Three coupled mechanisms distinguish the evidence: structural cytolinking at cellular attachment sites, microtubule binding/regulation, and neuronal association with dynactin-dependent retrograde transport. Human epithelial BP230 provides intracellular contacts with integrin beta4 and collagen XVII. In keratinocytes its depletion disrupts actin organization, polarity and directional persistence, without a significant change in average migration speed. These observations do not establish intrinsic exchange-factor activity. Muscle Z-disc evidence is positive but its complete original remains unavailable. PMID:10637308(https://pubmed.ncbi.nlm.nih.gov/10637308/), PMID:11375975(https://pubmed.ncbi.nlm.nih.gov/11375975/), PMID:19403692(https://pubmed.ncbi.nlm.nih.gov/19403692/), PMID:19932097(https://pubmed.ncbi.nlm.nih.gov/19932097/).
The BPAG1n4 tail experiments demonstrate calcium-sensitive EB1/plus-end versus lattice behavior. They do not establish identical regulation of every natural product, a direct calcium-affinity value, or a statistically significant calcium-dependent increase in microtubule cosedimentation. The neuronal transport experiments distinguish human-derived cDNA/library constructs from native mouse localization and transport assays. DST supplies a linker/adaptor role; dynein supplies motor force. PMID:22995871(https://pubmed.ncbi.nlm.nih.gov/22995871/), PMID:14581450(https://pubmed.ncbi.nlm.nih.gov/14581450/).
The nine official records are retained, including short products; the longest protein is not a substitute for the others. Names below are the official product numbers, with their listed synonyms. Historical long neural constructs are not identified solely from a shared short-product alias.
| UniProt product | Official name | Residues | Synonyms |
|---|---|---|---|
| Q03001-7 | 1 | 7570 | BPAG1-b |
| Q03001-8 | 2 | 5171 | BPAG1eA, Dystonin-1, eA |
| Q03001-3 | 3 | 2649 | BPAG1e, eBPAG1 |
| Q03001-9 | 4 | 3060 | BPAG1eA, eB |
| Q03001-10 | 5 | 304 | |
| Q03001-11 | 6 | 224 | BPAG1n1, BPAG1n2, Dystonin-2 |
| Q03001-12 | 7 | 81 | BPAG1n3 |
| Q03001-13 | 8 | 905 | |
| Q03001-14 | 9 | 5375 | BPAG1-a, BAPG1n4 |
The epithelial BP230 architecture is consistent with Q03001-3, while historical long n4 naming is consistent with Q03001-14; neither correspondence authenticates every residue of the old full-length clone. The deposited 3GJO 30-residue peptide exactly matches the tails of Q03001-7, -8 and -14 despite its historical accession annotation. These are segment matches, not evidence that all three complete natural products were assayed. The reproducible analysis retains both the historical label and actual sequence.
The sperm-nuclear source is positive contextual detection: the author-thesis table lists Q03001 with 1.4% coverage and three peptides after human sperm-nucleus preparation. It does not resolve an isoform or a nuclear mechanism. Mouse neuronal ER/nuclear-envelope and N-terminal stress-fiber/nuclear evidence are retained as distinct non-core pools. The H-zone assignment remains unresolved because the accessible source does not settle that narrower compartment; positive Z-disc evidence does not exclude it. PMID:21630459(https://pubmed.ncbi.nlm.nih.gov/21630459/), PMID:18638474(https://pubmed.ncbi.nlm.nih.gov/18638474/), PMID:14576348(https://pubmed.ncbi.nlm.nih.gov/14576348/).
Supported partner-class refinements identify EB1/plectin as cytoskeletal partners, collagen XVII as a collagen partner through its cytoplasmic domain, and dengue NS5 as an RNA-polymerase partner in a fragment screen. None assigns the partner’s downstream function to DST. ERBIN, DISC1 and APPL1 associations remain positive non-core where no warranted narrower binding class was established. Co-complex recovery is distinguished from a purified binary interface. The explicit CLINGEN project policy retains such supported generic binding, rather than removing it solely for low information content.
All 69 raw GOA contexts are preserved; deterministic normalization gives 68 review rows. The sole collapsed EB1/PMID:19632184 pair differs in identifier, assignment and date fields, not only date. All nine source IBDs and the full DST branch are authenticated; target self-grounding is legitimate. Actual rule records are retained, with unresolved FunFam-condition membership left unresolved. Reactome explicitly models a cytosolic BPAG1e:Plectin input, but its stale 7570-residue Q03001-3 coordinates are not used as product evidence.
Normal source files and cached full-text flags remain unchanged. Separately recovered originals are distinguished from abstract-only caches in the source guide. The genuine Falcon research attempt failed authentication; the configured fallback returned HTTP 401 insufficient_quota. No manual text was labeled as a provider report. Three concise source excerpts appear once each in the core; these notes and the guide paraphrase their findings.
The neuronal mechanism remains cytoskeletal adaptor activity (GO:0008093), the current official preferred label. Dynactin binding (GO:0034452) describes a complementary contact rather than replacing that transport-linking role. The new binding assertion uses human brain-derived interaction constructs, with native mouse p150/p50/dynein-complex association as corroboration. The human p150 partner is DCTN1/Q14203. The original Results and Figures 2–5 distinguish interaction mapping, native complexes and transport; the cloning Methods identify human cDNA and library provenance. Neither lysate overlay nor co-immunoprecipitation is represented as purified-complex reconstitution, and no complete current natural DST product is assigned. Existing broad cytoskeletal-partner refinements concern EB1/plectin; the added dynactin claim records a different partner and source context. PMID:14581450(https://doi.org/10.1083/jcb.200306075), GO:0008093, GO:0034452.
The dengue NS5 refinement is retained on the identified partner class and controlled fragment interaction evidence. Four curated records include sufficient-binding-region features extending through NS5 residue 835; they are not all short N-terminal peptides. The original viral-library construction, interaction-screening and pairwise-retest Methods describe sequence checking, independent reporter selection and exclusion of promiscuous activators. These controls do not establish a DST-specific infection mechanism or an orthogonal cellular validation. The normal abstract-only cache is unchanged; the separately recovered original and the literal interaction records provide the additional scope. Reference-sequence checksums and sufficient-binding-region features remain distinct from complete clone sequence proof. PMID:21911577(https://pmc.ncbi.nlm.nih.gov/articles/PMC3237087/).
The epithelial migration term is refined to keratinocyte migration (GO:0051546), both in the existing IMP review and the integrated structural core. The human keratinocyte Methods and Figure 6 Results report impaired persistence/turning after BPAG1e depletion, unchanged average speed and Rac1/cofilin bypass. This specifies the tested cell type without turning a migration phenotype into intrinsic motor or exchange-factor activity. Polarity remains separately represented, and the integrin/collagen contacts remain part of the same structural mechanism. PMID:19403692(https://doi.org/10.1091/mbc.e09-01-0051), GO:0051546.
The 68 imported source dictionaries, nine official product records, historical notes and three existing core quotations are preserved. The sole appended annotation is the dynactin-binding MF; no new process assertion is added. The public parser has been reformatted for readability without changing its syntax tree, inputs, interface or reproduced results.