System Design¶

Vision¶

It is very difficult to track the output and impact of the facilities at SciLifeLab. The Data Tracker is intended as an attempt to improve this by providing all generated data with unique identifiers that can be used to refer to the generated data.

The Data Tracker is designed based on the concept of three entities orders, datasets, and collections.

An order is simply an order to a facility. It may be a request for a microscope image, whole-genome sequencing of multiple samples, protein quantification, or any other service provided by a facility. Each order is expected to generate one or more datasets. Each dataset may refer to one or more analysed samples. All datasets generated from the same order are considered to be related. Datasets can be grouped together using collections, e.g. to show that orders from different facilities all are connected to the same research project.

Orders are intended to be private, while all information about datasets and collections is public.

All three entities will get one or more unique identifier, with the option of generating DOIs for specific datasets and collections.

Implementation¶

The Data Tracker is implemented as a backend in Flask. All interactions with the system are intended to be done via the API, but a frontend (written in Quasar) is also available to simplify visualisation and usage.

The data structure is listed on the Data Structure page.

{
  "DATA_EDIT": ("DATA_EDIT", "USER_ADD", "USER_SEARCH"),
  "OWNERS_READ": ("OWNERS_READ",),
  "USER_ADD": ("USER_ADD",),
  "USER_SEARCH": ("USER_SEARCH",),
  "USER_MANAGEMENT": ("USER_MANAGEMENT", "USER_ADD", "USER_SEARCH"),
  "DATA_MANAGEMENT": ("DATA_EDIT", "OWNERS_READ", "DATA_MANAGEMENT"),
 }