Data model

by Kees van Bochove last modified Oct 08, 2010 06:07 PM
This page describes the basic technical architecture of DbNP

Overview

The core of dbNP is formed by a module called GSCF, which stands for Generic Study Capture Framework. It is a web application that is designed to store the design of the biological studies in the database. This information is small in terms of disk size, but very valuable. GSCF serves as a linking hub for the other dbNP modules: it links to the data in the various assay modules. The data in these assay modules tends to be large in terms of disk space, but generally has a low information density.

 

Data model

Overview of the DbNP structure: the study design model in GSCF and the links to omics modules, ontologies etc.

 

The data in GSCF, the study capture module, forms the top layer of this graph. In GSCF, studies are defined. On a study level, meta-information about the study is defined, such as the people involved, study code, description, aim, associated publications etc. Also, the subjects (humans, mice, plants, cell cultures) of the study are defined. After that, the events are indicated that take place in the study. These can be treatments (such as medicine treatment), nutritional challenges, but also e.g. adverse effects can be described with an event. (See templates below on how to customize event descriptions for a particular area of biology). Also, a special type of event, called a 'sampling event', results in associated samples. For example, a sampling event 'Blood sampling' could lead to a blood sample for each subject on which this sampling event is applied. Events can be grouped in so-called 'event groups', which couple a group of subjects to a specific group of events. For example, a medicine intervention study in which one half of the subjects gets the actual medicine and the other half gets a placebo, could be described by two event groups, one with an associated medicine intervention event with the substance set to the actual medicine, and an event group with a medicine intervention event of which the substance is set to placebo. Finally, the samples are normally generated from the specified sampling events, but these can be customized if necessary, and then added to one or more assays, which reside in assay modules.

 

GSCF Data Model Overview

 

The following picture gives an overview of the main GSCF data entities. Most of them extend TemplateEntity, allowing them to store multiple custom field types. These field types form 'templates', which are editable from within the user interface using the 'template editor'.

 

GSCF data model overview

 

Screenshot of the template editor:

GSCF screenshot 2 - template editor for study properties

 

Also, a study links to a number of other entities, such a persons and publications. Those are highlighted in the image below. Also, here is shown how a particular study has a particular template which contains the custom fields.

GSCF data model study