The data model

The data model for the (scientific) metadata for each record consist of two parts:

1. General parameters
2. Method specific parameters

Overview of General parameters

General parameters describe both the metadata of the record (e.g. authors), as well biophysical properties expected to be present for all types of biophysical measurements (e.g. identities of measured species).

All records have the same fields in the general parameters section Independent of measurement technique, which means that the required fields in general parameters are present (and hence searchable) in all records.

More information about the general parameters can be found here

Overview of method specific parameters

The method specific parameters include the detailed information about how data was measured and how it was analyzed (e.g. measurement protocol). As this is inherently specific to the techniques used, the fields in method specific parameters should only be expected to exist in records using the same technique.

More information about the data model of each of the supported techniques can be found here:

• MST/TRIC/Spectral shift
• Biolayer interferometry (BLI)
• Surface plasmon resonance (SPR)
• Isothermal Titration Calorimetry (ITC)

Structure of the model

Reusable elements

The model makes heavy use of the programming technique composition. This means that certain elements are reused as often as possible, which has the advantages that you can find the same elements in many places. This makes comparisons easier as well as making it easier to gain intuition about which fields are available. A good example of this is "concentration", whenever you see this field, its subfields will allways be the same. It also has the advantage that errors only have to be corrected in one place. The disadvantage is that it increases nesting at makes the schema less flexible/specific. In some parts of the data models the shortcomings of composition outweigh their advantages.

Polymorphic elements

Some of the shortcomings of composition can favour use of the programming technique inheritance. Occasionally this technique is used in the data model, however as this is not trivial to implement and makes it harder to know which elements are available the usage is kept to minimum. One of the places it is used is to define the different types of publications. This has the advantage of making type-specific fields available for each of the publication types.

Files

The metadata of the files themselves (rather than the data they describe) are foundoutside the two part model described above in a separate part of the record see here.

Vocabularies

In certain places (like organisms) we rely on controlled vocabularies. You can think of these like multi-field preset options. They are described in detail here.