Dataflows for Neurohackweek: Reference

Storing scientific data in an open and useful format.

This is in line with our overall goal of making science more structured and reproducible.

Automate conversion of DICOM data to BIDS

The specification is still the key source of information

Validate data before running workflows.

Nipype can be a powerful tool for writing dataflows

Nipype is verbose, but that’s for your own good

Versioning data ensures that you can track changes to data

Git annex provides a convenient way to sync data

Standardized BIDS layout means you can grab info consistently across projects

The same workflow will run on multiple BIDS datasets

Apps can be any executable tool