Pathways · Database
Reactome
A curated, peer-reviewed knowledgebase of human biological pathways and molecular reactions for life scientists.
What Reactome is
Reactome is an open-access biological pathway database maintained by an international consortium spanning the Ontario Institute for Cancer Research (OICR), Oregon Health & Science University (OHSU), EMBL-EBI, and NYU Langone Medical Center (NYULMC). It provides detailed mechanistic representations of biological pathways and molecular processes in a computationally accessible structure, supporting basic research, genome analysis, disease studies, and systems biology.
The core unit of the Reactome data model is the reaction, grouping nucleic acids, proteins, complexes, and small molecules into interconnected biological networks. Human biological processes are broken down into discrete molecular events—including intermediary metabolism, binding, complex formation, cleavage, and subcellular transport—grounded in literature-backed experimental evidence or expert-reviewed inferences.
Beyond human biology, the database computationally projects pathways to other organisms through an orthology-based pipeline. Reactome also provides visual tools for inspecting molecular pathways, mapping identifiers, and performing multi-omics and expression analyses.
How it is organised
Biological processes are broken down into physical entities and reactions. Each event defines input physical entities (substrates), outputs (products), and any facilitating catalysts or enzymes. Molecules are defined by their subcellular localization using Gene Ontology terms, meaning identical chemical entities in different cellular compartments (such as cytosolic versus extracellular glucose) are tracked as distinct instances. Physical entities are cross-referenced to external resources such as UniProt for proteins and ChEBI for small molecules.
Human pathway steps are manually curated by PhD-level biologists in collaboration with Reactome editorial staff and verified by peer review, with each step linked to experimental literature citations. When human experimental data is unavailable, steps may be manually inferred from non-human data if validated by an expert author and reviewer. Pathways for non-human species are derived computationally from human pathways using orthology prediction. Pathway diagrams are presented using a Process Description Systems Biology Graphical Notation (SBGN) standard.
Key facts
- Core data unit
- Reaction
- Project leadership
- Lincoln Stein (OICR), Peter D'Eustachio (NYULMC), Henning Hermjakob (EMBL-EBI), Guanming Wu (OHSU)
- Release frequency
- Quarterly
- Export formats
- SBML, BioPAX, PSI-MITAB, Neo4j GraphDB, MySQL, PDF
- Diagram standard
- Systems Biology Graphical Notation (SBGN) Process Description
- Ontologies used
- Gene Ontology (GO), UniProt, ChEBI
What you can use it for
- Pathway over-representation analysis
- Submit single-column lists of protein or compound identifiers to discover statistically over-represented biological pathways.
- Expression and omics data overlay
- Project numeric values from differential gene expression, quantitative proteomics, single-cell RNA-seq, or GWAS onto pathway diagrams using spectral colour gradients across experimental conditions or timepoints.
- Disease network analysis with ReactomeFIViz
- Investigate functional relationships and pathway networks in cancer and other diseases using a Cytoscape plugin connected to the Reactome Functional Interaction network.
- Cross-species pathway inference
- Examine biological pathways computationally mapped from curated human models to other organisms using orthology relationships.
Query it here
The research agent queries Reactome directly — no account, no API key — and every answer links back to the record it came from. Ask in plain language and it writes the query.
Strengths and limits
Good at
- All human pathway steps are manually curated by domain experts and supported by peer-reviewed literature citations.
- Molecules in different subcellular compartments are treated as distinct physical entities to maintain accurate compartmental mechanics.
- Pathway models can be exported in multiple computational formats, including SBML, BioPAX, and Neo4j graph databases.
- Integrated analysis tools support multi-omics and single-cell RNA-seq datasets, including direct integrations with public repositories like EBI Expression Atlas and GEO.
Watch out for
- Non-human pathways are primarily derived computationally via orthology rather than independently curated from species-specific literature.
- Inclusion of non-human experimental evidence into human pathways requires manual inference and mutual agreement between an expert author and reviewer.
Questions
Can I search Reactome without writing code?
Yes. The research agent queries Reactome directly, with no account or API key, and cites the record behind every answer. Ask in plain language — for a gene, a protein, a compound or a study — and it builds the query for you.
Can I reuse data from Reactome?
CC0. Covers database records; illustrations and software have separate terms. Terms: https://reactome.org/license
How are reactions modelled in Reactome?
Reactions are modelled as discrete events converting input physical entities into output physical entities, with optional catalysis. These events encompass chemical transformations, binding, complex formation, cleavage, and movement between subcellular compartments.
What molecule identifiers does Reactome accept for analysis?
Reactome's tools accept single columns of protein and compound identifiers, with UniProt and ChEBI identifiers preferred, alongside support for mixed identifiers and gene symbols.
Can I cite individual pathways from Reactome?
Yes, completed pathways are assigned digital object identifiers (DOIs) and can be cited individually as publications with named authors and reviewers.
What is ReactomeFIViz?
ReactomeFIViz is a Cytoscape application that enables researchers to access Reactome pathways, execute pathway enrichment, and explore functional interaction networks associated with cancer and other diseases.
Written from Reactome's own website and reference material. Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.