Contactin-2 (CNTN2, TAG-1 or TAX-1) is a glycosylphosphatidylinositol-anchored neural cell-surface adhesion protein with six immunoglobulin-like and four fibronectin type III domains. Homophilic extracellular contacts support cell adhesion, while interactions with other neural surface proteins help organize axonal and synaptic membrane domains. In myelinated axons, contactin-2 cooperates with CASPR2 to maintain juxtaparanodal potassium-channel clusters. Neuronal and glial expression contributes to axon organization and axon–glia contacts. Human loss-of-function variants have been associated with epilepsy, and rodent studies reveal additional developmental and metabolic phenotypes.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0002021 response to dietary excess | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: The mouse ortholog has a diet-dependent energy-balance phenotype. Reason: The exact Q61330 donor has an IMP record from the diet study. Tag1-null mice resisted weight gain on a high-fat/simple-carbohydrate diet and differed in body composition and food consumption. This supports a peripheral physiological annotation, while the experiments do not identify a direct CNTN2 nutrient-sensing mechanism or establish the hypothalamic step. It is kept outside the adhesion core. |
| GO:0002023 reduction of food intake in response to dietary excess | IEA GO_REF:0000107 | UNDECIDED | Summary: The specific reduction-of-intake process is not resolved by the donor phenotype. Reason: The mouse study reports lower food intake in null mice early after high-fat feeding, with a later nonsignificant trend, and lower rather than increased energy expenditure after prolonged feeding. These observations do not identify CNTN2 as performing the defined reduction of feeding after detection of dietary excess. The original positive donor assertion and the real diet phenotype are retained while that precise functional direction remains unresolved. |
| GO:0005515 protein binding | IPI PMID:32814053 Interactome Mapping Provides a Network of Neurodegenerative ... | MODIFY | Summary: The positive APP interaction can be described by the partner adhesion class. Reason: Three exact IntAct records from the neurodegenerative-disease interaction screen support APP–CNTN2 association: array, prey-pooling and validated two-hybrid assays. APP is independently established as a cell adhesion protein by ectodomain bead aggregation and cell-contact experiments. Cell adhesion molecule binding is therefore a defensible partner-class refinement, without inferring a native APP–CNTN2 adhesion mechanism or amyloid-processing role. The recorded APP 670–671 mutation feature and neutral roles in the validated record remain explicit. Proposed replacements: cell adhesion molecule binding |
| GO:0005886 plasma membrane | IBA GO_REF:0000033 | ACCEPT | Summary: Plasma-membrane activity is supported by target experiments and the inherited family annotation. Reason: Human CNTN2 constructs are displayed at the cell surface and mediate adhesion. This independently corroborates the PAINT location. The target among the descendant sources is legitimate experimental grounding, and donor number is not used to judge the ancestral assertion. The historical tree and IBD placement were not reconstructed. Propagation Review Root cause: NO FAILURE CORE Sources checked: FB:FBgn0000635 SUPPORTS TRANSFER Official identity resolves to Fas2. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0004839 SUPPORTS TRANSFER Official identity resolves to otk. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0028482 SUPPORTS TRANSFER Official identity resolves to bdl. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0037240 UNRESOLVED Official identity resolves to Cont. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. MGI:MGI:104750 SUPPORTS TRANSFER Official identity resolves to Nrcam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:108077 UNRESOLVED Official identity resolves to Nptn. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. MGI:MGI:1196281 SUPPORTS TRANSFER Official identity resolves to Dscam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:1858223 SUPPORTS TRANSFER Official identity resolves to Cntn6. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:2444710 SUPPORTS TRANSFER Official identity resolves to Prtg. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:3042287 UNRESOLVED Official identity resolves to Cntn5. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. MGI:MGI:88208 SUPPORTS TRANSFER Official identity resolves to Bsg. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. PANTHER:PTN001296633 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. RGD:2220 SUPPORTS TRANSFER Official identity resolves to Bsg. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. RGD:620296 SUPPORTS TRANSFER Official identity resolves to Nptn. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. RGD:621067 SUPPORTS TRANSFER Official identity resolves to Emb. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. UniProtKB:P17790 SUPPORTS TRANSFER Official identity resolves to BSG. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. UniProtKB:P35613 SUPPORTS TRANSFER Official identity resolves to BSG. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. UniProtKB:Q02246 SUPPORTS TRANSFER CNTN2 is the target and a legitimate descendant source. Human surface-expression and synaptic-location evidence is independently assessed; for synapse organization the exact organizing step remains unresolved. Self inclusion is not circular, and the historical placement remains uninspected. UniProtKB:Q8TD84 SUPPORTS TRANSFER Official identity resolves to DSCAML1. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. WB:WBGene00004370 UNRESOLVED Official identity resolves to rig-3. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. ZFIN:ZDB-GENE-000209-4 SUPPORTS TRANSFER Official identity resolves to robo3. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. ZFIN:ZDB-GENE-020809-1 SUPPORTS TRANSFER Official identity resolves to robo4. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. ZFIN:ZDB-GENE-050310-7 SUPPORTS TRANSFER Official identity resolves to dscama. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. |
| GO:0005886 plasma membrane | IEA GO_REF:0000044 | ACCEPT | Summary: The mapped plasma-membrane location is well supported. Reason: The fetched UniProt location maps through SL-0039. GPI anchoring and human surface-expression experiments independently support the membrane assignment; the extracellular adhesion domains do not imply a cytosolic protein. |
| GO:0005886 plasma membrane | ISS GO_REF:0000024 | ACCEPT | Summary: The rat-based plasma-membrane transfer agrees with direct human evidence. Reason: The named rat protein P22063 is Cntn2. The original fractionation study directly detects rat TAG-1 in synaptic plasma membranes, and human constructs independently reach the cell surface. The rat tissue experiment is not relabeled as a human brain assay. |
| GO:0005886 plasma membrane | TAS PMID:8425542 cDNA cloning, structural features, and eucaryotic expression... | ACCEPT | Summary: Human TAG-1 is directly observed at the cell surface. Reason: The original abstract explicitly describes human cDNA expression in COS1 and mouse myeloma cells and surface immunoreactivity, with GPI attachment. Human construct identity is distinct from the host-cell species. Full text of this original study remains unavailable, but the positive abstract directly supports the location and later human experiments corroborate it. |
| GO:0007155 cell adhesion | NAS PMID:8425542 cDNA cloning, structural features, and eucaryotic expression... | ACCEPT | Summary: Cell adhesion is a directly demonstrated function of human CNTN2. Reason: The original NAS source introduces the cloned human adhesion molecule but its inspected abstract does not provide a direct adhesion assay. Independent human full-length and domain-mutant experiments demonstrate cell aggregation, homophilic binding and loss of adhesion after interface disruption with preserved surface expression. These establish the function while preserving the original source assertion. Supporting Evidence: PMID:39702996 cell adhesion could be observed by transfection |
| GO:0007156 homophilic cell-cell adhesion | IBA GO_REF:0000033 | ACCEPT | Summary: Homophilic adhesion is supported independently of the PAINT inference. Reason: Human ectodomain binding, cryo-EM and controlled cell-aggregation assays establish homophilic CNTN2 contacts. Current descendant-source records have positive homophilic-adhesion grounding. The newer Ig3–Ig6 interface experiments are stronger functional evidence than earlier crystal-packing models, without implying that all native assemblies have one fixed stoichiometry. Propagation Review Root cause: NO FAILURE CORE Sources checked: FB:FBgn0000635 SUPPORTS TRANSFER Official identity resolves to Fas2. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0028482 SUPPORTS TRANSFER Official identity resolves to bdl. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0265296 SUPPORTS TRANSFER Official identity resolves to Dscam2. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. PANTHER:PTN001296633 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. RGD:620296 SUPPORTS TRANSFER Official identity resolves to Nptn. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. UniProtKB:Q8TD84 SUPPORTS TRANSFER Official identity resolves to DSCAML1. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. |
| GO:0007411 axon guidance | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Axon guidance is a supported developmental context for CNTN2 adhesion. Reason: Positive current source annotations support the inherited process, and the mouse TAG-1 study reports premature sensory-axon entry and altered responses to ventral-cord repellents. Human TAG-1 substrates also influence mouse retinal-axon trajectories in stripe assays. These are developmental roles of the surface-recognition system, not evidence that CNTN2 itself is a diffusible guidance ligand. The historical PAINT placement remains uninspected. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: FB:FBgn0000635 UNRESOLVED Official identity resolves to Fas2. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. FB:FBgn0004839 SUPPORTS TRANSFER Official identity resolves to otk. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0010473 SUPPORTS TRANSFER Official identity resolves to tutl. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0033159 SUPPORTS TRANSFER Official identity resolves to Dscam1. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:104750 SUPPORTS TRANSFER Official identity resolves to Nrcam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:1196281 SUPPORTS TRANSFER Official identity resolves to Dscam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. PANTHER:PTN001296633 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. WB:WBGene00016354 UNRESOLVED Official identity resolves to rig-6. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. ZFIN:ZDB-GENE-000209-4 SUPPORTS TRANSFER Official identity resolves to robo3. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. |
| GO:0009986 cell surface | IEA GO_REF:0000107 | ACCEPT | Summary: Cell-surface localization is established for mouse and human CNTN2. Reason: The mouse donor study explicitly reports staining of developing axons and the surface of neuronal somata. Original human expression and later surface-controlled adhesion experiments independently corroborate the transfer. |
| GO:0022604 regulation of cell morphogenesis | IEA GO_REF:0000107 | KEEP AS NON CORE | Summary: CNTN2 influences neuronal and glial morphology in a developmental context. Reason: The mapped mouse study measures altered retinal-axon caliber, cytoskeletal organization and neighboring astroglial processes in Tag1-null nerves. Human TAX-expressing cells and purified human TAG-1 substrates also affect mouse retinal-axon adhesion and fasciculation. The broad morphology-regulation term is retained without assigning a cell-intrinsic cytoskeletal activity or a universal morphogenesis mechanism. |
| GO:0030246 carbohydrate binding | IEA GO_REF:0000107 | MODIFY | Summary: Lewis-x recognition supports a more specific trisaccharide-binding term. Reason: The mapped paper directly tests human TAX-1: Lewis-x–BSA pulls it down from transfected CHO lysates, whereas BSA and N-acetyllactosamine controls do not. Lewis-x is a trisaccharide, matching the current GO definition. This is a secondary ligand-recognition capability; the lysate assay is not purified-protein affinity measurement, and binding is not inferred merely from CNTN2 glycosylation. Proposed replacements: trisaccharide binding |
| GO:0030424 axon | IBA GO_REF:0000033 | ACCEPT | Summary: Axonal localization is consistent with family and target evidence. Reason: The inherited family annotation is supported by positive current descendant records. Mouse protein staining along developing retinal and other axons, together with human neural surface expression, supports the location. This current evidence check does not reconstruct the historical ancestral node. Propagation Review Root cause: NO FAILURE CORE Sources checked: FB:FBgn0000635 SUPPORTS TRANSFER Official identity resolves to Fas2. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0004839 SUPPORTS TRANSFER Official identity resolves to otk. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0033159 SUPPORTS TRANSFER Official identity resolves to Dscam1. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:104750 SUPPORTS TRANSFER Official identity resolves to Nrcam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:1196281 SUPPORTS TRANSFER Official identity resolves to Dscam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. PANTHER:PTN001296633 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. RGD:3209 SUPPORTS TRANSFER Official identity resolves to Nrcam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. UniProtKB:Q03696 SUPPORTS TRANSFER Official identity resolves to Q03696. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. WB:WBGene00004370 SUPPORTS TRANSFER Official identity resolves to rig-3. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. WB:WBGene00016354 SUPPORTS TRANSFER Official identity resolves to rig-6. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. |
| GO:0030424 axon | IEA GO_REF:0000107 | ACCEPT | Summary: The mouse-based axonal transfer has direct protein-localization support. Reason: The Q61330 donor is mouse Cntn2, and the source directly labels TAG-1 on retinal ganglion-cell neurites and optic fibers. The developmental expression window is preserved rather than treated as uniform expression in all mature axons. |
| GO:0033268 node of Ranvier | IEA GO_REF:0000107 | MODIFY | Summary: Juxtaparanode is the precise location established by direct protein staining. Reason: The mapped 2003 paper directly places TAG-1 at juxtaparanodes. Its separate X-gal node/myelin images report a tau-lacZ expression marker and do not by themselves place the TAG-1 protein in the nodal sodium-channel gap. The juxtaparanodal term captures the experimentally resolved mature localization; transient developmental or ectopic paranodal staining is not claimed to be impossible. Proposed replacements: juxtaparanode region of axon |
| GO:0033268 node of Ranvier | ISS GO_REF:0000024 | MODIFY | Summary: Juxtaparanode is the precise location established by direct protein staining. Reason: The mapped 2003 paper directly places TAG-1 at juxtaparanodes. Its separate X-gal node/myelin images report a tau-lacZ expression marker and do not by themselves place the TAG-1 protein in the nodal sodium-channel gap. The juxtaparanodal term captures the experimentally resolved mature localization; transient developmental or ectopic paranodal staining is not claimed to be impossible. Proposed replacements: juxtaparanode region of axon |
| GO:0043005 neuron projection | IEA GO_REF:0000107 | ACCEPT | Summary: Neuronal projections contain TAG-1 protein. Reason: The exact mouse donor study directly stains developing retinal neurites and axonal tracts. This supports the broader neuronal-projection location without requiring every projection type to express CNTN2. |
| GO:0043025 neuronal cell body | IEA GO_REF:0000107 | ACCEPT | Summary: Neuronal cell-body localization has positive developmental evidence. Reason: The original mouse abstract explicitly reports surface staining of the parent somata of differentiating neurons and strong labeling of somata in several migratory streams. This is a developmental surface-associated distribution, not a claim that CNTN2 is a soluble cell-body protein. |
| GO:0043194 axon initial segment | IEA GO_REF:0000107 | ACCEPT | Summary: Axon-initial-segment localization is supported with its actual species scope. Reason: The mapped article detects TAG-1 immunoreactivity at the axon initial segment of rat hippocampal cultures. That actual rat experiment is kept distinct from its current mouse annotation mapping. The same paper shows that AIS channel clustering uses mechanisms distinct from juxtaparanodes; TAG-1 localization is not assigned PSD-93 scaffold activity. |
| GO:0043209 myelin sheath | IEA GO_REF:0000107 | ACCEPT | Summary: A glial/myelin-associated CNTN2 pool is supported. Reason: The original source includes a myelinating-cell expression reporter, which is not sufficient alone for protein localization. Independent protein-specific immunostaining and myelin fractionation in mouse knockout animals expressing rat TAG-1 specifically in oligodendrocytes support the glial myelin-associated pool. The transgene rescues CNS juxtaparanodal organization; peripheral ectopic localization and failed rescue are kept separate. |
| GO:0043209 myelin sheath | ISS GO_REF:0000024 | ACCEPT | Summary: A glial/myelin-associated CNTN2 pool is supported. Reason: The original source includes a myelinating-cell expression reporter, which is not sufficient alone for protein localization. Independent protein-specific immunostaining and myelin fractionation in mouse knockout animals expressing rat TAG-1 specifically in oligodendrocytes support the glial myelin-associated pool. The transgene rescues CNS juxtaparanodal organization; peripheral ectopic localization and failed rescue are kept separate. |
| GO:0044224 juxtaparanode region of axon | IEA GO_REF:0000107 | ACCEPT | Summary: CNTN2 is a juxtaparanodal adhesion component. Reason: Direct rodent TAG-1 staining flanks paranodal markers with CASPR2 and Kv1 channels. Knockout and CNS glial rescue experiments independently corroborate this compartment. These observations support ortholog transfer without presenting the tissue staining as a native human experiment. |
| GO:0044224 juxtaparanode region of axon | ISS GO_REF:0000024 | ACCEPT | Summary: CNTN2 is a juxtaparanodal adhesion component. Reason: Direct rodent TAG-1 staining flanks paranodal markers with CASPR2 and Kv1 channels. Knockout and CNS glial rescue experiments independently corroborate this compartment. These observations support ortholog transfer without presenting the tissue staining as a native human experiment. |
| GO:0045163 clustering of voltage-gated potassium channels | IEA GO_REF:0000107 | ACCEPT | Summary: CNTN2 helps maintain potassium-channel clusters at juxtaparanodes. Reason: Tag1 loss disrupts juxtaparanodal Kv1/CASPR2 clusters despite preserved protein abundance. Rat TAG-1 expressed in mouse oligodendrocytes restores CNS clusters, and adhesion/complex experiments supply a mechanism beyond the knockout phenotype. CNTN2 organizes the surface complex; it is not a potassium-conducting pore. |
| GO:0045163 clustering of voltage-gated potassium channels | ISS GO_REF:0000024 | ACCEPT | Summary: CNTN2 helps maintain potassium-channel clusters at juxtaparanodes. Reason: Tag1 loss disrupts juxtaparanodal Kv1/CASPR2 clusters despite preserved protein abundance. Rat TAG-1 expressed in mouse oligodendrocytes restores CNS clusters, and adhesion/complex experiments supply a mechanism beyond the knockout phenotype. CNTN2 organizes the surface complex; it is not a potassium-conducting pore. |
| GO:0045202 synapse | IBA GO_REF:0000033 | ACCEPT | Summary: Synaptic localization is supported independently of the family transfer. Reason: The current PAINT source set includes Cntn5, Cntn6 and human CNTN2. Human TAG-1-GFP accumulates at postsynaptic contacts in rodent neuronal cultures, and rat TAG-1 is present in synaptic plasma-membrane fractions. This direct evidence supports the broad synapse location even where the current donor query does not reconstruct the original experimental chain. Propagation Review Root cause: NO FAILURE CORE Sources checked: MGI:MGI:1858223 UNRESOLVED Official identity resolves to Cntn6. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. MGI:MGI:3042287 UNRESOLVED Official identity resolves to Cntn5. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. PANTHER:PTN002560628 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. UniProtKB:Q02246 UNRESOLVED CNTN2 is the target and a legitimate descendant source. Human surface-expression and synaptic-location evidence is independently assessed; for synapse organization the exact organizing step remains unresolved. Self inclusion is not circular, and the historical placement remains uninspected. |
| GO:0045202 synapse | ISS GO_REF:0000024 | ACCEPT | Summary: The rat synaptic-localization source directly includes TAG-1. Reason: Although the paper centers on CNTNAP2, its original Figure 3D and accompanying text detect both rat Cntnap2 and TAG-1 in postnatal forebrain synaptic plasma-membrane fractions with compartment markers. The positive CNTN2 experiment is retained; cofractionation alone is not proof of a physical complex. |
| GO:0045211 postsynaptic membrane | IDA PMID:26217189 Inhibitory axons are targeted in hippocampal cell culture by... | ACCEPT | Summary: Human TAG-1-GFP localizes at postsynaptic membrane contacts. Reason: The Methods explicitly identify human TAG-1-GFP. In rat hippocampal cultures, live surface labeling places its clusters on shafts and spines facing synaptophysin-positive sites. This supports the IDA location with the transfected human construct and rodent cellular context stated separately. Supporting Evidence: PMID:26217189 TAG-1-GFP clusters were facing presynaptic sites |
| GO:0045211 postsynaptic membrane | IMP PMID:26217189 Inhibitory axons are targeted in hippocampal cell culture by... | ACCEPT | Summary: Perturbation and rescue support the postsynaptic TAG-1 context. Reason: Caspr2-Fc binding is weak in Tag1-null mouse neurons and restored by human TAG-1-GFP expression. Direct surface labeling of the same construct places it at postsynaptic sites. The result supports the preserved location assertion without treating an anti-Caspr2 patient-IgG phenotype as a direct CNTN2 presynaptic-assembly assay. Supporting Evidence: |
| GO:0045444 fat cell differentiation | IEA GO_REF:0000107 | MARK AS OVER ANNOTATED | Summary: Reduced adiposity does not demonstrate adipocyte differentiation. Reason: The mapped mouse study measures body weight, body composition, food intake and energy expenditure after diet challenge. It does not measure acquisition of adipocyte identity or differentiation of precursor cells. Extending the systemic Tag1-null lean phenotype to the defined fat-cell differentiation process is too specific; the positive dietary phenotype remains in the separate response annotation. |
| GO:0050808 synapse organization | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: The inherited synapse-organization role is retained outside the demonstrated adhesion core. Reason: The PAINT assertion represents an ancestral judgment, not a pairwise donor transfer. Human CNTN2 is a legitimate descendant source; its inclusion is not circular. Current target experiments establish synaptic positioning and partner recognition, while the clinical source does not resolve the precise organizing step. The historical IBD placement has not been reconstructed, so no evolutionary failure is inferred. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: MGI:MGI:3042287 UNRESOLVED Official identity resolves to Cntn5. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. PANTHER:PTN002560628 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. UniProtKB:Q02246 UNRESOLVED CNTN2 is the target and a legitimate descendant source. Human surface-expression and synaptic-location evidence is independently assessed; for synapse organization the exact organizing step remains unresolved. Self inclusion is not circular, and the historical placement remains uninspected. |
| GO:0070593 dendrite self-avoidance | IBA GO_REF:0000033 | KEEP AS NON CORE | Summary: Dendritic self-avoidance is retained as an inherited, non-core context. Reason: The named Turtle and DSCAM-family descendants have positive experimental self-avoidance evidence; actual mouse DSCAM/DSCAML1 experiments distinguish avoidance from synaptic specificity. Human CNTN2 self-avoidance has not been directly demonstrated in the inspected sources, and its homophilic adhesion alone does not establish repulsion. These limits justify excluding this role from the core, not rejecting a PAINT node from donor identity or count. Propagation Review Root cause: NO FAILURE NON CORE Sources checked: FB:FBgn0010473 SUPPORTS TRANSFER Official identity resolves to tutl. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. The inspected donor self-avoidance experiments concern Turtle or DSCAM proteins; target CNTN2 repulsion is not inferred merely from Ig-domain similarity. FB:FBgn0033159 SUPPORTS TRANSFER Official identity resolves to Dscam1. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. The inspected donor self-avoidance experiments concern Turtle or DSCAM proteins; target CNTN2 repulsion is not inferred merely from Ig-domain similarity. MGI:MGI:1196281 SUPPORTS TRANSFER Official identity resolves to Dscam. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. The inspected donor self-avoidance experiments concern Turtle or DSCAM proteins; target CNTN2 repulsion is not inferred merely from Ig-domain similarity. MGI:MGI:2150309 SUPPORTS TRANSFER Official identity resolves to Dscaml1. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. The inspected donor self-avoidance experiments concern Turtle or DSCAM proteins; target CNTN2 repulsion is not inferred merely from Ig-domain similarity. PANTHER:PTN001296633 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. |
| GO:0071205 protein localization to juxtaparanode region of axon | ISS GO_REF:0000024 | ACCEPT | Summary: CNTN2 contributes to localization of juxtaparanodal proteins. Reason: Rodent loss-of-function and glial rescue experiments demonstrate the presence or absence of CASPR2 and Kv1 clusters at juxtaparanodes without simply changing total abundance. The term includes establishment or maintenance and matches the demonstrated surface-organizing role. |
| GO:0071206 establishment of protein localization to juxtaparanode region of axon | IEA GO_REF:0000107 | MODIFY | Summary: The broader localization term captures recruitment or retention without requiring resolved transport. Reason: The mapped mouse experiments establish altered juxtaparanodal accumulation of CASPR2 and Kv1 proteins. They do not distinguish directed delivery from capture or maintenance after arrival. Protein localization to juxtaparanode region of axon accurately includes these alternatives, whereas the original establishment term specifically denotes directed movement. Proposed replacements: protein localization to juxtaparanode region of axon |
| GO:0097090 presynaptic membrane organization | IMP PMID:23518707 Autosomal recessive cortical myoclonic tremor and epilepsy: ... | UNDECIDED | Summary: The specific presynaptic-membrane organization assertion remains unresolved. Reason: The original clinical paper supports a segregating human frameshift and epilepsy phenotype. Inspected complete Methods and Results concern clinical assessment and genetic mapping; the inspected Discussion invokes earlier juxtaparanodal work. These observations do not identify a presynaptic axon-terminal assembly step performed by CNTN2. The original IMP assertion is preserved pending the precise supporting experiment; the uninspected remainder and supplement preclude a claim of miscitation or contradiction. Supporting Evidence: |
| GO:0098631 cell adhesion mediator activity | IEA GO_REF:0000107 | MODIFY | Summary: Cell-cell adhesion mediator activity specifies the demonstrated surface function. Reason: The mouse-derived broad cell-adhesion mediator term is supported, and direct human homophilic binding and cell-aggregation assays resolve adhesion between opposing cells. The specific cell-cell mediator term captures that work. No cadherin-binding or exclusively heterotypic function is inferred. Proposed replacements: cell-cell adhesion mediator activity |
| GO:0098632 cell-cell adhesion mediator activity | IBA GO_REF:0000033 | ACCEPT | Summary: CNTN2 directly mediates cell-cell adhesion. Reason: Human full-length constructs, extracellular-domain binding and surface-controlled interface mutants establish the physical work of adhesion. Current positive descendant evidence complements this target result; unresolved historical PAINT placement is not treated as a failure. The activity is distinct from merely changing the expression of another adhesion molecule. Propagation Review Root cause: NO FAILURE CORE Sources checked: FB:FBgn0010473 SUPPORTS TRANSFER Official identity resolves to tutl. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. FB:FBgn0028482 SUPPORTS TRANSFER Official identity resolves to bdl. Current exact-term annotations contain positive experimental or curator-stated grounding, with literal record identities retained in the evidence index. This confirms current source support rather than the historical placement. MGI:MGI:104750 UNRESOLVED Official identity resolves to Nrcam. The inspected exact-term records do not independently resolve the experimental chain. This does not refute the original descendant evidence or establish circular propagation. PANTHER:PTN001296633 UNRESOLVED The original node identifier is preserved. Historical PAINT tree, alignment and IBD placement were not independently reconstructed; no failure is inferred from source count. |
Loading supporting content…
Download this section (compressed HTML)Q: Which experiment establishes the specific presynaptic membrane organization assertion linked to the clinical CNTN2 paper, and how is it distinguished from juxtaparanodal organization?
Q: Does human CNTN2 mediate dendritic self-avoidance in a defined neuronal context, and how does that relate to its directly demonstrated adhesive contacts?
Q: Does CNTN2 control a nutrient-sensing or feeding step, and can any adipocyte-differentiation effect be separated from whole-animal growth and intake changes?
Experiment: Compare surface-matched WT human CNTN2 and adhesion-interface mutants in neuron–glia cocultures, measuring CASPR2/Kv1 arrival and residence at defined membrane domains.
Hypothesis: Adhesive contacts stabilize membrane-domain complexes, with distinguishable effects on initial delivery and retention.
Experiment: Use compartment-restricted CNTN2 rescue in neuronal cultures with direct pre- and postsynaptic membrane markers and quantified partner clustering.
Hypothesis: A specific synaptic organizing contribution can be separated from surface localization and from the juxtaparanodal role.
Loading supporting content…
Download this section (compressed HTML)Loading supporting content…
Download this section (compressed HTML)