Data Types
The Genomic Regions from the Experiments and Annotations, which we will now simply refer to as regions, are the DeepBlue central data. Every experiment has a set of regions. Each region must contain a chromosome, start, and end fields. Others fields, e.g., region name, scores, strand, etc. are also possible. DeepBlue allows any number of fields to be associated with a region. The annotations, that are auxiliary data, are also formed by a set of regions in same way as Experiments.
Every experiment included in DeepBlue contains at least 6 data fields as part of the experiment metadata:
- Experiment Name
- Genome Assembly
- Epigenetic Mark
- BioSource
- Project
- Technique
These fields are organized into controlled vocabularies. Controlled vocabularies are collections of unique terms with information about them. For example, the Genome Assembly Controlled Vocabulary contains the name of all genomic assemblies that can be used in experiment metadata. Terms can be included and removed from a controlled vocabulary, but not modified.
The next sections explain how experiments and annotations are organized before going through the DeepBlue controlled vocabularies individually.