COMP (thrombospondin-5) is a secreted, calcium-binding matrix glycoprotein that assembles through its N-terminal coiled-coil region into pentamers. It binds collagen and other matrix ligands and promotes early collagen fibril organization in cartilage, tendon and additional connective tissues. Integrin-dependent attachment and context-dependent survival and vascular effects extend its matrix role. Its calcium-binding repeats and C-terminal domain couple conformation to ligand interactions. Pathogenic variants disrupt matrix organization, secretion or cellular homeostasis and cause skeletal or tendon-associated disease. The shorter annotated splice product lacks most of the second EGF-like domain and a short adjacent segment.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0031012 extracellular matrix | IBA GO_REF:0000033 | ACCEPT | Reason: Extracellular matrix is the established location of secreted COMP and its matrix-organizing activity. The historical PAINT node was not independently reconstructed from an accessible annotated tree/MSA; acceptance rests on coherent target localization and matrix experiments, not the number of listed donors. No loss of this property in human COMP is evident. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN000810635 UNRESOLVED The current tree/MSA service attempts timed out; no independent reconstruction of the historical IBD placement is claimed. Target/donor biological support is assessed separately. |
| GO:0005509 calcium ion binding | IEA GO_REF:0000002 | ACCEPT | Reason: The calcium-binding EGF/type-3 signature supports this electronic assignment, independently corroborated by the original calcium-binding study and the human signature-domain structure. These results concern tested wild-type/domain constructs and do not establish equal calcium binding by both official splice products. |
| GO:0005576 extracellular region | IEA GO_REF:0000120 | ACCEPT | Reason: The signal peptide, secretion experiments and extracellular matrix detections support the extracellular region. This broad destination is compatible with later matrix incorporation and does not imply every translated or misfolded mutant molecule is extracellular. |
| GO:0006915 apoptotic process | IEA GO_REF:0000120 | MARK AS OVER ANNOTATED | Reason: This retired keyword-derived apoptotic-process assertion conflates effects on cell death with execution of apoptosis. The source-specific positive experiment supports suppression of apoptosis, while toxic-mutant stress can increase death. Neither establishes normal COMP as an apoptosis-executing component; the explicit anti-apoptotic assertion is retained separately. |
| GO:0007155 cell adhesion | IEA GO_REF:0000120 | KEEP AS NON CORE | Reason: COMP-coated substrate supports chondrocyte attachment through integrins. This is a real contextual matrix-ligand contribution to adhesion, not an inference that the secreted protein is itself a transmembrane adhesion receptor. The historical keyword row is retired because it is absent from current GOA, not because this activity is disproved. Supporting Evidence: |
| GO:0008201 heparin binding | IEA GO_REF:0000120 | KEEP AS NON CORE | Reason: Direct heparin-binding experiments support this ligand activity. The historical combined-method row is retained as retired provenance; the current experimental and ARBA assertions independently retain the same positive function. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:16611630 ADAMTS-12 associates with and degrades cartilage oligomeric ... | MODIFY | Reason: The original ADAMTS12 study identifies COMP–ADAMTS12 association using yeast two-hybrid, purified binding/pull-down and cellular association assays, with COMP EGF-like and ADAMTS12 C-terminal domains. ADAMTS12 is a protease, so protease binding is informative. COMP is the binding partner/substrate, not the enzyme performing cleavage. Proposed replacements: protease binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:17588949 Interaction of cartilage oligomeric matrix protein/thrombosp... | MODIFY | Reason: This retired aggregate protein-binding assertion is resolved by current partner-specific aggrecan records. The original experiment binds whole aggrecan, a chondroitin-sulfate proteoglycan; preserve that positive ligand class without inventing distinct human and bovine preparations from the two database accessions. Proposed replacements: chondroitin sulfate proteoglycan binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:18485748 Inhibition of ADAMTS-7 and ADAMTS-12 degradation of cartilag... | MODIFY | Reason: The curated positive COMP association concerns ADAMTS7/ADAMTS12, established proteases. The selected abstract/Introduction/Discussion describes their COMP-degradation context and antecedent binding work; current IntAct retrieval did not reproduce these exact pairs, which is not evidence against them. Refine the known partner class without asserting a cleavage activity of COMP. Proposed replacements: protease binding |
| GO:0005515 protein binding | IPI PMID:25416956 A proteome-scale map of the human interactome network. | MODIFY | Reason: The three positive Rolland yeast two-hybrid records identify human COMP and canonical 354-residue OTX1. Official OTX1 identity and its curated RNA-polymerase-II DNA-binding transcription-factor activity support this counterpart class. This preserves the screen interaction, without asserting native extracellular–nuclear encounter, COMP transcriptional activity or a mapped interface. Proposed replacements: RNA polymerase II-specific DNA-binding transcription factor binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:31515488 Extensive disruption of protein interactions by genetic vari... | MODIFY | Reason: The positive Fragoza record identifies OTX1 and the COMP construct carrying the residue-386 variant, annotated as a mutation with no effect on this tested interaction. OTX1 is a DNA-binding RNA-polymerase-II transcription factor. The refinement is a partner-class description of the recorded yeast assay, not a wild-type-only, native nuclear or COMP transcription claim. Proposed replacements: RNA polymerase II-specific DNA-binding transcription factor binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | KEEP AS NON CORE | Reason: This retired aggregate HuRI assertion collapsed seven distinct current partner records. Preserve the genuine positive binding scope here without selecting one replacement for the whole heterogeneous group; the seven exact current records are adjudicated individually. Retirement records source granularity, not biological rejection. Supporting Evidence: |
| GO:0001501 skeletal system development | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: Human COMP dysplasia genetics and mouse mutant growth-plate/matrix abnormalities support a broad skeletal-development contribution. This is contextual structural biology, not a claim that COMP instructs skeletal cell fate. The old Ensembl-transfer row is retired; its original missing support list has not been reconstructed as if newly supplied by GOA. |
| GO:0001503 ossification | IEA GO_REF:0000107 | UNDECIDED | Reason: Current mouse donor annotations include calcification and bone-related experiments, but calcification inhibition in vascular smooth-muscle contexts does not by itself resolve this historical human ossification transfer. Positive donor phenotypes are preserved; the complete relevant developmental assay and historical transfer premise were not recovered for this retired assertion. |
| GO:0002020 protease binding | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The current mouse donor has experimental protease-binding records, and human COMP–ADAMTS12 evidence independently supports the same partner class. Retain binding without confusing a protease substrate or inhibitor with a protease catalytic function. |
| GO:0002063 chondrocyte development | IEA GO_REF:0000107 | UNDECIDED | Reason: The mouse mutant study measures growth-plate organization, BrdU incorporation and cell stress/death. These are positive phenotypes, but they do not uniquely resolve normal COMP participation in chondrocyte development versus altered matrix or toxic-mutant effects. Additional donor developmental papers remain incompletely accessed for this historical transfer. Supporting Evidence: |
| GO:0003416 endochondral bone growth | IEA GO_REF:0000107 | UNDECIDED | Reason: The current mouse donor carries an endochondral-growth annotation from PMID20138147, but the original experiment supporting that specific transfer was not recovered. Matrix-dependent growth is plausible; neither dysplasia alone nor retirement from current GOA is used to reject the function. |
| GO:0003417 growth plate cartilage development | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: Actual mutant-mouse growth plates show disordered chondron organization, altered matrix and mislocalized matrix partners. Together with performed matrix organization, this supports a contextual contribution to growth-plate cartilage development; it does not establish a lineage-instructing mechanism or equal effects of all mutations. |
| GO:0005178 integrin binding | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: Integrin-dependent attachment to COMP-coated substrate establishes a contextual ligand function. The original experiment distinguishes calcium-replete COMP/alpha5beta1 from calcium-depleted and DTT-reduced COMP/alphavbeta3; retain this broad electronic binding assertion without extrapolating identical receptor usage to every native conformation. |
| GO:0005615 extracellular space | IEA GO_REF:0000107 | MODIFY | Reason: The historical extracellular-space identifier is preserved unchanged in the source object. Current extracellular region represents the supported secreted location; COMP occurs in tissue matrix and extracellular deposits. This is a current-term replacement, not evidence that the old localization was biologically wrong. Proposed replacements: extracellular region |
| GO:0006986 response to unfolded protein | IEA GO_REF:0000107 | MARK AS OVER ANNOTATED | Reason: In the actual mutant-mouse study, abnormal COMP provokes an unfolded-protein response measured through chaperones and stress mediators. This does not establish wild-type COMP as an effector of that response. Preserve the positive mutant stress observations while marking the historical electronic response-function inference as over-specific. Supporting Evidence: |
| GO:0007596 blood coagulation | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The donor thrombin study reports direct binding, inhibition of thrombin-dependent endpoints and platelet-derived COMP effects in mice. This supports contextual involvement in coagulation regulation, with an inhibitory direction; it is not a claim that COMP catalyzes clot-forming proteolysis. Full Methods were unavailable, so no finer mechanistic replacement is forced. Supporting Evidence: |
| GO:0009306 protein secretion | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The complete abstract reports intracellular assistance of collagen secretion, retention of collagen in COMP-null fibroblasts and restoration by wild-type COMP. This is more than COMP merely being secreted. Retain the existing contextual secretion contribution with the abstract-only assay/construct limit, without adding a new transport mechanism. Supporting Evidence: |
| GO:0010468 regulation of gene expression | IEA GO_REF:0000107 | UNDECIDED | Reason: The historical regulation-of-gene-expression transfer has current mouse donor records, but the relevant full primary experiments were not recovered. Expression changes downstream of matrix or stress effects do not alone resolve the specific regulatory participation asserted here; no transcription-factor activity is inferred for COMP. |
| GO:0014829 vascular associated smooth muscle contraction | IEA GO_REF:0000107 | UNDECIDED | Reason: Current donor evidence names vascular smooth-muscle contraction, but the relevant original PMID27498005 assay was not recovered. A matrix influence on vascular phenotype is plausible; this retired human transfer remains unresolved rather than being removed from an extracellular-only assumption. |
| GO:0016485 protein processing | IEA GO_REF:0000107 | UNDECIDED | Reason: The donor thrombin source reports inhibition of proteolytic endpoints and direct binding. The broad protein-processing transfer may capture regulation rather than catalytic processing, but the full donor Methods and precise historical premise remain unavailable. COMP is not assigned peptidase activity from being bound or cleaved. Supporting Evidence: |
| GO:0030199 collagen fibril organization | IEA GO_REF:0000107 | ACCEPT | Reason: COMP performs collagen-organization work: original bovine experiments compare pentameric COMP with an N-terminally truncated monomer and measure fibril formation, while mouse tendon experiments show altered fibril architecture. The electronic human assertion fits the integrated matrix function; these corroborating studies are not mislabeled human IDA. |
| GO:0030282 bone mineralization | IEA GO_REF:0000107 | UNDECIDED | Reason: The available donor abstracts support effects on vascular calcification and osteochondrogenic transition, but these are not automatically equivalent to bone mineralization in the historical human transfer. Relevant complete assay details were not recovered; no broad denial of COMP-mediated mineralization regulation is made. |
| GO:0030500 regulation of bone mineralization | IEA GO_REF:0000107 | UNDECIDED | Reason: COMP binding to BMP2 and inhibition of vascular calcification are positive donor observations. Whether the historical bone-mineralization-regulation assertion refers to actual bone or an osteogenic vascular model remains unresolved from the available abstract and current records. Supporting Evidence: |
| GO:0030509 BMP signaling pathway | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The donor abstract explicitly reports direct COMP–BMP2 binding, impaired receptor engagement and reduced osteogenic signaling. This supports a contextual contribution to BMP signaling regulation rather than COMP itself being a BMP receptor or transcriptional effector. Full Methods and exact human equivalence were not established. Supporting Evidence: |
| GO:0032991 protein-containing complex | IEA GO_REF:0000107 | MODIFY | Reason: The broad complex annotation can be specified as thrombospondin complex, whose definition accommodates homopentameric thrombospondins. Full-length COMP oligomerization supports this choice. The isolated human A225–A757 crystal-packing trimer is explicitly nonbiological and is not the basis for calling native COMP a trimer. Proposed replacements: thrombospondin complex |
| GO:0035264 multicellular organism growth | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: Mutant mice develop altered skeletal growth together with matrix and growth-plate abnormalities. Retain broad organism-growth context as a secondary consequence of structural tissue function, without treating a disease phenotype as a distinct growth-promoting molecular activity. Supporting Evidence: |
| GO:0035988 chondrocyte proliferation | IEA GO_REF:0000107 | UNDECIDED | Reason: The original mouse study quantifies reduced BrdU labeling in mutant growth plates but discusses both matrix-mediated and intracellular-stress mechanisms. The specific normal COMP contribution to proliferation is unresolved; positive mutant data are retained without converting them automatically into a wild-type proliferative function. Supporting Evidence: |
| GO:0035989 tendon development | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: Mouse mutant tendon studies show altered fibril organization, tendon dimensions and mechanics, supporting a structural contribution to tendon development. The historical transfer remains retired metadata; it is not rejected because the phenotype involves a secreted matrix component. |
| GO:0036122 BMP binding | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The current mouse donor has the exact BMP-binding annotation and the original abstract explicitly reports direct C-terminal COMP–BMP2 binding. Retain the ligand class at that donor/abstract scope, without claiming autonomous BMP signaling activity or a defined human splice-product binding site. Supporting Evidence: |
| GO:0043588 skin development | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The source abstract reports human skin matrix association and functional mouse fibroblast/skin effects of COMP loss, including altered fibrils and secretion rescue. This supports contextual skin biology, while the full developmental timing and cell-type scope remain unresolved. |
| GO:0048844 artery morphogenesis | IEA GO_REF:0000107 | UNDECIDED | Reason: The exact current mouse donor has an artery-morphogenesis record, but the complete source-specific remodeling assay was not recovered. Vascular matrix and calcification findings alone do not settle this historical morphogenesis transfer. Supporting Evidence: |
| GO:0050881 musculoskeletal movement | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The mouse study directly measures grip weakness and tendon cyclic-strain abnormalities. A matrix contribution to force transmission plausibly supports this broad movement annotation; no motor, neurotransmission or muscle-intrinsic activity is inferred. Supporting Evidence: |
| GO:0050905 neuromuscular process | IEA GO_REF:0000107 | UNDECIDED | Reason: Mouse musculoskeletal pathology and weakness are real, but the historical neuromuscular-process scope is not resolved by tendon abnormalities alone. The complete relevant neuromuscular assay/transfer premise was not independently established. |
| GO:0051216 cartilage development | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: COMP matrix organization and actual mutant growth-plate architecture support a contextual cartilage-development contribution. Retain that broad existing structural-process interpretation without asserting direct chondrocyte fate instruction or ignoring intracellular mutant toxicity. |
| GO:0051260 protein homooligomerization | IEA GO_REF:0000107 | ACCEPT | Reason: Full-length COMP forms homooligomeric pentamers through its N-terminal region. This electronic assertion is consistent with target assembly data; the shorter official isoform retains that region in sequence, but equal oligomerization competence has not been tested. |
| GO:0055001 muscle cell development | IEA GO_REF:0000107 | UNDECIDED | Reason: The source literature documents mild myopathy localized near force-transmission structures, with tendon pathology as a proposed origin. That is not yet a resolved direct muscle-cell-development contribution by COMP; the historical electronic transfer remains uncertain. |
| GO:0060173 limb development | IEA GO_REF:0000107 | UNDECIDED | Reason: Mouse limb shortening and matrix defects are positive developmental phenotypes, while the human localization study leaves functional limb participation speculative. These different sources do not uniquely establish the normal performed limb-development step behind this retired transfer. |
| GO:0060349 bone morphogenesis | IEA GO_REF:0000107 | UNDECIDED | Reason: Current mouse donor records include bone morphogenesis, but the source-specific full PMID18191556 experiment was not recovered. COMP dysplasia and cartilage structure make the process plausible without settling the exact historical transfer. |
| GO:0070527 platelet aggregation | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The thrombin study reports inhibition of thrombin-induced platelet aggregation and related endpoints. Preserve the existing contextual process association with its inhibitory direction and abstract-only detailed-assay limit; COMP is not claimed to initiate platelet aggregation. Supporting Evidence: |
| GO:0090398 cellular senescence | IEA GO_REF:0000107 | UNDECIDED | Reason: The current donor has a cellular-senescence record from PMID27498005, but its full experimental premise was not recovered. Retain uncertainty rather than infer a senescence program from generic vascular or stress effects. |
| GO:0097084 vascular associated smooth muscle cell development | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: COMP–BMP2 interaction and modulation of the osteochondrogenic transition of vascular smooth-muscle cells are explicitly reported in the donor abstract. Retain this contextual phenotype-maintenance contribution, not an assertion of a universal vascular differentiation instruction. Supporting Evidence: |
| GO:0098868 bone growth | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The mouse mutant growth and cartilage-architecture phenotypes support a broad secondary contribution to bone growth through connective-tissue organization. This does not assert COMP catalyzes mineral deposition or that every mutation has the same mechanism. Supporting Evidence: |
| GO:1900047 negative regulation of hemostasis | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: The donor study explicitly tests clotting/bleeding and thrombin-dependent platelet endpoints, with direct COMP–thrombin binding and an inhibitory result. This supports negative regulation of hemostasis at the reported mouse/biochemical scope; full detailed Methods were unavailable. Supporting Evidence: |
| GO:1902732 positive regulation of chondrocyte proliferation | IEA GO_REF:0000107 | UNDECIDED | Reason: Reduced proliferation in COMP-mutant growth plates does not by itself distinguish positive physiological regulation from toxic-mutant or matrix-secondary effects. Other current donor proliferation records were not fully recovered; keep the historical specific positive-regulation inference unresolved. Supporting Evidence: |
| GO:1990079 cartilage homeostasis | IEA GO_REF:0000107 | KEEP AS NON CORE | Reason: COMP contributes to cartilage matrix maintenance, but current donor and target disease experiments mix structural disruption with mutant secretion/stress effects. Retain cartilage homeostasis as contextual biology; no universal requirement for normal gross cartilage or separate homeostatic core is inferred. |
| GO:0031012 extracellular matrix | IDA PMID:18285447 Extracellular matrix retention of thrombospondin 1 is contro... | ACCEPT | Reason: ECM localization is independently established for COMP. The cited abstract foregrounds TSP1 and a conserved family mechanism, and complete COMP-specific experiments were not recovered; that limited access is not evidence of a wrong-gene IDA annotation. Retain the supported location with explicit curator deference. |
| GO:0005576 extracellular region | IDA PMID:32747625 Author Correction: Mutations in COMP cause familial carpal t... | ACCEPT | Reason: The GOA source points to an affiliation correction, which is preserved unchanged. Its verified original PMID32686688 directly tests full-length human COMP secretion and extracellular deposition, supporting this location. The correction itself contains no localization experiment. Supporting Evidence: |
| GO:0051260 protein homooligomerization | IDA PMID:32747625 Author Correction: Mutations in COMP cause familial carpal t... | ACCEPT | Reason: The affiliation notice is not experimental evidence, but its verified original directly examines human full-length COMP complex formation. A separate purified coiled-coil construct is rat-derived and is not mislabeled human full length. These observations support homooligomerization without equating an isolated signature-domain crystal contact with the native pentamer. Supporting Evidence: |
| GO:1990079 cartilage homeostasis | IDA PMID:32747625 Author Correction: Mutations in COMP cause familial carpal t... | KEEP AS NON CORE | Reason: The corrected original establishes disease-associated matrix/secretion effects, including R718W and V66E contexts. These support a cartilage-maintenance contribution, but toxic-mutant disease and tendon-focused assays do not establish universal normal cartilage necessity. Preserve the original notice identifier and use the verified original for biological support. Supporting Evidence: |
| GO:0043394 proteoglycan binding | IDA PMID:29030641 Lubricin binds cartilage proteins, cartilage oligomeric matr... | KEEP AS NON CORE | Reason: Human cartilage in-situ proximity ligation detects COMP–lubricin association with antibody-omission controls. The same paper’s purified solid-phase experiments concern lubricin with fibronectin/collagen II, not a new purified COMP affinity measurement. Retain the curated proteoglycan-binding context without asserting that COMP is uniquely required to anchor lubricin. Supporting Evidence: |
| GO:0005201 extracellular matrix structural constituent | RCA PMID:20551380 Proteomics characterization of extracellular space component... | ACCEPT | Reason: Original human aorta Table I and Table II identify COMP in NaCl and guanidine extracellular fractions, respectively. Detection alone is not a mechanics assay, but it agrees with the independent ligand/assembly evidence for an ECM structural glycoprotein. The general structural term is preferable to importing collagen-specific tensile-strength terminology. |
| GO:0005576 extracellular region | HDA PMID:27068509 Extracellular matrix remodelling in response to venous hyper... | ACCEPT | Reason: The curated human vein HDA observation is consistent with COMP acting in the extracellular matrix, the established primary location. The actual study analyzes salt/SDS/guanidine fractions, but its target-specific supplemental COMP row remains unrecovered; this acceptance defers to the curated detection and independent matrix evidence rather than claiming an inspected target table. No remodeling mechanism or functional vesicle role is inferred. Supporting Evidence: |
| GO:0005615 extracellular space | HDA PMID:20551380 Proteomics characterization of extracellular space component... | MODIFY | Reason: The original aorta tables positively detect COMP in extracellular extracts. Preserve the retired GOA identifier, replacing the obsolete extracellular-space term with current extracellular region; no false localization or disappearance of the protein is inferred. Proposed replacements: extracellular region Supporting Evidence: |
| GO:0031012 extracellular matrix | HDA PMID:20551380 Proteomics characterization of extracellular space component... | ACCEPT | Reason: Original Table I/II positively identify COMP in human aortic extracellular extracts, with two and four unique peptides respectively. The tissue extraction and database-search Methods support matrix association, consistent with independently established COMP matrix function; no splice-product resolution is claimed. Supporting Evidence: |
| GO:0070062 extracellular exosome | HDA PMID:23533145 In-depth proteomic analyses of exosomes isolated from expres... | KEEP AS NON CORE | Reason: The original study isolates extracellular vesicle/exosome-enriched fractions from human expressed-prostatic-secretions urine using differential centrifugation and sucrose processing. Its exact COMP supplemental target row was not recovered. Preserve the experimental cargo-associated localization with this access limit, without asserting an intrinsic exosome-biogenesis role or rejecting it because COMP is an ECM protein. Supporting Evidence: |
| GO:0002020 protease binding | IPI PMID:18485748 Inhibition of ADAMTS-7 and ADAMTS-12 degradation of cartilag... | KEEP AS NON CORE | Reason: This retired aggregate protease-binding assertion is replaced in current GOA by two literal ADAMTS7/ADAMTS12 records. Both protease classes and the positive COMP association are retained; retirement denotes the source grouping change, not loss of binding. Supporting Evidence: |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2424252 | ACCEPT | Reason: The current Reactome event and actual P49747 component paths place the COMP pentamer extracellularly on both input and product sides. This is the established location of its primary matrix role, so the broad location is accepted consistently with the secretion and matrix studies. The event is pathway-curated context, not an independent assay of every listed ligand interaction. Supporting Evidence: Reactome:R-HSA-2424252 |
| GO:0005576 extracellular region | TAS Reactome:R-HSA-2426259 | ACCEPT | Reason: The receptor-interaction event and actual extracellular P49747 component paths agree with the established location of COMP matrix activity. Accept the location while distinguishing the direct integrin-attachment evidence from uninspected receptor combinations; the event title does not establish a separate signaling mechanism or CD47-binding assay. |
| GO:0005509 calcium ion binding | IDA PMID:10852928 Cartilage oligomeric matrix protein is a calcium-binding pro... | ACCEPT | Reason: Direct calcium-binding experiments with expressed wild-type COMP and the MUT3 disease variant support this activity. Calcium-dependent conformation contributes to the primary matrix role, making ACCEPT consistent with the electronic calcium annotation and integrated structural core. This is not calcium signaling and does not establish equal stoichiometry for both official splice products. The complete abstract is available; the full original Methods remain unrecovered. Supporting Evidence: PMID:10852928 wild-type COMP, when purified in the presence of calcium, is a calcium-binding protein. |
| GO:0005509 calcium ion binding | IDA PMID:11084047 Mutations in cartilage oligomeric matrix protein causing pse... | ACCEPT | Reason: The complete original abstract directly reports calcium binding by the type-3-repeat fragment and changes in disease variants, alongside collagen I/II/IX binding. This calcium-dependent structural chemistry contributes to the established primary matrix role, so it is accepted consistently with the other calcium annotations. Full Methods remain unrecovered: preserve the experimental curator and avoid finer control, equal-splice-product or universal stoichiometry claims. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:12225811 Matrix-matrix interaction of cartilage oligomeric matrix pro... | MODIFY | Reason: The complete original abstract explicitly reports concentration-dependent, saturable COMP–fibronectin binding by solid-phase assay and maps the interacting regions. Fibronectin binding is the exact measured ligand specificity and is more informative here than general extracellular matrix binding. Its current typed ancestry does not make it a child of GO:0050840. The abstract does not itself specify ionic requirements; no simultaneous multi-ligand bridge, untested interface or native processing condition is inferred. Proposed replacements: fibronectin binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:15075323 Interactions between the cartilage oligomeric matrix protein... | MODIFY | Reason: The retired aggregate matrilin interaction is biologically positive. MATN1/MATN3/MATN4 are ECM components; the original abstract reports copurification/co-immunoprecipitation and purified binding. Current individual source records preserve each partner, while this historical aggregate retains the same matrix-component class. Proposed replacements: extracellular matrix binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:16051604 Cartilage oligomeric matrix protein/thrombospondin 5 support... | MODIFY | Reason: The retired aggregate integrin interaction is more precisely represented by integrin binding in cell-matrix adhesion because attachment was tested on COMP-coated substrate. Retain the separate calcium-replete alpha5beta1 versus calcium-depleted-and-reduced alphavbeta3 conditions. Proposed replacements: integrin binding involved in cell-matrix adhesion Supporting Evidence: |
| GO:0005518 collagen binding | IDA PMID:11084047 Mutations in cartilage oligomeric matrix protein causing pse... | ACCEPT | Reason: The source directly reports COMP binding to collagens I, II and IX, and the later pentamer-versus-monomer experiments connect collagen engagement to performed fibril organization. Collagen binding is a molecular component of the primary matrix-organizing role, so it is accepted within the existing single structural core. The acellular assay does not require a collagen-V-specific or cell-matrix-adhesion child. Full original Methods remain unavailable; no finer binding-site or equal-splice-product claim is added. Supporting Evidence: PMID:11084047 In addition to collagen I and II, collagen IX also binds to COMP with high affinity. |
| GO:0008201 heparin binding | IDA PMID:17588949 Interaction of cartilage oligomeric matrix protein/thrombosp... | KEEP AS NON CORE | Reason: Actual affinity co-electrophoresis establishes heparin binding, with heparin preparations and COMP conformation affecting affinity. Retain this ligand activity as contextual matrix chemistry; it is distinct from the free heparan-sulfate assay and does not establish binding to every intact HSPG. Supporting Evidence: |
| GO:0043066 negative regulation of apoptotic process | IDA PMID:17993464 Cartilage oligomeric matrix protein protects cells against d... | KEEP AS NON CORE | Reason: Human cell experiments show reduced apoptotic endpoints, reduced active caspase-3 and increased IAP proteins; original Figure 8/Results include XIAP and survivin knockdown controls. This supports contextual survival activity. RGD sensitivity does not identify a particular integrin, and neither direct COMP–IAP binding nor direct caspase inhibition was demonstrated. Full-paper access remains partial. Supporting Evidence: PMID:17993464 down-regulation of survivin and XIAP protein levels by small interfering RNAs blocks the ability of COMP to enhance survival. |
| GO:0043395 heparan sulfate proteoglycan binding | IDA PMID:17588949 Interaction of cartilage oligomeric matrix protein/thrombosp... | MODIFY | Reason: Actual affinity co-electrophoresis tested free bovine-kidney heparan sulfate. The original HSPG term requires a proteoglycan, whereas the separately tested whole proteoglycan was aggrecan with chondroitin/keratan sulfate. Replace with heparan sulfate binding, preserving the positive glycosaminoglycan interaction rather than asserting universal HSPG nonbinding. Proposed replacements: heparan sulfate binding Supporting Evidence: |
| GO:0060173 limb development | IDA PMID:16542502 Cartilage oligomeric matrix protein is involved in human lim... | UNDECIDED | Reason: Controlled human embryonic localization identifies COMP in developing cartilage and joint regions, but the actual Methods/Results do not test an effect on limb formation and the Discussion leaves that role speculative. Matrix organization makes participation plausible; the unresolved issue is the performed developmental contribution, not absence of localization or a requirement for fate instruction. Supporting Evidence: |
| GO:0009887 animal organ morphogenesis | TAS PMID:7713493 Characterization of human and mouse cartilage oligomeric mat... | UNDECIDED | Reason: The complete abstract of this sequence/structure paper supports COMP identity and matrix composition, but does not establish the stated organ-morphogenesis participation. Full original recovery failed. Preserve the TAS assertion as unresolved rather than remove it from title/abstract absence or infer every developmental process from matrix expression. Supporting Evidence: |
| GO:0001501 skeletal system development | TAS PMID:7670472 Pseudoachondroplasia and multiple epiphyseal dysplasia due t... | KEEP AS NON CORE | Reason: The original human genetics abstract links COMP mutations to skeletal dysplasia, and later matrix/animal experiments provide a plausible structural developmental contribution. Retain this broad existing process as contextual rather than a defining molecular function; genetics alone does not prove wild-type indispensability or a specific developmental step. |
| GO:0005201 extracellular matrix structural constituent | TAS PMID:7713493 Characterization of human and mouse cartilage oligomeric mat... | ACCEPT | Reason: COMP is an extracellular matrix glycoprotein with performed matrix-organization activity. The original sequence/structural abstract is consistent with this role and independent fibrillogenesis experiments provide direct functional support. The general structural term is supported without assuming collagen-like tensile mechanics. |
| GO:0005509 calcium ion binding | TAS PMID:7670472 Pseudoachondroplasia and multiple epiphyseal dysplasia due t... | ACCEPT | Reason: The original genetics abstract identifies mutations in a calcium-binding motif and proposes a structural role. Independent direct calcium-binding and signature-domain structural evidence establish the same primary structural chemistry, supporting ACCEPT without misrepresenting this TAS source as a purified binding assay or inferring equal activity of both splice products. |
| GO:0031012 extracellular matrix | TAS PMID:7713493 Characterization of human and mouse cartilage oligomeric mat... | ACCEPT | Reason: The original structural/sequence abstract identifies a cartilage matrix pentamer, and independent human tissue proteomics confirms extracellular matrix presence. This is core spatial context; no complete original Methods read is claimed. |
| GO:0002020 protease binding | IPI PMID:18485748 Inhibition of ADAMTS-7 and ADAMTS-12 degradation of cartilag... | KEEP AS NON CORE | Reason: The literal current source identifies ADAMTS12, a protease, as the positive COMP-binding counterpart. The original selected text and antecedent binding study support this class. Preserve the exact record even though the current IntAct query did not return the pair; COMP is not assigned protease activity. |
| GO:0002020 protease binding | IPI PMID:18485748 Inhibition of ADAMTS-7 and ADAMTS-12 degradation of cartilag... | KEEP AS NON CORE | Reason: The literal current source identifies ADAMTS7, a protease, as the positive COMP-binding counterpart. The original selected text and antecedent binding study support this class. Preserve the exact record even though the current IntAct query did not return the pair; COMP is not assigned protease activity. |
| GO:0005201 extracellular matrix structural constituent | IBA GO_REF:0000033 | ACCEPT | Reason: The PAINT structural inference agrees with target matrix localization and performed collagen organization. The exact RGD donor identity is rat Comp/P35444; historical PTN placement/MSA could not be reconstructed after failed service retrieval. That limitation is recorded separately and is not treated as evidence of target divergence. Propagation Review Root cause: NO FAILURE CORE Sources checked: PANTHER:PTN000810694 UNRESOLVED The current tree/MSA service attempts timed out; no independent reconstruction of the historical IBD placement is claimed. Target/donor biological support is assessed separately. |
| GO:0005515 protein binding | IPI PMID:15075323 Interactions between the cartilage oligomeric matrix protein... | MODIFY | Reason: The literal partner is MATN3, an established extracellular matrix protein. The original study reports COMP–matrilin biochemical association and purified binding, supporting ECM component binding. Retain the exact partner and assay scope without requiring a mapped native interface or claiming simultaneous matrix bridging. Proposed replacements: extracellular matrix binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:15075323 Interactions between the cartilage oligomeric matrix protein... | MODIFY | Reason: The literal partner is MATN4, an established extracellular matrix protein. The original study reports COMP–matrilin biochemical association and purified binding, supporting ECM component binding. Retain the exact partner and assay scope without requiring a mapped native interface or claiming simultaneous matrix bridging. Proposed replacements: extracellular matrix binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:15075323 Interactions between the cartilage oligomeric matrix protein... | MODIFY | Reason: The literal partner is MATN1, an established extracellular matrix protein. The original study reports COMP–matrilin biochemical association and purified binding, supporting ECM component binding. Retain the exact partner and assay scope without requiring a mapped native interface or claiming simultaneous matrix bridging. Proposed replacements: extracellular matrix binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:16051604 Cartilage oligomeric matrix protein/thrombospondin 5 support... | MODIFY | Reason: The actual assay measures chondrocyte attachment to COMP-coated substrate and implicates ITGB3 within alphavbeta3 using blocking controls. The relevant condition is calcium-depleted and DTT-reduced COMP. Thus integrin binding involved in cell-matrix adhesion is supported; receptor immunoprecipitation profiles surface receptors and is not a purified COMP–integrin affinity assay. Proposed replacements: integrin binding involved in cell-matrix adhesion Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:16051604 Cartilage oligomeric matrix protein/thrombospondin 5 support... | MODIFY | Reason: The actual assay measures chondrocyte attachment to COMP-coated substrate and implicates ITGA5 within alpha5beta1 using blocking controls. The relevant condition is calcium-replete COMP. Thus integrin binding involved in cell-matrix adhesion is supported; receptor immunoprecipitation profiles surface receptors and is not a purified COMP–integrin affinity assay. Proposed replacements: integrin binding involved in cell-matrix adhesion Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:17588949 Interaction of cartilage oligomeric matrix protein/thrombosp... | MODIFY | Reason: The source record names aggrecan (bovine P13608); aggrecan belongs to the chondroitin-sulfate-proteoglycan class. Actual main-text assays tested whole Sigma aggrecan, distinct from free bovine-kidney heparan sulfate. Preserve the accession without claiming that two separate species preparations were documented by the Methods. Proposed replacements: chondroitin sulfate proteoglycan binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:17588949 Interaction of cartilage oligomeric matrix protein/thrombosp... | MODIFY | Reason: The source record names aggrecan (human P16112); aggrecan belongs to the chondroitin-sulfate-proteoglycan class. Actual main-text assays tested whole Sigma aggrecan, distinct from free bovine-kidney heparan sulfate. Preserve the accession without claiming that two separate species preparations were documented by the Methods. Proposed replacements: chondroitin sulfate proteoglycan binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | KEEP AS NON CORE | Reason: Three positive HuRI yeast two-hybrid records identify the exact 144-residue human CYSRT1 product and COMP. Current partner identity does not establish a more specific applicable binding class. Preserve the positive assay without inventing a matrix, antimicrobial or physiological colocalization role for COMP. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MODIFY | Reason: Three positive HuRI records identify the exact 68-residue ATOX1 product. Its reviewed FUNCTION and GO:0001228 ISS annotation infer a copper-dependent sequence-specific transcription-factor class from mouse evidence, distinct from copper delivery; a direct human transcription-factor assay was not established. This class refinement describes the tested counterpart, not copper-dependent transcription by COMP or demonstrated native nuclear encounter. Proposed replacements: RNA polymerase II-specific DNA-binding transcription factor binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MODIFY | Reason: The exact 269-residue AQP1 product is a verified transmembrane water transporter, and the positive HuRI records support its association with COMP in the yeast assay. Transmembrane transporter binding describes that partner class. No channel gating/regulation, native membrane interface or secretory processing compatibility was assayed. Proposed replacements: transmembrane transporter binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MODIFY | Reason: Positive HuRI records identify canonical 354-residue OTX1, whose current official annotation establishes DNA-binding RNA-polymerase-II transcription-factor activity. Preserve the interaction as binding to that counterpart class, without asserting COMP acts in transcription or that a native extracellular–nuclear encounter occurs. Proposed replacements: RNA polymerase II-specific DNA-binding transcription factor binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | MODIFY | Reason: The exact positive Q6FHY5 MEOX2 product has 303 residues and an intact homeobox. Sequence comparison with curated 304-residue P50222 differs only by one glutamine before the identical homeobox; the established DNA-binding RNA-polymerase-II transcription-factor class is retained without claiming equal transactivation or changing the tested accession. COMP transcription and native encounter remain unestablished. Proposed replacements: RNA polymerase II-specific DNA-binding transcription factor binding Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | KEEP AS NON CORE | Reason: The three positive HuRI records identify canonical 695-residue NUFIP2. The saved RNA-associated partner identity does not establish a narrower applicable binding class for this interaction. Retain the positive assay without inferring COMP RNA processing, native compartment overlap or a mapped interface. Supporting Evidence: |
| GO:0005515 protein binding | IPI PMID:32296183 A reference map of the human binary protein interactome. | KEEP AS NON CORE | Reason: The three positive records identify canonical 304-residue SGTB. Official human and rat functions establish a co-chaperone that binds HSC70/HSP70 and regulates ATPase activity; they do not resolve SGTB itself as the nascent/unfolded-protein-binding chaperone class required by GO:0051087. Preserve the interaction without converting association with a chaperone into that class. Supporting Evidence: |
| GO:0005576 extracellular region | HDA PMID:20551380 Proteomics characterization of extracellular space component... | ACCEPT | Reason: Original human aorta tables identify COMP in NaCl/guanidine extracellular fractions, consistent with its established extracellular matrix location. Accept that broad location as part of the primary matrix role while preserving the limited HDA detection scope; peptide detection alone does not establish a remodeling mechanism. This current source remains separate from the retired obsolete-space assertion. Supporting Evidence: |
| GO:0007155 cell adhesion | IEA GO_REF:0000002 | KEEP AS NON CORE | Reason: The InterPro family/domain inference is consistent with actual attachment to COMP-coated substrate through integrins. This supports contextual cell adhesion by an extracellular ligand, without making all domain-bearing proteins equivalent or asserting COMP is an integral membrane adhesion receptor. Supporting Evidence: |
| GO:0008201 heparin binding | IEA GO_REF:0000117 | KEEP AS NON CORE | Reason: Current ARBA00027326 branches 0 and 1 match the official target InterPro/FunFam and primate lineage. Independent heparin-binding experiments support the biological claim. The present-day rule match is not a reconstruction of historical rule training, and does not generalize to every GAG/proteoglycan. Supporting Evidence: |
| GO:0035989 tendon development | IBA GO_REF:0000033 | KEEP AS NON CORE | Reason: Mouse Comp mutant tendon experiments show altered collagen fibrils, tendon dimensions and mechanics, providing a structural developmental premise. Historical PTN008302756 placement and the zebrafish donor source were not independently reconstructed; donor count is not used against the IBA. Retain contextual tendon development without claiming it is the sole or universal route of tendon formation. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: PANTHER:PTN008302756 UNRESOLVED The current tree/MSA service attempts timed out; no independent reconstruction of the historical IBD placement is claimed. Target/donor biological support is assessed separately. |
| GO:0140149 non-collagenous component of interstitial matrix | TAS PMID:36399478 MatrisomeDB 2.0: 2023 updates to the ECM-protein knowledge d... | ACCEPT | Reason: COMP is a noncollagenous interstitial matrix glycoprotein. MatrisomeDB is a curated database/classification source, not a new COMP localization experiment; independent human tissue detection and matrix-ligand studies support this specific structural location. |
| GO:1990341 thrombospondin complex | IEA GO_REF:0000117 | ACCEPT | Reason: Current ARBA00097011 branches 1 and 3 match the official target domain/FunFam and primate lineage. Full-length COMP homopentamer evidence supports thrombospondin-complex membership; an isolated C-terminal crystal-packing contact is not substituted for biological assembly. |
| GO:1990341 thrombospondin complex | NAS PMID:19276170 The crystal structure of the signature domain of cartilage o... | ACCEPT | Reason: The paper places COMP in the thrombospondin family and discusses its established full-length pentamer. Its actual human A225–A757 structure is monomeric in solution with nonbiological crystal packing, so the NAS complex assertion is accepted using known full-length membership, not by labeling that crystal trimer physiological. |
| GO:1990341 thrombospondin complex | TAS PMID:15094109 The thrombospondins. | ACCEPT | Reason: The review describes the thrombospondin family, and independent COMP identity/full-length assembly support its specific complex membership. This TAS classification does not imply every family member has the same stoichiometry or that both COMP splice products were assayed. |
Loading supporting content…
Download this section (compressed HTML)Q: Does the 704-residue splice product lacking canonical residues129–181 fold, secrete and assemble like displayed COMP, and which calcium-dependent ligand activities are retained?
Q: Which human limb-development step is directly supported by COMP matrix activity beyond its patterned embryonic localization and disease association?
Q: How do COMP conformation, calcium occupancy and reduction state alter integrin usage in native tissue, compared with coated-substrate assays?
Q: Which historical developmental transfers reflect normal matrix participation, and which instead describe toxic-mutant stress or secondary tissue pathology?
Q: Do the screen-positive intracellular counterparts encounter COMP physiologically, and how do secretory processing and product identity constrain these interactions?
Q: What target-specific exosome and venous-proteomics records underlie the current COMP detections, and can the original supplements resolve their product and fraction scope?
Experiment: Compare purified displayed and shorter COMP products for secretion, pentamer formation, calcium binding and collagen/heparan-sulfate interactions using matched expression and folding controls.
Experiment: Test matrix-mediated developmental contribution with temporally controlled COMP perturbation and structural rescue that separates extracellular assembly from intracellular mutant stress.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)