Home

Welcome openMINDS, the open Metadata Initiative for Neuroscience Data Structures powered by the HBP (Human Brain Project) and EBRAINS (European Brain ReseArch INfraStructure).

openMINDS is a collection of ontology-based metadata schemas for neuroscience data structures, developed and maintained by the curation team of the HBP Neuroinformatics Platform (NIP). Currently supported data structures are data generated in experimental laboratories (across all neuroscience modalities) as well as the code of computational models and software.

The schemas are the architectural building blocks of the Knowledge Graph (KG) database, a unique data sharing framework powered by HBP and hosted on EBRAINS. As a metadata standard, openMINDS is flexible enough to capture the specific aspects of the data structures, yet strict enough to guarantee comparability across them to facilitate queries via the KG Search UI. Moreover, wherever it is applicable, openMINDS metadata (properties and values) are linked to neuroscience terminologies or ontologies that are also used outside of the KG database.

The openMINDS schemas can be used to describe the origin, context, content, and physical location of individual or entire sets of data files in a modular fashion. In general, each schema captures ...

• ... the information context of the contained metadata (e.g, subject),
• ... a set of required or optional metadata (e.g., for subject: species, biological sex, etc.),
• ... the expected value type of each metadata entry (e.g., reference to another schema, date, etc.),
• ... and in some cases even a drop-down list of possible values for a specific metadata entry.

All openMINDS schemas are defined in JSON-Schema and can be serialized in JSON-LD. The latter can also be directly digested as machine-readable data descriptions into the database of the KG. More information about the technical implementation of the openMINDS schemas and their usage can be found in Implementation & Usage.

Using openMINDS metadata schemas for describing data will increase their findability and interoperability according to the FAIR guiding principles for scientific data management and stewardship (Wilkinson et al. 2016).