Predicting functional networks from region connectivity profiles in task-based versus resting-state fMRI data

Scripts have to be run in the following way:

download and preprocess resting fmri

1- sh shen_time_series_native_fmri_icafix.sh

download and preprocess motor task fmri

2- sh shen_time_series_native_task_icafix.sh

perform cross validation on task data

3- python cross_validation_main.py

after best model picked, train on task and predict on resting

4- python test_resting.py

generate the plots

5- python generate_plots.py

Software required to run all the scripts:

python 2, numpy, pandas, scikit-learn, keras, nilearn