jrasero / predicting-icns

Results and plots replications for paper in prediction of Intrinsic Connectivity Networks in task versus resting fMRI

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

Description:

Code written in python and used for downloading and preprocessing the data and generating results and plots of the work:

Predicting functional networks from region connectivity profiles in task-based versus resting-state fMRI data
Javier Rasero, Hannelore Aerts, Marlis Ontivero Ortega, Jesus M. Cortes, Sebastiano Stramaglia, Daniele Marinazzo
bioRxiv 259077, that can be found in the link: https://doi.org/10.1371/journal.pone.0207385

Software Requirements:

python 2.7, numpy, pandas, scikit-learn, keras 2.0, nilearn, fsl and FIX

Usage:

Scripts have to be run in the following order (steps 1, 2 and 3 are to be run in a cluster given the time of computation):

1. Download and preprocess resting fmri: sh shen_time_series_native_fmri_icafix.sh
2. Download and preprocess motor task fmri: sh shen_time_series_native_task_icafix.sh
3. Perform cross validation on task data: python cross_validation_main.py
4. Train on task and predict on resting after best model selected from previous step: python test_resting.py
5. Generate the plots (Figure 1 of the paper was included using Libreoffice Impress) python generate_main_plots.py and python generate_suppl_plots.py

Please do not hesitate to contact us for any issue running the code, suggestions and remarks to [email protected]