This project implements entities maintained by the overview builder. Each of the
centres submit a large amount of data. These are submitted as XML documents
using a data format that makes standardisation and data validation easier. When
the PhenoDCC receives these XML documents, they are stored verbatim in the
phenodcc_raw
database. However, for the most frequent queries that are used by
the annotation generator and data display, it is faster to prepare overview
records set that are queried frequently. The phenodcc_overviews
database stores
the collection of data that are queried most frequently.
The database schema is defined elsewhere in the overview builder. In this project we only define entities that are used by Phenoview and the QC tool.