Core principles for the implementation of the neurodata without borders data standard

Background: The Neurodata Without Borders data standard (NWB) unifies diverse modalities of neurophysiology data in a single format. Integrating NWB with a database unleashes its full potential to promote collaboration, standardize analyses, capitalize on historical data, and ensures data integrity by maintaining process transparency. NWB database technology is the bedrock of analytical systems used by academic leaders including the Allen Institute and the International Brain Laboratory. Here we present the benefits of incorporating NWB design principles in a big data analytics application.

New method: Data standards and databases are the foundation of big data analytics. To demonstrate the benefits of using these systems together, we implemented NWB in Jupyter notebooks using DataJoint to streamline database operations.

Results: We demonstrate the utility of combining the NWB with DataJoint in a Jupyter-based electronic lab journal. We convert open-field behavioral data (using X, Y coordinates) to NWB format and process it with a DataJoint pipeline. Additional notebooks demonstrate working NWB files, data sharing, combining data from diverse sources, and retrospective analyses with data query filtering techniques.

Comparison with existing methods: NWB describes how to structure and store neurophysiology data and is streamlined for research settings. In contrast to other data standards, combining NWB with DataJoint's database interface can dramatically increase data analytical capabilities.

Conclusions: The joint use of NWB with DataJoint transforms traditional laboratory datasets and workflows. Our Jupyter notebooks showcase the analytical and collaborative advantages of adopting big data analytics and can be tailored to other modalities by researchers interested in evaluating NWB.