Mondo

The Mondo Disease Ontology harmonizes disease concepts and identifiers across clinical and research resources. It covers rare and common diseases, including genetic and non-genetic disorders. Mondo is coordinated by the Monarch Initiative community.

Mondo is available in Athena as a non-Standard source vocabulary (vocabulary_id = 'Mondo'). It enables recognition and preservation of Mondo-coded source data and evaluation of disease specificity in OMOP. The released vocabulary contains Condition and Observation concepts and has no Mondo-to-Standard mappings. See the Vocabularies release notes.

Sources

Mondo is distributed as versioned OWL, OBO, and JSON files under CC BY 4.0. The Mondo download page provides the ontology distributions and associated documentation.

For OMOP ETL and analysis, use the Mondo content in the installed OHDSI vocabulary release. Its VOCABULARY record identifies the vocabulary version. Record that version when reporting analyses or exchanging concept sets.

Scope

Mondo reconciles disease concepts across resources including OMIM, Orphanet/ORDO, MeSH, NCIT, Disease Ontology, GARD, and ICD and SNOMED CT alignments. A native Mondo cross-reference does not establish an OMOP mapping.

Mondo complements HPO and ORPHAcodes in rare disease work: HPO describes phenotypic abnormalities, ORPHAcodes provide a rare-disease-specific nomenclature, and Mondo reconciles disease concepts across resources. Mondo's inclusion as a source vocabulary does not change the current OMOP Standard representation of conditions.

Concept Names and Codes

concept_name identifies the Mondo concept. Resolve source codes using both concept_code and vocabulary_id = 'Mondo'.

MONDO identifiers retain their numeric component and leading zeros. OMOP uses the underscore-form code:

Identifier representation	OMOP concept_code
http://purl.obolibrary.org/obo/MONDO_0008764	MONDO_0008764
MONDO:0008764	MONDO_0008764

Retain the original source value even when normalizing its representation for lookup.

Standard Concepts and Mappings

Mondo concepts are non-Standard (standard_concept IS NULL). They identify source values in appropriate *_source_concept_id fields and must not be used as Standard concepts in clinical event fields such as condition_concept_id.

The released vocabulary has no Mondo-to-Standard Maps to relationships. Native Mondo xrefs, SSSOM predicates, and ontology equivalence assertions do not substitute for these relationships. Any separately curated local mapping requires validation and provenance and must be distinguished from a released OHDSI mapping.

Mondo's native disease hierarchy does not itself provide a Standard analytical hierarchy in CONCEPT_ANCESTOR. Standard concept-set expansion uses the released Standard and Classification hierarchies. Analyses using Mondo source identifiers require explicit source-concept selection.

Domains

The August 2026 OHDSI vocabulary release contains the following Mondo concepts:

Domain	Concepts
Condition	23,224
Observation	4,529
Total	27,753

Source: Vocabularies release notes. These figures describe the released OMOP vocabulary, not the native Mondo rare disease subset.

Use the domain_id and concept_class_id of each released CONCEPT record. Mondo concepts must not all be treated as Condition concepts.

Rare Disease Use Cases

Mondo supports recognition of source codes from registries, clinical genomics, Phenopackets, and other Mondo-aligned data. Preserving the source identifier retains disease and subtype information needed to evaluate semantic loss during ETL.

For cohort discovery and landscape studies, distinguish analyses using Standard concepts from analyses explicitly using Mondo source concepts or separately curated mappings. Adding Mondo to the vocabulary cannot recover disease specificity already lost upstream or make existing ICD/SNOMED-coded records automatically Mondo-coded. Source-based studies need documented coverage, provenance, and validation across sites.

Native Mondo provides a rare subset combining authority-specific annotations, manual curation, and ontology inference. For a rare disease analysis, select the annotation appropriate to the study, record the source release, and link the selected subset to OMOP Mondo concepts by identifier. The subset files also contain classification ancestors, so file membership alone does not establish rarity: filter on the relevant subset annotation. See Mondo's rare disease subset documentation.

Membership in the OMOP Mondo vocabulary does not identify a concept as rare. Rarity must be established through the study's selected subset definition.

Instructions for ETL

When source data contain a Mondo code:

1. Normalize its representation for lookup and resolve it using vocabulary_id = 'Mondo' in the installed vocabulary release. Check the concept's validity and domain.
2. Preserve its OMOP concept_id in the appropriate *_source_concept_id field and the original code in *_source_value.
3. If mappings exist in future releases, use a valid Maps to relationship to an active Standard concept according to OMOP ETL rules, including the target domain, while retaining the Mondo source concept and original source value.
4. For an unmapped record, set the event's Standard concept field to 0 while retaining its source fields. For example, an unmapped Mondo-coded diagnosis in CONDITION_OCCURRENCE has condition_concept_id = 0, the Mondo concept ID in condition_source_concept_id, and the original code in condition_source_value. If the source code itself cannot be resolved, use 0 for the source concept ID and retain the source value.
5. Do not infer an OMOP mapping from a native xref or exactMatch assertion. If a separately validated local mapping is used, preserve its provenance and apply normal OMOP rules for the Standard target and its domain.

Apply the OMOP CDM field conventions for the destination table. A Mondo identifier assigned through a local crosswalk is derived information: retain the original source coding and the crosswalk provenance rather than implying that the source system recorded Mondo directly.

Stewardship

The OHDSI contribution is maintained by the Monarch Initiative / TISLab team at the University of North Carolina at Chapel Hill, in collaboration with the OHDSI Rare Disease Working Group and Thesaurus Health. The contribution issue records the vocabulary's submission and maintenance discussion.