ChEBI Download

What is ChEBI Download?

ChEBI Download is a record-retrieval tool for ChEBI, EMBL-EBI's open chemical database and ontology for biologically relevant chemical entities. It retrieves the JSON record behind one stable ChEBI accession, such as CHEBI:15365 for acetylsalicylic acid. ChEBI records combine a chemical definition with ontology relationships, making them useful identifiers for metabolites, ligands, reaction participants, and biomedical annotations.

ChEBI represents more than a catalogue of structures. Entries can describe individual molecules, ions, complexes, molecular groups, and classes. The ontology connects an entity to its chemical classes and biological roles, while the record can also carry names, synonyms, formula, charge, mass, structure identifiers, cross-references, and curation information. Which fields appear depends on the entry.

This tool retrieves an existing record. It does not search for a compound by name, create a new structure record, or determine whether a chemical is present in a sample.

How to use ChEBI Download online

ProteinIQ retrieves a ChEBI record online from a known CHEBI: accession and returns the ChEBI API response as a downloadable JSON file. Enter one identifier, such as CHEBI:15365, with ChEBI selected as the database. The result includes the JSON document, a source link, and file metadata for keeping the ontology record alongside an analysis or dataset.

Input

Input	Description
Entry ID	One ChEBI accession. Use the prefixed form, for example CHEBI:15365, CHEBI:30616, or CHEBI:17544. It is an ontology identifier, not a compound name, SMILES string, InChIKey, PubChem CID, or ChEMBL ID.

ChEBI uses stable identifiers to distinguish entities that may share a common name. Supplying the accession is especially important for automated workflows, where a name such as “caffeine” can hide differences in salt form, protonation state, stereochemistry, or whether the intended record is a molecular entity or an ontology class.

Settings

Setting	Description
Database	Defaults to ChEBI. Selecting another database changes the source and requires an identifier recognized by that database. Keep ChEBI selected for CHEBI: accessions.

Results

Output	Format	Contents
ChEBI record	JSON (.json)	The response for the requested entity from the ChEBI public API. Available fields vary by record.
Result details	Metadata	Database, provider, entry ID, record title, record type, download label, filename, format, source type, and ChEBI record URL.

The file name uses the provider, accession, and download label. For CHEBI:15365, the JSON file is named chebi-CHEBI-15365-json.json. The content is fetched from ChEBI rather than converted into a different chemical file format.

Understanding a ChEBI record

The returned JSON is designed for programmatic use, so nested objects are common. Field names and available values can change as ChEBI evolves, but the following parts are often useful when inspecting a molecular-entity record.

JSON area	Meaning
chebi_accession	The stable ChEBI identifier for the record. Preserve this field when joining data across resources.
name and ascii_name	The ChEBI name and an ASCII-compatible rendering. These are labels, not interchangeable identifiers.
definition	A curated natural-language description of the entity or class. It helps distinguish closely related chemicals but does not replace experimental evidence for a specific sample.
chemical_data	Chemical properties when available, including formula, charge, mass, and monoisotopic_mass.
default_structure	Structure identifiers such as smiles, standard InChI, and InChIKey when ChEBI provides a structure. Ontology classes may not have a concrete structure.
names	Names and synonyms from the source record. Synonym matching is useful for discovery but should not replace the accession in data exchange.
ontology_relations and roles_classification	Relationships to chemical parents, children, and biological roles. A role or class relationship describes ChEBI's ontology assertion, not a measured property of a particular experiment.
database_accessions and secondary_ids	Cross-references and retired or alternate identifiers. Check the primary accession before treating a secondary ID as the current record.
stars	ChEBI's manual-curation rating. A higher rating indicates more manual curation, not a confidence score for an assay result or structure prediction.

For example, a downloaded aspirin record can include a neutral molecular formula and a default SMILES value, while a broad chemical class may instead be useful mainly for its definition and ontology links. The JSON should be parsed defensively because neither set of fields is guaranteed for every identifier.

Using ChEBI identifiers in data workflows

ChEBI accessions work well as durable chemistry references in tables that combine data from different biological resources. Rhea uses ChEBI entities to identify reaction participants, so a known participant accession can be retrieved with ChEBI Download while the reaction itself is retained through Rhea reaction records. This keeps the reaction definition distinct from the chemical-entity definition.

The record is also useful when a project needs to retain the source context behind a metabolite or ligand label:

• Capture the ChEBI accession, retrieved JSON, and retrieval date with a pathway, metabolomics, or annotation export.
• Use formula, charge, and structure identifiers as cross-checks when reconciling compound records from different databases.
• Follow ontology relations when grouping compounds by chemical class or biological role, while retaining the specific entity accession in the underlying data.
• Extract a returned SMILES value into an SDF structure only when the record contains a concrete molecular structure and that format is needed downstream.

ChEBI is curated and released over time. The accession is stable, but the definition, cross-references, relationships, and other annotations can be revised. The downloaded JSON is therefore a retrieval-time snapshot, not a frozen copy of all future ChEBI content.