Sunburst visualisation of an ontology representing classes of sets of metabolic objects

Python 3.10 recommended

Requirements from requirements.txt

• numpy>=1.26.1
• plotly>=5.17.0
• scipy>=1.11.3
• SPARQLWrapper>=2.0.0
• pandas>=1.5.3

Need Apache Jena Fuseki SPARQL server for ChEBI and GO requests and their OWL files.

• Download Apache Jena Fuseki : https://jena.apache.org/download/index.cgi
• Download ChEBI ontology : https://ftp.ebi.ac.uk/pub/databases/chebi/ontology/ (chebi.owl or chebi_lite.owl)
• Download GO ontology : https://geneontology.org/docs/download-ontology/ (go.owl)

pip install ontosunburst

Inside the cloned repository :

pip install -r requirements.txt
pip install -e .

Execute followed bash script to launch server.

#!/bin/bash

FUSEKI_PATH=/path/to/apache-jena-fuseki-x.x.x
CHEBI_PATH=/path/to//chebi_lite.owl

${FUSEKI_PATH}/fuseki-server --file=${CHEBI_PATH} /chebi

#!/bin/bash

FUSEKI_PATH=/path/to/apache-jena-fuseki-x.x.x
GO_PATH=/path/to/go.owl

${FUSEKI_PATH}/fuseki-server --file=${GO_PATH} /go

With local files :

• MetaCyc (compounds, reactions, pathways)
• EC (EC-numbers)
• KEGG Ontology (modules, pathways, ko, ko_transporter, metabolite, metabolite_lipid)

With SPARQL server :

• ChEBI (chebi roles)
• Gene Ontology (<1000 go terms recommended in the interest set)

Personal ontology possible :

• Define all the ontology classes relationship in a dictionary {class: [parent classes]}
• Define the root : unique class with no parents

• Topology (1 set + 1 optional reference set) : displays proportion (number of occurrences) representation of all classes
• Enrichment (1 set + 1 reference set) : displays enrichment analysis significance of a set according to a reference set of metabolic objects

View full documentation here : https://github.com/AuReMe/Ontosunburst/wiki