Skip to content

Human Phenotype Ontology Annotations (HPOA)

The Human Phenotype Ontology group curates and assembles over 115,000 annotations to hereditary diseases using the HPO ontology. Here we create Biolink associations between diseases and phenotypic features, together with their evidence, and age of onset and frequency (if known).

The monarch-phenotype-profile-ingest repository produces four files from HPOA and Orphanet: 'disease-to-phenotype', 'disease-to-mode-of-inheritance' and 'gene-to-phenotype' from the HPO Annotation files, and 'gene-to-disease' from Orphanet product 6.

The 'disease-to-phenotype' and 'disease-to-mode-of-inheritance' parsers currently only process the "abnormal" annotations. Association to "remarkable normality" may be added in the near future.

The 'disease-to-mode-of-inheritance' ingest script parses 'inheritance' record information out from the annotation file.

Gene to Disease

Gene-to-disease associations come from Orphanet's gene–disorder file, product 6 (en_product6.xml), producing orphanet_gene_to_disease_edges.tsv. HPOA's genes_to_disease.txt is no longer used for this: it types every Orphanet row UNKNOWN, while product 6 carries a per-gene association type and assessment status. OMIM gene-to-disease associations come from the separate omim-ingest (omim_gene_to_disease_edges.tsv).

Genes are resolved to NCBIGene through the HGNC complete set (by HGNC id, falling back to the symbol); the few rows with no NCBIGene id (withdrawn HGNC loci, snoRNA clusters) are skipped. An Orphanet association type the ingest does not know fails the transform, so a new upstream category gets reviewed rather than silently mapped.

Orphanet association type status association predicate
Disease-causing germline mutation(s) [/ (loss of function) / (gain of function)] in Assessed CausalGeneToDiseaseAssociation biolink:causes (loss/gain of function in subject_form_or_variant_qualifier)
Disease-causing somatic mutation(s) in Assessed CausalGeneToDiseaseAssociation biolink:causes (not treated as Mendelian for gene-to-phenotype)
Major susceptibility factor in, Modifying germline mutation in Assessed CorrelatedGeneToDiseaseAssociation biolink:contributes_to
Candidate gene tested in, Biomarker tested in, Part of a fusion gene in, Role in the phenotype of any CorrelatedGeneToDiseaseAssociation biolink:gene_associated_with_condition
any Not yet assessed CorrelatedGeneToDiseaseAssociation biolink:gene_associated_with_condition

Biolink Captured

  • biolink:CausalGeneToDiseaseAssociation or biolink:CorrelatedGeneToDiseaseAssociation (per the table above)
    • subject (NCBIGene id)
    • predicate (per the table above)
    • object (Orphanet disorder, mapped to Mondo in the merge)
    • original_predicate (the Orphanet association type)
    • original_subject (HGNC id)
    • qualifiers (the Orphanet assessment status)
    • publications (PMIDs from SourceOfValidation)
    • primary_knowledge_source (infores:orphanet)
    • aggregator_knowledge_source (["infores:monarchinitiative"])

Disease to Phenotype

phenotype.hpoa: A description of this file is found here, has the following fields:

  • 'database_id'
  • 'disease_name'
  • 'qualifier'
  • 'hpo_id'
  • 'reference'
  • 'evidence'
  • 'onset'
  • 'frequency'
  • 'sex'
  • 'modifier'
  • 'aspect'
  • 'biocuration'

Note that we're calling this the disease to phenotype file because - using the YAML file filter configuration for the ingest - we are only parsing rows with Aspect == 'P' (phenotypic anomalies), but ignoring all other Aspects.

Frequencies

The 'Frequency' field of the aforementioned phenotypes.hpoa file has the following definition, excerpted from its Annotation Format page:

8. Frequency: There are three allowed options for this field. (A) A term-id from the HPO-sub-ontology below the term “Frequency” (HP:0040279). (since December 2016 ; before was a mixture of values). The terms for frequency are in alignment with Orphanet. * (B) A count of patients affected within a cohort. For instance, 7/13 would indicate that 7 of the 13 patients with the specified disease were found to have the phenotypic abnormality referred to by the HPO term in question in the study referred to by the DB_Reference; (C) A percentage value such as 17%.

Previously this ingest attempted to map the quantitative frequency values to HPO frequency terms. The only frequency mapping that occurs is between numerical values now, if a fraction is given it will be split into has_count and has_total, and has_quotient will be calculated by division of has_count/has_total, and has_percentage will be calculated by multiplying has_quotient by 100.

Reference

The phenotype.hpoa format includes a DB_reference field described as:

 5. DB_Reference: This required field indicates the source of the information used for the annotation. This may be the clinical experience of the annotator or may be taken from an article as indicated by a PubMed id. Each collaborating center of the Human Phenotype Ontology consortium is assigned a HPO:Ref id. In addition, if appropriate, a PubMed id for an article describing the clinical abnormality may be used.

This ingest preserves values of DB_reference which are not duplicates of the database_id captured as the subject, and will ultimately be found in original_subject in the graph when the subject is mapped via SSSOM to a Mondo identifier.

Biolink captured

  • biolink:DiseaseToPhenotypicFeatureAssociation
    • id (random uuid)
    • subject (disease.id)
    • predicate (has_phenotype)
    • negated (True if 'qualifier' == "NOT")
    • object (phenotypicFeature.id)
    • publications (List[publication.id])
    • has_evidence (List[Note [1]]),
    • sex_qualifier (Note [2])
    • onset_qualifier (Onset.id)
    • frequency_qualifier (Note [3])
    • aggregating_knowledge_source (["infores:monarchinitiative"])
    • primary_knowledge_source ("infores:hpo-annotations")

Notes: 1. CURIE of [Evidence and Conclusion Ontology(https://bioportal.bioontology.org/ontologies/ECO)] term 2. female -> PATO:0000383, male -> PATO:0000384 or None 3. See the Frequencies section above.

Disease to Modes of Inheritance

Same as above, we again parse the phenotype.hpoa file.

However, we're calling this the 'disease to modes of inheritance' file because - using the YAML file filter configuration for the ingest - we are only parsing rows with Aspect == 'I' (inheritance), but ignoring all other Aspects.

Biolink captured

  • biolink:DiseaseOrPhenotypicFeatureToGeneticInheritanceAssociation
    • id (random uuid)
    • subject (disease.id)
    • predicate (has_mode_of_inheritance)
    • object (geneticInheritance.id)
    • publications (List[publication.id])
    • has_evidence (List[Note [1]]),
    • aggregating_knowledge_source (["infores:monarchinitiative"])
    • primary_knowledge_source ("infores:hpo-annotations")

Gene to Phenotype

The gene-to-phenotype ingest processes the tab-delimited HPOA gene_to_phenotype.txt file. It keeps only genes with a Mendelian (causal germline) association to the disease, taking OMIM association types from HPOA and Orphanet types from product 6 (see Gene to Disease), so Orphanet-curated causal genes contribute gene-to-phenotype edges. The file has the following fields:

  • 'ncbi_gene_id'
  • 'gene_symbol'
  • 'hpo_id'
  • 'hpo_name'

Biolink captured

  • biolink:GeneToPhenotypicFeatureAssociation
    • id (random uuid)
    • subject (gene.id)
    • predicate (has_phenotype)
    • object (phenotypicFeature.id)
    • aggregating_knowledge_source (["infores:monarchinitiative"])
    • primary_knowledge_source (infores:hpo-annotations)

Citation

Sebastian Köhler, Michael Gargano, Nicolas Matentzoglu, Leigh C Carmody, David Lewis-Smith, Nicole A Vasilevsky, Daniel Danis, Ganna Balagura, Gareth Baynam, Amy M Brower, Tiffany J Callahan, Christopher G Chute, Johanna L Est, Peter D Galer, Shiva Ganesan, Matthias Griese, Matthias Haimel, Julia Pazmandi, Marc Hanauer, Nomi L Harris, Michael J Hartnett, Maximilian Hastreiter, Fabian Hauck, Yongqun He, Tim Jeske, Hugh Kearney, Gerhard Kindle, Christoph Klein, Katrin Knoflach, Roland Krause, David Lagorce, Julie A McMurry, Jillian A Miller, Monica C Munoz-Torres, Rebecca L Peters, Christina K Rapp, Ana M Rath, Shahmir A Rind, Avi Z Rosenberg, Michael M Segal, Markus G Seidel, Damian Smedley, Tomer Talmy, Yarlalu Thomas, Samuel A Wiafe, Julie Xian, Zafer Yüksel, Ingo Helbig, Christopher J Mungall, Melissa A Haendel, Peter N Robinson, The Human Phenotype Ontology in 2021, Nucleic Acids Research, Volume 49, Issue D1, 8 January 2021, Pages D1207–D1217, https://doi.org/10.1093/nar/gkaa1043