schema_info:
title: Reactome Pathway Database RDF
description: |
Open-source, curated knowledgebase of biological pathways and processes.
Contains 22,000+ pathways across 30+ species with molecular interactions,
biochemical reactions, protein complexes, and disease associations.
Based on BioPAX Level 3 ontology.
endpoint: https://rdfportal.org/backend/ebi/sparql
base_uri: http://www.reactome.org/biopax/
graphs:
- http://rdf.ebi.ac.uk/dataset/reactome
version:
mie_version: "1.0"
mie_created: "2025-12-08"
data_version: "Release 88"
update_frequency: "Quarterly"
license:
data_license: "CC BY 4.0"
license_url: "https://creativecommons.org/licenses/by/4.0/"
access:
rate_limiting: "No strict limits"
max_query_timeout: "60 seconds"
shape_expressions: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
PREFIX rdfs: <http://www.w3.org/2000/01/rdf-schema#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
<PathwayShape> {
a [ bp:Pathway ] ;
bp:displayName xsd:string ;
bp:organism @<BioSourceShape> ? ;
bp:comment xsd:string * ;
bp:pathwayComponent @<PathwayShape> * ;
bp:pathwayComponent @<BiochemicalReactionShape> * ;
bp:dataSource @<ProvenanceShape> * ;
bp:xref @<XrefShape> *
}
<BiochemicalReactionShape> {
a [ bp:BiochemicalReaction ] ;
bp:displayName xsd:string ? ;
bp:left @<PhysicalEntityShape> * ;
bp:right @<PhysicalEntityShape> * ;
bp:eCNumber xsd:string * ;
bp:comment xsd:string * ;
bp:xref @<XrefShape> *
}
<PhysicalEntityShape> {
a [ bp:Protein bp:Complex bp:SmallMolecule bp:Dna bp:Rna ] ;
bp:displayName xsd:string ? ;
bp:entityReference @<EntityReferenceShape> ? ;
bp:cellularLocation @<CellularLocationShape> ? ;
bp:feature @<EntityFeatureShape> *
}
<EntityReferenceShape> {
a [ bp:ProteinReference bp:SmallMoleculeReference bp:DnaReference bp:RnaReference ] ;
bp:name xsd:string * ;
bp:xref @<UnificationXrefShape> *
}
<ComplexShape> {
a [ bp:Complex ] ;
bp:component @<PhysicalEntityShape> + ;
bp:componentStoichiometry @<StoichiometryShape> *
}
<XrefShape> {
a [ bp:UnificationXref bp:PublicationXref bp:RelationshipXref ] ;
bp:db xsd:string ? ;
bp:id xsd:string ?
}
<UnificationXrefShape> {
a [ bp:UnificationXref ] ;
bp:db xsd:string ;
bp:id xsd:string
}
<BioSourceShape> {
a [ bp:BioSource ] ;
bp:name xsd:string * ;
bp:xref @<UnificationXrefShape> *
}
sample_rdf_entries:
- title: "Pathway with Hierarchy"
description: "Top-level pathway with sub-pathways and organism."
rdf: |
reactome:Pathway227 a bp:Pathway ;
bp:displayName "Platelet homeostasis" ;
bp:organism reactome:BioSource1 ;
bp:pathwayComponent reactome:Pathway234 ,
reactome:Pathway235 .
- title: "Biochemical Reaction"
description: "Reaction with substrates, products, and EC number."
rdf: |
reactome:BiochemicalReaction1154 a bp:BiochemicalReaction ;
bp:displayName "Autophosphorylation of PDGFRA" ;
bp:left reactome:Complex630 ;
bp:right reactome:Complex1445 ;
bp:eCNumber "2.7.10.1" .
- title: "Protein Complex"
description: "Multi-component protein complex with stoichiometry."
rdf: |
reactome:Complex630 a bp:Complex ;
bp:component reactome:Protein728 ,
reactome:Protein965 ;
bp:componentStoichiometry [
a bp:Stoichiometry ;
bp:physicalEntity reactome:Protein728 ;
bp:stoichiometricCoefficient 2
] .
- title: "Protein with UniProt Reference"
description: "Protein entity with external database cross-reference."
rdf: |
reactome:Protein728 a bp:Protein ;
bp:displayName "PDGFRA" ;
bp:entityReference reactome:ProteinReference728 .
reactome:ProteinReference728 a bp:ProteinReference ;
bp:name "PDGFRA" ;
bp:xref [
a bp:UnificationXref ;
bp:db "UniProt" ;
bp:id "P16234"
] .
- title: "Small Molecule in Reaction"
description: "Chemical compound with ChEBI reference."
rdf: |
reactome:SmallMolecule1234 a bp:SmallMolecule ;
bp:entityReference [
a bp:SmallMoleculeReference ;
bp:name "ATP" ;
bp:xref [
a bp:UnificationXref ;
bp:db "ChEBI" ;
bp:id "CHEBI:15422"
]
] .
sparql_query_examples:
- title: "Search Pathways by Keyword with bif:contains"
description: "Full-text search for pathways with relevance scoring"
question: "Which pathways are related to cancer?"
complexity: "basic"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
SELECT ?pathway ?name
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
?pathway a bp:Pathway ;
bp:displayName ?name .
?name bif:contains "'cancer'" option (score ?sc)
}
ORDER BY DESC(?sc)
LIMIT 20
- title: "Get Pathway Hierarchy"
description: "Retrieve sub-pathways and parent pathways"
question: "What are the sub-pathways of Platelet homeostasis?"
complexity: "basic"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
PREFIX reactome: <http://www.reactome.org/biopax/68/49646#>
SELECT ?subPathway ?name
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
reactome:Pathway227 bp:pathwayComponent ?subPathway .
?subPathway a bp:Pathway ;
bp:displayName ?name .
}
- title: "Find Reactions by EC Number"
description: "Search biochemical reactions by enzyme classification"
question: "Which reactions have EC number 2.7.10.1 (protein tyrosine kinases)?"
complexity: "intermediate"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
SELECT ?reaction ?name ?ecNumber
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
?reaction a bp:BiochemicalReaction ;
bp:eCNumber ?ecNumber .
OPTIONAL { ?reaction bp:displayName ?name }
FILTER(CONTAINS(?ecNumber, "2.7.10.1"))
}
LIMIT 50
- title: "Get Pathway Proteins with UniProt IDs"
description: "Retrieve all proteins in a pathway with external references"
question: "What are the UniProt IDs of proteins in a specific pathway?"
complexity: "intermediate"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
PREFIX reactome: <http://www.reactome.org/biopax/68/49646#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
SELECT DISTINCT ?proteinName ?uniprotId
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
reactome:Pathway227 bp:pathwayComponent*/bp:left ?entity .
?entity bp:entityReference ?proteinRef .
?proteinRef a bp:ProteinReference ;
bp:name ?proteinName ;
bp:xref ?xref .
?xref a bp:UnificationXref ;
bp:db "UniProt"^^xsd:string ;
bp:id ?uniprotId .
}
LIMIT 100
- title: "Search Pathways by Boolean Keywords"
description: "Complex keyword search with AND/OR operators"
question: "Find pathways about kinase signaling but not apoptosis"
complexity: "intermediate"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
SELECT ?pathway ?name
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
?pathway a bp:Pathway ;
bp:displayName ?name .
?name bif:contains "('kinase' AND 'signaling' AND NOT 'apoptosis')" option (score ?sc)
}
ORDER BY DESC(?sc)
LIMIT 20
- title: "Find Protein Complexes with Stoichiometry"
description: "Retrieve complexes with component counts"
question: "Which complexes have specific protein stoichiometry?"
complexity: "advanced"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
SELECT ?complex ?componentName ?coefficient
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
?complex a bp:Complex ;
bp:componentStoichiometry ?stoich .
?stoich bp:physicalEntity ?component ;
bp:stoichiometricCoefficient ?coefficient .
?component bp:entityReference/bp:name ?componentName .
FILTER(?coefficient > 1)
}
LIMIT 50
- title: "Pathway-Drug Integration via Guide to Pharmacology"
description: "Link pathways to drugs via protein targets"
question: "Which cancer pathways have proteins with known drug targets?"
complexity: "advanced"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
SELECT DISTINCT ?pathway ?pathwayName ?drugId
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
?pathway a bp:Pathway ;
bp:displayName ?pathwayName ;
bp:pathwayComponent*/bp:left ?entity .
?entity bp:entityReference ?proteinRef .
?proteinRef bp:xref ?drugXref .
?drugXref bp:db "Guide to Pharmacology"^^xsd:string ;
bp:id ?drugId .
?pathwayName bif:contains "'cancer'" option (score ?sc)
}
ORDER BY DESC(?sc)
LIMIT 50
cross_references:
- pattern: "bp:xref with bp:db"
description: |
Unified cross-references to external databases via bp:xref.
All entities (pathways, proteins, small molecules) link to external resources.
Use ^^xsd:string type restriction for database name matching.
databases:
proteins: "UniProt (87K proteins), Ensembl, RefSeq"
pathways: "GO, KEGG Pathway, PANTHER"
chemicals: "ChEBI (28K small molecules), PubChem (8K compounds), COMPOUND (14K)"
publications: "PubMed (268K evidence citations)"
drugs: "Guide to Pharmacology (8K drug targets)"
organisms: "NCBI Taxonomy"
sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
SELECT ?entity ?externalId
FROM <http://rdf.ebi.ac.uk/dataset/reactome>
WHERE {
?entity bp:xref ?xref .
?xref a bp:UnificationXref ;
bp:db "UniProt"^^xsd:string ;
bp:id ?externalId .
}
LIMIT 50
architectural_notes:
schema_design:
- "BioPAX Level 3 ontology for biological pathway representation"
- "Hierarchical pathways via bp:pathwayComponent relationships"
- "Reactions link to proteins, complexes, and small molecules"
- "Cross-references provide external database integration"
performance:
- "Use bif:contains for pathway name searches (full-text index)"
- "Boolean operators work: AND, OR, NOT in bif:contains"
- "Always specify FROM graph for multi-graph endpoints"
- "EC number filters efficient with CONTAINS or REGEX"
data_integration:
- "UniProt for protein sequences and annotations"
- "ChEBI for small molecule structures"
- "PubMed for evidence and citations"
- "GO for biological process mappings"
- "Guide to Pharmacology for drug-target information"
data_quality:
- "All pathways manually curated with evidence"
- "Computational inference noted in bp:comment"
- "Regular quarterly updates"
- "Organism-specific pathway instances"
- "CRITICAL: Use ^^xsd:string for bp:db comparisons to handle datatype issues"
data_statistics:
total_pathways: 22000
total_reactions: 11000
total_proteins: 226000
total_complexes: 101000
total_small_molecules: 50000
coverage:
pathways_with_organisms: ">95%"
reactions_with_ec_numbers: "~60%"
proteins_with_uniprot: "~90%"
pathways_with_pubmed: "~85%"
cardinality:
avg_subpathways_per_pathway: 5.3
avg_proteins_per_complex: 3.2
avg_reactions_per_pathway: 8.7
performance_characteristics:
- "bif:contains pathway search very fast (<1s)"
- "Hierarchy traversal efficient with specific starting pathway"
- "Cross-reference queries fast when filtered by db with ^^xsd:string"
anti_patterns:
- title: "Using FILTER Instead of bif:contains"
problem: "REGEX/CONTAINS is slow and not optimized for text search"
wrong_sparql: |
SELECT ?pathway ?name WHERE {
?pathway a bp:Pathway ;
bp:displayName ?name .
FILTER(CONTAINS(LCASE(?name), "cancer"))
}
correct_sparql: |
SELECT ?pathway ?name WHERE {
?pathway a bp:Pathway ;
bp:displayName ?name .
?name bif:contains "'cancer'" option (score ?sc)
}
ORDER BY DESC(?sc)
explanation: "Use bif:contains for Virtuoso-optimized full-text search with relevance ranking"
- title: "Unbounded Pathway Traversal"
problem: "bp:pathwayComponent* without starting point causes timeout"
wrong_sparql: |
SELECT ?pathway ?subPathway WHERE {
?pathway bp:pathwayComponent* ?subPathway
}
correct_sparql: |
SELECT ?subPathway WHERE {
reactome:Pathway227 bp:pathwayComponent+ ?subPathway .
?subPathway a bp:Pathway .
}
LIMIT 100
explanation: "Start from specific pathway and add type filter and LIMIT"
- title: "Missing Type Restriction for Database Names"
problem: "Direct string comparison with bp:db fails due to datatype issues"
wrong_sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
SELECT ?entity ?id WHERE {
?entity bp:xref ?xref .
?xref bp:db "UniProt" ;
bp:id ?id .
}
correct_sparql: |
PREFIX bp: <http://www.biopax.org/release/biopax-level3.owl#>
PREFIX xsd: <http://www.w3.org/2001/XMLSchema#>
SELECT ?entity ?id WHERE {
?entity bp:xref ?xref .
?xref bp:db "UniProt"^^xsd:string ;
bp:id ?id .
}
explanation: "Use ^^xsd:string type restriction to match database names correctly"
common_errors:
- error: "Query returns empty results for cross-references"
causes:
- "Direct comparison with bp:db value without type restriction"
- "Datatype mismatch between literal strings and RDF values"
- "Wrong xref type (PublicationXref vs UnificationXref)"
solutions:
- "Use ^^xsd:string type restriction: bp:db \"UniProt\"^^xsd:string"
- "Alternative: Use FILTER(STR(?db) = \"UniProt\") if variable binding needed"
- "Use bp:UnificationXref for external database IDs"
- "Use OPTIONAL for xrefs that may not exist"
- error: "Query timeout on pathway search"
causes:
- "Using FILTER/REGEX instead of bif:contains"
- "Missing LIMIT clause"
- "Unbounded transitive queries (bp:pathwayComponent* without start)"
solutions:
- "Use bif:contains for keyword searches"
- "Add LIMIT 50-100 for exploratory queries"
- "Start transitive queries from specific entity"
- "Use bp:pathwayComponent+ instead of * when possible"
- error: "Missing pathway components or proteins"
causes:
- "Not following bp:pathwayComponent* for nested pathways"
- "Missing bp:left/bp:right for reaction participants"
- "Not checking both Protein and Complex types"
solutions:
- "Use bp:pathwayComponent+ for all descendants"
- "Include both bp:left and bp:right for complete reactions"
- "Use UNION for comprehensive entity retrieval"
- "Check entity types with explicit type filters"
