110137-110512  in_110286-110719  jm_100042-120402  ld_110370-110922  mn_080208-110718  nc_110174-110421  sa_130042-130618
ap_110299-120405  cl_100240-110623  fp_110067-110301  jd_110235-130625  jm_100109-110412  ma_130185-130529  mp_110340-110908  pc_110210-110517  tl_110313-110906
bd_120417-130625  cm_110222-110809  ga_130053-130610  jh_100405-110308  kr_080082-111011  mb_110421-111206  mr_080072-130620  pe_110338-110912  vr_100551-110510
de246814@is210297:/neurospin/meg/meg_tmp/Dynacomp_Ciuciu_2011$ ls 2015_Denis/meg_study_dynamic_scales/data/subjects/
aa_100234-110531  cd_100449-120403  fp_110067-110301  jd_110235-130625  kr_080082-111011  mj_130216-130529  na_110353-111006  rg_110386-130606
ak_130184-130619  cl_100240-110623  fsaverage         jh_100405-110308  ks_110142-110609  mn_080208-110718  nc_110174-110421  sa_130042-130618
ap_110299-120405  cm_110222-110809  ga_130053-130610  jh_110224-110630  ld_110370-110922  morph-maps        ns110383          tl_110313-110906
bd_120417-130625  da_110453-111124  gc_100388-110309  jm_100042-120402  ma_130185-130529  mp_110340-110908  pc_110210-110517  vr_100551-110510
bl_110396-111025  fb_110137-110512  in_110286-110719  jm_100109-110412  mb_110421-111206  mr_080072-130620  pe_110338-110912
de246814@is210297:/neurospin/meg/meg_tmp/Dynacomp_Ciuciu_2011$ ls 2015_Denis/meg_study_dynamic_scales/data/MEG
MEG/    MEG-ER/ 
de246814@is210297:/neurospin/meg/meg_tmp/Dynacomp_Ciuciu_2011$ ls 2015_Denis/meg_study_dynamic_scales/data/MEG-ER/
aa_100234-110531  cd_100449-120403  fb_110137-110512  in_110286-110719  jm_100042-120402  ld_110370-110922  mp_110340-110908  pc_110210-110517  vr_100551-110510
ak_130184-130619  cl_100240-110623  fp_110067-110301  jd_110235-130625  jm_100109-110412  mb_110421-111206  mr_080072-130620  pe_110338-110912
ap_110299-120405  cm_110222-110809  ga_130053-130610  jh_100405-110308  kr_080082-111011  mj_130216-130529  na_110353-111006  sa_130042-130618
bl_110396-111025  da_110453-111124  gc_100388-110309  jh_110224-110630  ks_110142-110609  mn_080208-110718  nc_110174-110421  tl_110313-110906
dengemann commented 8 years ago

As explained in the readme 'data' is not tracked by git, but all code assumes that the data are there. It means if you clone this repo locally just make sure to put the links in data and you're fine. For /neurospin/meg I did it as you can see.

virvw commented 8 years ago

great - have you solved the missing empty room?

On Fri, Nov 27, 2015 at 3:22 PM, Denis A. Engemann <notifications@github.com

wrote:

As explained in the readme 'data' is not tracked by git, but all code assumes that the data are there. It means if you clone this repo locally just make sure to put the links in data and you're fine. For /neurospin/meg I did it as you can see.

— Reply to this email directly or view it on GitHub https://github.com/neurospin/meg_study_dynamic_scales/issues/1#issuecomment-160149545 .

Virginie van Wassenhove Virginie.van.Wassenhove@gmail.com work: +33(0)1.69.08.1667 skype: virginie_vw https://sites.google.com/site/virginievanwassenhove/

dengemann commented 8 years ago

great - have you solved the missing empty room?

No and it's not a blocker now, porting actual code now to do actual analyses.

virvw commented 8 years ago

ok nice - and it looks terrific on github! great w-e to you!

On Fri, Nov 27, 2015 at 5:48 PM, Denis A. Engemann <notifications@github.com

wrote:

great - have you solved the missing empty room?

No and it's not a blocker now, porting actual code now to do actual analyses.

— Reply to this email directly or view it on GitHub https://github.com/neurospin/meg_study_dynamic_scales/issues/1#issuecomment-160173738 .

Virginie van Wassenhove Virginie.van.Wassenhove@gmail.com work: +33(0)1.69.08.1667 skype: virginie_vw https://sites.google.com/site/virginievanwassenhove/

philouc commented 8 years ago

Tanks Denis for your efforts and involvement. If the file you're looking for is "empty_raw_sss.fif" for subject rg_110386 then you can get it in the subfolder Results/data/SSS/AVr/rg_110386/130606 I hope this helps. Phil

dengemann commented 8 years ago

So I'd say we have a set of ICA solutions to work with.

The info summary, the html reports and the ica.fif files can be found here:

/neurospin/meg/meg_tmp/Dynacomp_Ciuciu_2011/2015_Denis/meg_study_dynamic_scales/results/run_preproc_ica/current-2015-12-02_12-26-31

Out of the previous attempt the one that has 'current' in it is the one to use.

For convenience here is the automatically generated summary, detailing 2 misses for ECG according to minimum amplitude ranges (Magneotmeter). We discussed this already, it means there are not sufficient artifacts to model and remove them.

ica_info.xlsx