Reviewed all 43 seeded annotations and preserved both machine-derived protein isoforms. The normal fetch initially stalled during reference retrieval; its partial output was retained and the same CLI completed sequentially with a transport timeout. Falcon and the explicit perplexity-lite fallback both genuinely timed out; these notes record manual primary-source research. No provider report was fabricated.
The standing project instruction governs generic binding. This seed has no GO:0005515 rows.
CDH11 supplies the adhesive interaction itself. Existing cadherin binding is refined to cadherin binding involved in cell-cell adhesion, preserving IBA provenance. Calcium binding remains a correct, less informative supporting activity. Broad adhesion and membrane ancestors remain accepted; the core lists the informative processes and locations rather than every parent.
PMID:33811546(https://pmc.ncbi.nlm.nih.gov/articles/PMC9245547/) Methods and Fig5 distinguish CDH11-transduced mouse L cells adhering to human CDH11-Fc from E279Q patient fibroblast migration/focal-adhesion phenotypes. Neither experiment establishes direct collagen binding or an intrinsic Rho-family regulatory activity. Its genuine normal full-text cache is available.
PMID:16525026(https://pmc.ncbi.nlm.nih.gov/articles/PMC1446095/) Methods and Results/Figs1–3 establish full-length human CDH11 and tail mutants in mouse L cells. The canonical tail recruits catenins and organizes junctions. Its normal cache is abstract-only; authentic external full HTML was read, including exact construct and assay descriptions, but figure pixels and every supplement were not newly inspected. PMID:10320525(https://pubmed.ncbi.nlm.nih.gov/10320525/) distinguishes intact, tail-truncated and proteolytically secreted products. The truncated isoform is not assigned canonical beta-catenin binding. The secreted fragment justifies the separate extracellular-region row. PMID:9556063(https://pubmed.ncbi.nlm.nih.gov/9556063/) independently supports human osteoblast junctional CDH11/catenin association; its HAV-peptide perturbation is not treated as CDH11-specific.
Valve/migration PMID:26188246(https://pmc.ncbi.nlm.nih.gov/articles/PMC4841269/) was read externally at Methods and selected Results depth, with a normal abstract-only cache. WITH F1RFU7 is pig, and the paper also uses mouse and chick experiments. PMID:11450702(https://pubmed.ncbi.nlm.nih.gov/11450702/) adds mouse bone-density/mineralization evidence to the early suggestive skeletal annotations; those remain non-core rather than being promoted to a defining molecular function.
Exact QuickGO mouse donor records trace neural transfers to PMID:10860580(https://pubmed.ncbi.nlm.nih.gov/10860580/). The normal abstract supports hippocampal synaptic localization and CA1 potentiation; this is retained as ortholog-inferred context. The donor cytoplasmic observation from PMID:12139922(https://pubmed.ncbi.nlm.nih.gov/12139922/) remains unresolved. The full PMID:29112946(https://pmc.ncbi.nlm.nih.gov/articles/PMC5675431/) studies pathological hypodermal fibrosis, so the electronic transfer to normal skin development is marked over-annotated after checking the GO definition. The PMID:23533145(https://pmc.ncbi.nlm.nih.gov/articles/PMC3773505/) exosome main text was read, but its exact CDH11 table entry was not recovered; table downloads returned a preparation page, and Europe PMC reported the supplement unavailable.
Reactome events were checked individually. The R-NUL event deliberately combines human CDH11 with mouse catenins; it is not a wrong-species attribution. Product-localization events do not imply that CDH11 performs translation or cleavage of itself. The accompanying evidence JSON retains exact machine donor/participant fields and source URLs/hashes. These records corroborate provenance, not independent experimental replication.
Supporting quotations are sparse and source-specific, with no repetition in these notes. The 25-word aggregate authoring limit is not represented as a repository validation rule.
The final MF uses GO:0098641 because the adhesive ligand is itself a cadherin, a more specific fit than generic cell-cell adhesion mediator activity. Both unqualified cell-migration assertions are accepted as core, preserving the pig/valve scope of the ISS evidence. The skin-development decision concerns process scope, not rejection of the experimental fibrosis phenotype. No PAINT-node reassessment is claimed.
The cell-substrate reason now separates the original assays explicitly. In PMID:33811546, human-CDH11-bearing mouse L cells bind immobilized human CDH11-Fc; that geometry demonstrates homophilic trans adhesion. Patient fibroblast initial attachment is comparable to controls through two hours (Fig.S4C), despite altered spreading and immature paxillin-positive focal adhesions (Fig.5b–d). The earlier reason’s plate-assay explanation is superseded by the source-grounded rationale below.
Selected full Langhe et al., PMID:26952325 Results and Methods resolve the independent mechanism cited in the original Discussion. Xenopus neural-crest loss/rescue, flipping and single-cell force assays establish a structural contribution to fibronectin attachment through cadherin-11 transmembrane/cytoplasmic regions and a syndecan-4 complex. Syndecan-4 supplies the extracellular fibronectin-recognition role. Endogenous human HFF-1 focal-adhesion localization is distinct supporting evidence; Xenopus constructs transfected into human cell lines remain Xenopus protein experiments. This supports retaining the existing process as non-core with explicit species and assay limits, not assigning CDH11 an intrinsic fibronectin- or integrin-binding function.
The core description now states the already accepted adhesion-coupled movement context. The human osteoblast co-IP quote is explicitly corroborative because the source discusses cadherins 4 and 11 together. Cytoplasmic localization remains unresolved: absence from a UniProt location summary is not proof of absence. Other actions, source assertions, products, earlier evidence JSON and source caches are unchanged. A genuine normal full cache for PMID:26952325 is added; no new scientific quotations are introduced.
Round 4 of projects/INVOLVED_IN_PRECOMPOSITION.md (2026-11-07): the core-function molecular function now uses the genus GO:0045296 cadherin binding instead of the precomposed GO:0098641; the adhesion context is carried by GO:0044331 cell-cell adhesion mediated by cadherin in the core functions. The cadherin-binding IBA row is accepted as it stands rather than refined, so the earlier statements above that it is refined to GO:0098641 and that the final MF uses GO:0098641 are superseded.