Skip to content
No description, website, or topics provided.
Jupyter Notebook Other
Branch: master
Clone or download
Pull request Compare This branch is 1 commit ahead of NuriaQueralt:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
bot
curation/kylo/neo4j/networks/v20180118
monarch/1shell-animal
neo4j-graphs
ontologies
regulation
utils
wikibase-rdf
.gitignore
README.md

README.md

The NGLY1 Deficiency Knowledge Graph

This is the repository for the NGLY1 Deficiency Knowledge Graph, the reasoning context to support hypothesis discovery for NGLY1 Deficiency-CDDG (DOID:0060728) research. The user can navigate the knowledge in the graph in the Neo4j Browser website. This knowledge graph is a structured review around the research question what is the mechanism underpinning the NGLY1-AQP1 regulation association and explaining the reduced transcriptomic expression of multiple Aquaporins in NGLY1 deficient cells?. The graph v3.2 is the first deployed in the Wikibase application for community contribution and curation.

We also provide the related projects on GitHub:

  • The BioKnowledge reviewer: a Python library to build structured reviews around NGLY1 Deficiency research questions as knowledge graphs.
  • Neo4j Browser Guides: the asciidoc/html files and scripts for the creation of our Neo4j Browser Guides to assist the user on the exploration of the graph in Neo4j. Neo4j.
  • Krusty: a library to synchronize the NGLY1 Deficiency graph in the Knowledge Navigation (Neo4j) and Knowledge Contribution (Wikibase) components.
  • Queralt-Rosinach et al. 2019: Manuscript describing the project.

Data

Versions of the knowledge graph are in the neo4j-graphs/ directory.

Workflow to create the graph v3.2

We first built structured reviews using this workflow. To reproduce the creation of the NGLY1 Deficiency knowledge graph v3.2 the user might run the following workflow. Note that the BioKnowledge reviewer library was prototyped using this workflow. We encourage the user to use the library to reproduce the creation of the graph.

Prerequisites
  • Python 3.
  • Jupyter Notebooks framework and the IPython kernel.

PREPARING EDGES

Curated network

data: curation/

project-folder: regulation/

decompress the mondo ontology at: ontologies/

run jupyter-notebooks:

1. `graph.ipynb`

The curated network was built in the fly during the execution of the notebook that builds the knowledge graph from all edges reviewed.

Monarch network

project-folder: monarch/1shell-animal/

run jupyter-notebooks:

1. `get_monarch_connections.ipynb`

2. `add_connections_to_net.ipynb`

Regulation network

project-folder: regulation/

run jupyter-notebooks:

1. `regulation.ipynb`

RNA expression network

This is a data-driven network built from the NGLY1 knock out transcriptome from fly.

project-folder: regulation/

run jupyter-notebooks:

Prepare individual expression datasets:

1. `./ngly1-fly-chow-2018/chow2018_ngly1_fly_exploratory_analysis.ipynb`

Build network:

2. `transcriptomics.ipynb`

Graph nodes

project-folder: regulation/

run jupyter-notebooks:

1. `graph_nodes.ipynb`

Increase graph connectivity

project-folder: monarch/1shell-animal/

run jupyter-notebooks:

1. `get_monarch_connections_regulation_graph.ipynb`

Graph edges

project-folder: regulation/

run jupyter-notebooks:

  1. `graph.ipynb

Graph versions

Each graph version is stored.

You can’t perform that action at this time.