This is the public repository for code related to the Big Data for Discovery Science effort http://bd2k.ini.usc.edu/. The current BDDS software tools and services are available here.
The PBDA_PD folder contains the complete Predictive Big Data Analytics protocol (Pipeline workflow for obtaining a vector of hundreds of derived neuroimaging biomarkers and an R-script for machine-learning diagnostic classification and forecasting) that can be applied for Parkinson's Disease studies or any other clinical or translational biomedical investigation involving heterogeneous datasets.