Data sets or collections are often composed of multiple files that are related. Files may have come from (or still be stored in) a relational database, and the relationships among the data tables or other entities are important if the data are to be reused. These relationships should be documented for a repository.
Describe the overall organization of your data set or collection. Often, a data set or collection contains a large number of files, perhaps organized into a number of directories or database tables. By describing and documenting this organization, files and data can be easily located and used.
At a minimum, the organization and relationships between the directories and files, or database tables and other supporting materials, need to be fully described. Use a description of the data set or collection (e.g, an abstract) to describe what tables contain, where the supporting material, metadata, or other documentation are located, and/or descriptions of directory contents. Consider describing the logical relationships between data entities using an entity relationship diagram (ERD).
Associated specimens: if specimens (e.g., taxonomic vouchers, DNA samples) were collected with the data, include the name of the repository in which these specimens reside.
Relationships among data entities should be described documented to enable understanding by future users and repositories.
Specimen repositories: http://www.biorepositories.org/
Describing data table constraints with Ecological Metadata Language (EML): http://knb.ecoinformatics.org/software/eml/eml-2.1.0/eml-constraint.html