Research products

openMINDS supports the representation of various research products (e.g., datasets from human, animal or simulation studies, computational models, software tools). All research products are versioned, have a common set of bibliographic properties, and (typically) link to a file repository pointing to the stored data (cf. next chapter).

Versioning of research products

openMINDS distinguishes between a conceptual (version-independent) research product and it’s respective research product version(s). Both entities, research product and research product version, can be assigned with a unique, persistent digital identifier (e.g., a digital object identifier; DOI) which enables the citation of the overall concept (referring to all versions at once) or the explicit citation of a specific version.

A research product is composed of properties that are typically valid across all versions (e.g., title, abstract, homepage) and has to link to at least one research product version.

A research product version can inherit or overwrite all version-independent properties. Additionally, it includes properties that are (usually) version-specific (e.g., studied specimen, used technique).

Note

Although inheritance and overwrite rules for version-independent properties are defined by openMINDS schemas, their enforcment has to be either performed manually or defined in an automated validation workflow.

If a research product is composed of multiple versions, these versions can refer to each other as updates or alternatives:

If the usage of version B should be preferred over version A, version B is (the) new version of version A. For example, a software version could be preferred over another, because it includes more application features, or a dataset version could be preferred over another, because it studied a larger cohort.

If the usage of version A and B depends on the users preference, the versions should reference each other as alternative version of. For example, two model versions describe the same computational model, but using different model description languages, or two dataset versions could present the same data content, but using different data formats.

Grouping of research products

In addition to versioning of research products, openMINDS supports the grouping of research products and / or research product versions into a joint project.

For example, a project could group multiple research products (with automatically all respective versions) of the same type (e.g., datasets), because they were collected as part of one research study. Alternatively, a project could group multiple research products or research product versions of different types (e.g., dataset, model, software, and a workflow recipe), because their data (including code) depend on each other.