Home - PennBBL/conte GitHub Wiki

Welcome to the CONTE wiki!

If you're interested in learning all about CONTE fMRI I would read the wikis in the following order:

  1. Fear Conditioning Paradigm to learn about the task

  2. Model Overview to learn an overview of what analysis models were run on CONTE and what they mean

  3. FSL Analysis Pipeline to then learn how to run those models described in wiki #2 in FSL Feat

  4. XCP Pipeline to alternatively learn how to run those models described in wiki #2 using the XCP Engine

The FreeSurfer wikis detail you guessed it, FreeSurfer structural processing. The XCP Engine also details structural processing but for ANTS, gmd and jlf volume.

The ratings wiki will give you more information on the design and processing of the Behavioral task Ratings.

If you're interested in learning about Resting State EEG processing, click here for more information on preprocessing, post processing, and data extraction for the resting state EEG only.

EEG data was also collected for two tasks. For information on the two tasks, click here. However this wiki only includes information about the task set up and does not include information on processing and analysis as this was done by Anna Watters. For more information, see her.

For running a subject Incase data will be collected again, there is a file on the code page of the Wiki labeled RunningSubjectInstructions.docx which details all of the steps required for acquiring EEG, MRI.

There are also two wikis to describe these steps as well:

Running EEG Subjects
Running MRI Subjects

For information on clinical scores that were collected and where they are stored, click here to learn more about the REDCap project and variables that are included.

Storage of forms, consents, etc. are found on Saturn, and more information about that can be found here.

These wikis were written by MQ (fMRI) and PR (EEG and Clinical/REDCap) in 2017 to record information about CONTE before leaving the BBL. This is why the wikis only really focus on run1 fear conditioning, freesurfer, EEG, and behavioral ratings design, processing, and analysis. CONTE also has a reversal fear conditioning paradigm, resting BOLD imaging, and pupilometry. However, that information is not detailed in these wikis as MQ/PR did not process or analyze that data.

All CONTE aggregated data is stored in the CONTE2 project on RedCap. For MRI: the raw data files, processed data, and scripts are on both monstrum (/import/monstrum/conte_815814) and on CFN (/data/joy/BBL/studies/conte and /data/joy/BBL/projects/conteReproc2017). If you are new to CONTE and looking for more organized and finalized scripts and data you should work off of CFN rather than monstrum. This is why these wikis, although they mention some monstrum paths, focus on CFN paths and processing. Monstrum contains all that is on CFN (except for XCP Engine processing) but it is quite messy and also contains a lot of other intermediary analyses, data, and processing that may make sorting through or continuing the project more confusing. However, if you're looking for a script or analysis that used to be run to revive it, it will live on monstrum and not on CFN.

For EEG: raw data files and processed data can be found on Odin (Electrophysiology/CONTE_DATA/eeg_v3_6_2016/) and include resting state as well as task data and eye tracking data. Raw files will be in the folder labeled Raw_Files. These are divided by subject. In the Processed_Files folder, the files have now been divided into the individual tasks (AVC, RVC, OPA, OPB, eyes open resting, eyes closed resting). The data also lives on the EEG computer down on the ground floor of gates in the Conte folder on the desktop. Eye tracking data lives on the eye tracking computer. To grab eye tracking data, turn on the eye tracking computer. An error message will pop up on the screen, press F1. On the next screen that appears, click advanced mode then discard changes and exit. The next command screen will have two options: to start Windows or to start Eyelink. Select Windows to grad data (select Eyelink when recording data).

Dan Wolf and Kosha Ruparel would be the best individuals to provide information on fMRI CONTE that is missing and must be known, Karthik also handled the pupilometry analyses and so he would be a useful resource for that. Sean Gallagher and Monica Calkins handle the clinical side, and Bruce Turetsky oversees EEG.