family_id: PANTHER:PTHR23192 family_name: OLFACTOMEDIN-RELATED preferred_name: Olfactomedin-domain extracellular proteins summary: Olfactomedin-domain proteins include OLFML2A/2B, noelins, myocilin and gliomedin. Extracellular and membrane-associated architectures occur, with distinct ligand interactions, tissue distributions and matrix or signaling roles. functional_coherence: HETEROGENEOUS coherence_reason: Mouse OLFML2A/photomedin experiments support extracellular localization and glycosaminoglycan binding. This is a justified starting point for OLFML2A ortholog comparison, but matrix residence, matrix assembly and specific binding each require their own evidence. The short primate target and the horse gene-model differences require separate completeness checks; no single matrix function is safe across the entire olfactomedin family. subfamilies: - subfamily_id: PANTHER:PTHR23192:SF29 label: OLFACTOMEDIN-LIKE PROTEIN 2A description: Recorded placement of A0A8C9H4D2, OLFML2A, OLFML2A. Mouse OLFML2A/photomedin experiments support extracellular localization and glycosaminoglycan binding. This is a justified starting point for OLFML2A ortholog comparison, but matrix residence, matrix assembly and specific binding each require their own evidence. The short primate target and the horse gene-model differences require separate completeness checks; no single matrix function is safe across the entire olfactomedin family. representative_members: - id: UniProtKB:A0A8C9H4D2 gene_symbol: A0A8C9H4D2 organism: Piliocolobus tephrosceles gene_review: genes/9PRIM/A0A8C9H4D2/A0A8C9H4D2-ai-review.yaml - id: UniProtKB:A0A9L0SKW1 gene_symbol: OLFML2A organism: Equus caballus gene_review: genes/HORSE/OLFML2A/OLFML2A-ai-review.yaml - id: UniProtKB:Q68BL7 gene_symbol: OLFML2A organism: Homo sapiens gene_review: genes/human/OLFML2A/OLFML2A-ai-review.yaml term_assessments: - assessed_term: id: GO:0031012 label: extracellular matrix scope: UNRESOLVED scope_reason: Photomedins are experimentally secreted extracellular proteins that bind selected glycosaminoglycans. Extracellular presence and matrix binding support an ECM-associated hypothesis for OLFML2A orthologs but do not establish stable ECM membership for every olfactomedin protein. The short primate sequence and altered horse leader require product-specific secretion checks. supported_by: - reference_id: PMID:15836428 supporting_text: preferentially bound to chondroitin sulphate-E and heparin - reference_id: PMID:15836428 supporting_text: 'These proteins, named photomedin-1 and photomedin-2, were secreted as disulphide-bonded dimers (photomedin-1) or oligomers/multimers (photomedin-2) with O-linked carbohydrate chains' - reference_id: PMID:15836428 supporting_text: 'Among a panel of ECM components, including glycosaminoglycans, photomedins preferentially bound to chondroitin sulphate-E and heparin.' - reference_id: PMID:15836428 supporting_text: 'We screened more than 60000 RIKEN mouse cDNAs for novel ECM (extracellular matrix) proteins' - reference_id: file:genes/HORSE/OLFML2A/OLFML2A-bioinformatics/RESULTS.md supporting_text: share 91.3% identity among 598 paired residues. - assessed_term: id: GO:0030198 label: extracellular matrix organization scope: UNRESOLVED scope_reason: An ECM-screening study measured matrix assembly among several distinct candidate functions, but the fact that a protein was screened is not a positive assembly result. The photomedin binding study establishes extracellular interactions, not by itself extracellular-matrix organization. An actual assembly phenotype or member-specific functional result is required before extending this process. supported_by: - reference_id: PMID:15836428 supporting_text: preferentially bound to chondroitin sulphate-E and heparin - reference_id: PMID:18757743 supporting_text: "These assays screened \nthe candidate proteins for ECM-assembling activities" - reference_id: PMID:15836428 supporting_text: 'Among a panel of ECM components, including glycosaminoglycans, photomedins preferentially bound to chondroitin sulphate-E and heparin.' - reference_id: PMID:15836428 supporting_text: 'These proteins, named photomedin-1 and photomedin-2, were secreted as disulphide-bonded dimers (photomedin-1) or oligomers/multimers (photomedin-2) with O-linked carbohydrate chains' - reference_id: file:genes/HORSE/OLFML2A/OLFML2A-bioinformatics/RESULTS.md supporting_text: share 91.3% identity among 598 paired residues. - assessed_term: id: GO:0042802 label: identical protein binding scope: UNRESOLVED scope_reason: Photomedin-1 is experimentally secreted as disulfide-bonded dimers, providing a concrete OLFML2A self-association result. Other olfactomedins have different oligomeric architectures and proteolytic processing. Retention of the relevant association region, not merely the olfactomedin domain, determines whether this interaction transfers. supported_by: - reference_id: PMID:15836428 supporting_text: 'These proteins, named photomedin-1 and photomedin-2, were secreted as disulphide-bonded dimers (photomedin-1) or oligomers/multimers (photomedin-2) with O-linked carbohydrate chains' - reference_id: file:genes/HORSE/OLFML2A/OLFML2A-bioinformatics/RESULTS.md supporting_text: share 91.3% identity among 598 paired residues. - reference_id: file:genes/HORSE/OLFML2A/OLFML2A-bioinformatics/RESULTS.md supporting_text: Human UniProt annotates residues 1–27 as the signal peptide; the horse sequence starts `MDSQVFGDMDQVRMTSEGSDCRCKCIMRPLSKDA`, without the hydrophobic human leader. - reference_id: file:genes/human/OLFML2A/OLFML2A-uniprot.txt supporting_text: 'DE AltName: Full=Photomedin-1;' - reference_id: file:genes/human/OLFML2A/OLFML2A-uniprot.txt supporting_text: FT SIGNAL 1..27 - reference_id: PMID:15836428 supporting_text: 'were secreted as disulphide-bonded dimers (photomedin-1) or oligomers/multimers (photomedin-2)' - assessed_term: id: GO:0050840 label: extracellular matrix binding scope: UNRESOLVED scope_reason: Photomedins preferentially bind chondroitin sulfate-E and heparin among tested ECM components. This directly supports matrix-component binding for those characterized proteins without specifying all olfactomedin ligands or functions. The complete binding-positive branch boundary and integrity of the selected products remain unresolved. supported_by: - reference_id: PMID:15836428 supporting_text: 'Among a panel of ECM components, including glycosaminoglycans, photomedins preferentially bound to chondroitin sulphate-E and heparin.' - reference_id: PMID:15836428 supporting_text: 'We screened more than 60000 RIKEN mouse cDNAs for novel ECM (extracellular matrix) proteins' - reference_id: file:genes/HORSE/OLFML2A/OLFML2A-bioinformatics/RESULTS.md supporting_text: share 91.3% identity among 598 paired residues. - reference_id: file:genes/HORSE/OLFML2A/OLFML2A-bioinformatics/RESULTS.md supporting_text: Human UniProt annotates residues 1–27 as the signal peptide; the horse sequence starts `MDSQVFGDMDQVRMTSEGSDCRCKCIMRPLSKDA`, without the hydrophobic human leader. - reference_id: file:genes/human/OLFML2A/OLFML2A-uniprot.txt supporting_text: FT SIGNAL 1..27 - reference_id: PMID:15836428 supporting_text: 'photomedins preferentially bound to chondroitin sulphate-E and heparin.' references: - id: PMID:15836428 title: 'Identification and characterization of photomedins: novel olfactomedin-domain-containing proteins with chondroitin sulphate-E-binding activity.' - id: PMID:18757743 title: Transcriptome-based systematic identification of extracellular matrix proteins. - id: file:genes/HORSE/OLFML2A/OLFML2A-bioinformatics/RESULTS.md - id: file:genes/human/OLFML2A/OLFML2A-uniprot.txt - id: file:interpro/panther/PTHR23192/PTHR23192-entries.csv - id: file:interpro/panther/PTHR23192/PTHR23192-paint.tsv review_status: COMPLETE reviewed_by: codex review_date: '2026-09-10' notes: '9PRIM/A0A8C9H4D2: requested accession A0A8C9H4D2; current record A0A8C9H4D2; sequence length 439; source cohort ARGO50. HORSE/OLFML2A: requested accession A0A9L0SKW1; current record A0A9L0SKW1; sequence length 599; source cohort HORSE40. human/OLFML2A: requested accession Q68BL7; current record Q68BL7; sequence length 652; source cohort HORSE40_HUMAN_PAIR. Family boundaries were assessed using the fetched member set and the cited domain, literature and analytical evidence. Gene-review verdicts and ARBA assertions are not independent validation. PANTHER subfamily labels and membership are recorded from the frozen UniProt records and panther.obo; they do not by themselves prove functional uniformity within a subfamily. PAINT rows were inspected as ancestral assertions, but node placement is not adjudicated without the relevant descendant tree/alignment. No residue requirement is encoded without a verified sequence-version anchor. A target-level refutation is not a family-wide NOT_APPLICABLE judgment. Exact protein-name-only predictions are treated as family-identification hypotheses rather than invented GO predictions.'