BioGraph: a web application and a graph database for querying and analyzing bioinformatics resources

Abstract

Background: Several online databases provide a large amount of biomedical data of different biological entities. These resources are typically stored in systems implementing their own data model, user interface and query language. On the other hand, in many bioinformatics scenarios there is often the need to use more than one resource. The availability of a single bioinformatics platform that integrates many biological resources and services is, for those reasons a fundamental issue.

Description: Here, we present BioGraph, a web application that allows to query, visualize and analyze biological data belonging to several online available sources. BioGraph is built upon our previously developed graph database called BioGraphDB, that integrates and stores heterogeneous biological resources and make them available by means of a common structure and a unique query language. BioGraph implements state-of-the-art technologies and provides pre-compiled bioinformatics scenarios, as well as the possibility to perform custom queries and obtaining an interactive and dynamic visualization of results.

Conclusion: We present a case study about functional analysis of microRNA in breast cancer in order to demonstrate the functionalities of the system. BioGraph is freely available at http://biograph.pa.icar.cnr.it . Source files are available on GitHub at https://github.com/IcarPA-TBlab/BioGraph.

Keywords: BioGraphDB; Bioinformatics databases; Graph databases; Integrated databases; miRNA.

Fig. 1

BioGraphDB scheme. The graph data model of BioGraphDB consists of a set of nodes/vertices classes matching the biological entities provided by the used data sources. Relationships between entities are modeled as relations, or edges

Fig. 2

BioGraph architectural stack. The pictures gives an overview on the state-of-the-art technology behind BioGraph. Used tools are grouped into three levels. From the bottom to the top: the Graph Data level overlooked by Apache Tinkerpop, the Microservices level, and the Web Application level. Protocols and data formats of inter-level communications are also highlighted

Fig. 3

The Templates tab. Templates are simple predefined queries given as examples of how an user can traverse BioGraphDB. The queries are customizable and grouped by category

Fig. 4

The Scenarios tab. The four proposed scenarios are examples of how complex Gremlin queries can help in the analysis of non-trivial bioinformatics problems

Fig. 5

The Gremlin Workbench. It is basically composed of four panes: 1) the Gremlin query pane, 2) Tree View, 3) Graph view-port, 4) the Details pane

Fig. 6

Case study scenario. The functional analysis of miRNAs in breast cancer can be done starting from the first scenario in the Scenarios tab and personalizing the values of parameters

Fig. 7

Custom Gremlin query. Gremlin Workbench allows the user to manually type and execute any query he wants

Fig. 8

Results for the case study scenario. The user can immediately study the results given in graphical and tree form. He can also (1) export them in several formats or (2) perform some data analysis, when applicable

Fig. 9

GO analysis Analysis of data through p-value calculation. As the figure shows, it is possible to deeply investigate a specific field of the interrogation. In the shown case study figure, it is possible to investigate about a specific annotation linked to a particular gene target