mnarayan / cpac-utils

Utility functions to fetch and organize preprocessed fMRI timeseries

cpac-utils

Utility functions to fetch and organize time-series derivatives after data has been preprocessed using C-PAC

Requirements

Uses python 3 as well as the following python packages

• awscli
• numpy
• scipy
• pandas

pip install -r requirements.txt

Usage

Download Data

Preprocessed data is expected to be located at in data/cpac

|--data
  |--cpac
    |--derivatives
|--src
|--requirements.txt
|--README

For example data preprocessed at OpenNeuro.org can be downloaded as follows

./src/download_data.sh

Fetch and Organize ROI Time-Series

Given a textfile path_roi_timeseries_mask_<ATLASNAME>.txt containing the list of paths to all outputs from C-PAC such as roi_stats.csv, the function collect_timeseries.py will load, clean and reorganize into .mat and .pickle data structures by participant and task.

python3 collect_timeseries.py path_roi_timeseries_mask_$ATLASNAME.txt

The example script src/organize_timeseries.sh creates a textfile of pathnames and then calls the above command.

After downloading and organizing the timeseries, the outputs look as follows:

data/cpac/derivatives/pipeline_analysis/compcor_ncomponents_5_selector_pc10.linear1.wm0.global0.motion1.quadratic1.gm0.compcor1.csf1/sub-01_ses-post:
sub-01_ses-post_task-rest_run-01_CC200.mat     sub-01_ses-post_task-rest_run-02_CC200.pickle
sub-01_ses-post_task-rest_run-01_CC200.pickle  sub-01_ses-post_task-rest_run-03_CC200.mat
sub-01_ses-post_task-rest_run-02_CC200.mat     sub-01_ses-post_task-rest_run-03_CC200.pickle

data/cpac/derivatives/pipeline_analysis/compcor_ncomponents_5_selector_pc10.linear1.wm0.global0.motion1.quadratic1.gm0.compcor1.csf1/sub-01_ses-pre:
sub-01_ses-pre_task-rest_run-01_CC200.mat     sub-01_ses-pre_task-rest_run-02_CC200.pickle
sub-01_ses-pre_task-rest_run-01_CC200.pickle  sub-01_ses-pre_task-rest_run-03_CC200.mat
sub-01_ses-pre_task-rest_run-02_CC200.mat     sub-01_ses-pre_task-rest_run-03_CC200.pickle