gpiantoni / baseproj Goto Github PK

Template script to start a complete analysis on EEG/MEG data, using Fieldtrip.

The analysis can run on:
1) event-based data (using the eventbased repo)
2) MRI data to create personal leadfields (using the mri2lead repo)
3) sleep to detect spindles and slow waves (using the detectsleep repo)

To start a new complete analysis, called "NICKNAME", you should go into the script folder (/data1/projects/PROJNAME/scripts/):
  git clone git://github.com/gpiantoni/baseproj.git
Then rename baseproj into NICKNAME. In the folder NICKNAME, you'll find a file called "nickname.m", which you should rename to your NICKNAME.m, and a file called "info_nickname.m", which you should rename to your info_NICKNAME.m. 

Depending on what you want to do, you need to get the other tools for the potential analysis.

1) If you want to do event-based analysis, go into your NICKNAME folder and type in command line:
  git clone git://github.com/gpiantoni/eventbased.git
The artifact-cleaning of the data depends on German's toolbox, called eegcore. To get the repository, you need reading access, through bitbucket. Then, on the command-line:
  hg clone https://username:[email protected]/gomezher/eegcore
Furthermore, two functions of eventbased requires custom-made functions: (a) seldata requires a trialfun_XXX to read the data in one big trial and create events (b) redef requires a event2trl_XXX to create smaller, event-based trials. 
These two functions and other functions you may want to add should go into a directory called NICKNAME_private. I strongly recommend to use a version control (git) on that folder as well.

2) If you want to create subject-specific forward model based on the MRI, go into your NICKNAME folder and type in command line:
  git clone git://github.com/gpiantoni/mri2lead.git

3) If you want to detect slow waves and spindles, go into your NICKNAME folder and type in command line:
  git clone git://github.com/gpiantoni/detectsleep.git

Then you can start modifying the NICKNAME.m and info_NICKNAME.m for your analysis