search

nrnb / GoogleSummerOfCode

Main documentation site for NRNB GSoC project ideas and resources
114 stars 38 forks source link

Metabolite plugin for PathVisio #46

Closed egonw closed 6 years ago

egonw commented 8 years ago

Background

PathVisio is a Java-based pathway editor, closely linked to WikiPathways. During a past Google Summer of Code a student implemented a plugin for showing information about metabolites in pathways. This information came from the HMDB database. Unfortunately, in the last month of the project the format in which this information was provided has considerably changed. There was not enough time to update the plugin.

Much earlier, someone else wrote a PathVisio plugin to visualize the chemical structure of metabolites in a backpage too. This plugin took advantage of the SMILES strings provided by the BridgeDb metabolite ID mapping database and the Chemistry Development Kit (CDK) to show this 2D diagram. This plugin needs updating for recent ID mapping database files as well as the recent CDK versions.

Goal

The goal of this project is to merge to code of both plugin projects and create an advanced metabolite plugin for PathVisio that can show data from an external database using modern technologies, as well as depict the 2D chemical structure of the structure using the latest CDK version.

The below screenshot shows PathVisio and the default backpage. Your plugin will create one or more additional backpages to show further information (data, 2D depiction, etc) for a metabolite selected in the left side pathway (image from the PathVisio website):

image

Technologies

PathVisio is written in Java, as is the CDK. The source code of the previous plugins is available from the PathVisio source code repository and will act as a starting point. BridgeDb identifier mappings will be used to get identifiers for a remote database from which information will be retrieved. Preferably, modern web service technologies like REST and SPARQL will be used to retrieve relevant information. The first database to be used for input is Wikidata.

Difficulty level: 2

Prior experience with REST and SPARQL in Java is highly recommended. Familiarity with molecular structure is a plus.

Potential mentors

Egon Willighagen

Contact

Egon Willighagen

egonw commented 6 years ago

I'm archiving this one. We tried for two years, with no significant interest. Java is out.