Here we document standard practices we’ve developed while working with lots of messy data. The goal is to have a accurately-curated, provenance-tracked, usable data set as a basis for studies.

Table of contents

• Fetching Your Data
• Curating BIDS Datasets
• Interfacing with the clusters via Jupyter and MacFuse
• Atom on the clusters (with GitHub)
• Running Pipelines with DataLad