README.md

INDRA CoGEx

INDRA CoGEx (Context Graph Extension) is an automatically assembled biomedical knowledge graph which integrates causal mechanisms from INDRA with non-causal contextual relations including properties, ontology, and data.

Content

Knowledge source	Relation	Description
INDRA	indra_rel	The source regulates or interacts with the target according to an INDRA Statement.
INDRA Ontology	isa / partof	The source node is a subclass or part of the target node.
Gene Ontology	associated_with	The gene represented by the source is associated with the GO term represented by the target.
BGee	expressed_in	The gene represented by the source is expressed in the tissue/cell type represented by the target.
CCLE	mutated_in / copy_number_altered_in	The gene represented by the source is mutated or its copy number is altered in the cancer cell line represented by the target.
CCLE	sensitive_to	The cancer cell line represented by the source is sensitive to the drug represented by the target.
ClinicalTrials.gov	tested_in	The drug represented by the source is tested in the clinical trial represented by the target.
ClinicalTrials.gov	has_trial	The disease/condition represented by the source has a clinical trial represented by the target.
ChEMBL	has_indication	The chemical represented by the source has been studied for use against the indication represented by the target.
SIDER	has_side_effect	The chemical represented by the source has a side effect represented by the target.
Reactome	haspart	The pathway represented by the source node contains the gene represented by the target node.
WikiPathways	haspart	The pathway represented by the source node contains the gene represented by the target node.
Enzyme Codes	has_activity	The relation between a gene and its enzyme class(es).
Human Phenotype Ontology	has_phenotype	The disease represented by the source has a phenotype represented by the target.
Human Phenotype Ontology	phenotype_has_gene	The phenotype represented by the source has an associated gene represented by the target.
NIH reporter	has_publication	The NIH Reporter Project represented by the source has an associated publication represented by the target.
NIH reporter	has_clinical_trial	The NIH Reporter Project represented by the source has an associated clinical trial represented by the target.
NIH reporter	has_patent	The NIH Reporter Project represented by the source has an associated patent represented by the target.
DisGeNet	gene_disease_association	Literature curated associations between genes and diseases
DisGeNet	variant_disease_association	Literature curated associations between variants (e.g., identified by dbSNP) and diseases
DisGeNet	variant_gene_association	Literature curated associations between genes and variants (e.g., from dbSNP)
GWAS	variant_phenotype_association	Large literature curated collection of human genome-wide association studies. Phenotype and variant associations (e.g., from dbSNP) are extracted.
PubMed	annotated_with	The publiction represented by the source is annotated with the MeSH entity represented by the target.
PubMed	has_citation	The Evidence represented by the source is supported by the publication represented by the target.
PubMed	published_in	The publication represented by the source is published in the journal represented by the target.
CellMarker	has_marker	The cell type represented by the source node has the gene marker represented by the target node.
InterPro	has_domain	The gene represented by the source node has a protein domain represented by the target node.
DepMap	codependent_with	The gene represented by the source node is codependent with the gene represented by the target node in cancer cell lines.

Installation

Install the indra_cogex package with:

git clone https://github.com/bgyori/indra_cogex
cd indra_cogex
pip install -e .

Build

Build the graph then bulk import into Neo4j with:

python -m indra_cogex.sources
sh import.sh

Funding

The development of this project is funded under the DARPA Young Faculty Award (ARO grant W911NF2010255).