Summary

Research object self descriptions are intended to provide a machine readable interface to power interoperable and autonomous workflows for computational sciences. A variety of standards have been published and achieved community-level use in the fields of neuroimaging, bioinformatics, and behavioral sciences. Here we outline the standards, their metadata, and potential integration with tokenized assets in the Ocean Marketplace.

Standards

The Brain Imaging Data Structure (BIDS)

The BIDS standard was introduced to solve issues regarding data sharing and reproducibility within the neuroimaging field. All BIDS datasets must follow a specific folder hierarchy that includes grouping data by research participant, the session for the experiment, and finally the modality of the imaging protocol (i.e., 3D structural, 4D functional).

Individual files can be further distinguished by the run of acquisition, the space they are registered into, or other metadata that describes specific transformations performed on the data.

The BIDS standard is engineered to be machine-readable for containerized applications while remaining human-readable to facilitate manual quality control, manipulation, and review of datasets.

BIDS-Apps are containerized applications written to perform specific transformations on BIDS data and that conform to a standard input/output schema allowing for deployment both on local and high performance server environments.

https://bids-apps.neuroimaging.io/apps/

Some examples of BIDS-Apps include

GitHub - nipreps/mriqc: Automated Quality Control and visual reports for Quality Assessment of structural (T1w, T2w) and functional MRI of the brain

GitHub - nipreps/fmriprep: fMRIPrep is a robust and easy-to-use pipeline for preprocessing of diverse fMRI data. The transparent workflow dispenses of manual intervention, thereby ensuring the reproducibility of the results.

Experiment Factory

The experiment factory is a standard specification for psychological, psychometric and other generic behavioral experiments compiled as web applications. A key challenge with building reproducible psychology experiments are package requirements that change between environments. Experiment factory solves this issue with a template docker container that allows users to build and share experiments .