Monarch Gene-Disease-Phenotype Associations Scraper

Extract genes, diseases, phenotypes and the associations between them from the Monarch Initiative knowledge graph - the cross-resource, cross-species association layer that integrates OMIM, Orphanet, ClinGen, HPO, MGI, ZFIN and dozens of other sources into one graph. No API key, no login.

Runs against the free Monarch v3 API (api.monarchinitiative.org). Roughly ~$0.50 per 1,000 records on the Apify platform.

What it does

Two modes:

Mode	Input	Output
search	free-text queries (disease / gene / phenotype)	matching entities with ids, categories, synonyms, taxa, descriptions
associations	entityIds (biolink CURIEs like MONDO:0007947)	association edges - subject, predicate, object, evidence, publications, knowledge source

In associations mode both directions are queried (subject and object), so a disease id returns its gene-to-disease edges (disease as object) and its disease-to-phenotype edges (disease as subject).

Example (verified live)

search for Marfan syndrome -> MONDO:0007947. Feed that into associations (category gene to disease) and you get:

FBN1  causes  Marfan syndrome   (source: OMIM)
FBN1  causes  Marfan syndrome   (source: ClinGen)

...plus 180+ disease-to-phenotype edges (Mitral regurgitation, Emphysema, ...).

How is this different from the HPO Phenotype Scraper?

• HPO Phenotype Scraper scrapes the Human Phenotype Ontology itself - the controlled vocabulary of phenotypic abnormalities (HP terms, definitions, ontology hierarchy). It is a single ontology.
• This Monarch scraper is the cross-resource association layer: it does not own an ontology, it connects them - linking genes (HGNC) to diseases (MONDO) to phenotypes (HP) across species, with evidence and provenance. Use HPO to browse the phenotype vocabulary; use Monarch to ask "which genes cause this disease and what phenotypes does it present?"

Input

Field	Type	Notes
mode	select	search (default) or associations
queries	array	search mode; prefilled ["diabetes","Marfan syndrome"]
entityIds	array	associations mode; e.g. ["MONDO:0007947"]
associationCategory	select	any / gene_to_disease / gene_to_phenotype / disease_to_phenotype
searchCategory	select	any / Disease / Gene / PhenotypicFeature
maxItems	integer	default 200, max 3000

Output

Entity (search): entity_id, name, category, description, symbol, taxon, synonyms, url, source, scraped_at.

Association (associations): association_id, subject_id, subject_label, predicate, object_id, object_label, association_category, evidence_count, publication_count, publications, knowledge_source, source, scraped_at.

The biolink: prefix is stripped from categories and predicates.

Use cases

• Rare disease research - resolve a disease to its causal genes and full phenotype profile in one run.
• Gene prioritisation - rank candidate genes by disease/phenotype associations and evidence counts.
• Knowledge graphs - bulk-export subject-predicate-object triples with provenance to seed your own graph.
• Translational medicine - bridge model-organism findings to human disease via cross-species associations.

Related actors

• HPO Phenotype Scraper - the phenotype ontology
• NCBI E-utilities Scraper - PubMed / gene / nucleotide records
• Clinical Trials Scraper - ClinicalTrials.gov studies
• Reactome Pathways Scraper - biological pathways

Notes

Data belongs to the Monarch Initiative and its source resources; respect their licences. This actor is an unofficial convenience wrapper and is not affiliated with Monarch.