https://github.com/dhimmel/myelinet
Hetnet prediction of candidate remyelinating compounds
https://github.com/dhimmel/myelinet
baranzini-lab hetionet multiple-sclerosis myelin
Last synced: about 1 month ago
JSON representation
Hetnet prediction of candidate remyelinating compounds
- Host: GitHub
- URL: https://github.com/dhimmel/myelinet
- Owner: dhimmel
- Created: 2016-03-02T20:09:38.000Z (over 10 years ago)
- Default Branch: master
- Last Pushed: 2016-03-31T22:47:57.000Z (about 10 years ago)
- Last Synced: 2025-01-28T23:50:07.615Z (over 1 year ago)
- Topics: baranzini-lab, hetionet, multiple-sclerosis, myelin
- Language: Jupyter Notebook
- Size: 70.3 KB
- Stars: 2
- Watchers: 3
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Hetnet prediction of candidate remyelinating compounds
Multiple sclerosis destroys the myelin coating neurons leading to disability. Here, we aim to find small molecules compounds that will promote remyelination. We base our search on our drug repurposing hetnet -- a network with multiple node and relationship types -- which provides a comprehensive framework for data integration. The hetnet is stored in a graph database called neo4j, which we use to compute network features.
Our approach to identify candidate compounds is unsupervised. We look for compounds that highly connected to myelination or oligodendrocyte differentiation. We specify the types of paths we are interested in. For example we look for paths where a _Compound - targets - Gene - interacts - Gene - participates - Biological Process_ where the biological process is myelination.
See [`query.ipynb`](query.ipynb) for the queries and [`data/queries.tsv`](data/queries.tsv) for the results.
## Queries for visualization
These queries should be run in the neo4j browser. They were chosen to produce
illustrative and appropriately sized visualizations.
```cypher
// Find paths between Marimastat and myelination
MATCH paths =
(s1:Compound)-[:BINDS_CbG]-(n1)-[:INTERACTS_GiG]-()
-[:PARTICIPATES_GpBP]-(t1:BiologicalProcess)
WHERE s1.identifier = 'DB00786' // Marimastat
AND t1.identifier = 'GO:0042552' // myelination
// Expressed in the central nervous system
AND exists((:Anatomy {identifier: 'UBERON:0001017'})-[:EXPRESSES_AeG]-(n1))
RETURN paths
UNION ALL
MATCH paths = (s2:Compound)-[:BINDS_CbG]-(n1)-[:INTERACTS_GiG]-()
-[:REGULATES_BPrG]-(t2:BiologicalProcess)
WHERE s2.identifier = 'DB00786' // Marimastat
AND t2.identifier = 'GO:0048709' // oligodendrocyte differentiation
RETURN paths
```
## Installation
This repository assumes a neo4j server containing the hetio-ind database is
running on port 7474.
## License
All original content in this repository is released under [CC0 1.0](https://creativecommons.org/publicdomain/zero/1.0/ "Creative Commons · Public Domain Dedication").