DZIP1L is associated with centrioles and ciliary basal bodies and with SEPTIN2. Human evidence supports regulation of polycystin localization, while effects on cilium assembly vary by cell type and organism. The 2009 human study requires combined DZIP1/DZIP1L depletion for the reported reduction in ciliogenesis; human localization uses GFP-DZIP1L IMAGE4940443 in hTERT-RPE1. The 2017 study includes endogenous human reciprocal SEPTIN2 co-immunoprecipitation and N-terminal fragment mapping. Patient B155 fibroblasts need not lose cilia to show altered PC1/PC2 distribution. Mouse and zebrafish phenotypes are kept distinct. These results support one integrated SEPTIN2-binding and ciliary-organization core without assigning SEPTIN2 GTP hydrolysis to DZIP1L (PMID:19852954; PMID:28530676).
The 2018 mouse study separately supports Cby positioning and ciliary-bud formation. Mouse Dzip1l and Cby constructs in HEK293 cells remain mouse-protein experiments. The mutant cells retain Tctn2 and IFT recruitment despite defects in other steps. Thus neither universal cilium absence nor a generalized failure of every recruitment event is claimed (PMID:29487109).
All 28 imported assertions are preserved. The decisions are 11 ACCEPT, 10 MODIFY and seven KEEP_AS_NON_CORE; there are no NEW annotations. The project-specific binding instruction retains supported generic associations as non-core where no evidence-backed informative subclass is available. A supported pair plus independently verified partner class can justify a binding refinement without demanding a purified interface or native physiological role solely for the class term. This does not assign the partner's activity to DZIP1L. CEP76 and CBY1 are independently curated centriolar/basal-body components; AMOTL2 coordinates signaling proteins, but the tested partner fragment does not establish its full-length activities. SEPTIN2, TRIM37 and AMOTL2 support GTPase-, ubiquitin-ligase- and scaffold-binding refinements respectively. CBY2 is not substituted by CBY1, and TEPSIN's AP-4 association does not make it the AP-4 complex (PMID:25416956; PMID:27107012; PMID:28530676; PMID:29892012; PMID:31515488; PMID:32296183; PMID:33961781).
The source guide exposes exact IntAct IDs, PSI-MI methods, feature sides and independently curated class records. The portable analysis reconstructs 35 matching pair records and 70 participant ROGIDs. These authenticate database reference sequences, not whole experimental inserts. In particular, PRPF40A, LRRN4 and AMOTL2 binding-associated regions belong to the partners; the four no-effect variant features belong to DZIP1L. The BioPlex CBY1 association retains its co-IP and spoke scope.
HPA reports supported basal-body localization with antibody HPA030404; individual-location labels for microtubules, cytokinetic bridge, mitotic spindle, primary cilium, nucleoplasm and cytosol are uncertain. The overall supported score is not inherited by each location. XML assay metadata includes basal body in RPTEC/TERT1, and no image pixels were interpreted. The broader cytoplasmic IBA is refined to independently supported centriole and basal-body pools without inventing a current cytoplasm ancestry edge (HPA target page).
The three PAINT assertions join literal IBDs on the target path, with an exact 767-residue target alignment. Donor records retain their actual evidence types; propagated entries are not labelled direct experiments, and target self-donorship is expected. ARBA00027837's first clause matches IPR013087, IPR032714 and Chordata. The UniProt SL-0046 mapping is checked separately. Each methodological GO_REF has LOW biological relevance in its own reference assessment while the target evidence remains positive.
The official products are 767 and 539 residues. The shorter product retains the zinc-finger, coiled-coil and N-terminal fragment-associated region, but this does not establish equivalent whole-product function. The archived IMAGE4940443/BC033308 translation differs from the current long sequence at three positions; AK057406 similarly differs from the short sequence at three positions. The exact genuine ENA records and position comparisons are bundled in the portable analysis. Neither archived sequence is relabelled as a new natural isoform, and sequencing of the actual historical expressed insert is not established.
ROOT independently read the two reserved central originals (PMID:19852954 and PMID:28530676), supplied authenticated source advice and synthesized the final core and product descriptions. The owner independently read the mouse original and the accessible screen Methods/Results and literal target records. The actual per-reference access scope is recorded in the source guide. PMID:29892012 has a machine cache marked as full text but only partial text was available; the separately attempted PMC response is a challenge page, not a recovered original. Its positive curated pair is retained with that explicit limit. The three sequence-resource publications were read at abstract/metadata scope only. No primary paper figure pixels were interpreted in this owner lane.
The genuine Falcon attempt failed with an Edison authorization error; the genuine perplexity-lite fallback failed for insufficient quota. No provider report was fabricated. Manual research and independent primary consultation supplied the evidence above. Raw failures, intact cache bytes, the corrected acquisition-check receipt and source/runtime preimages remain in the private provenance package. No newly authored supporting-text quotations are introduced. Canonical validation, status normalization, rendering, append-only history and publication are performed separately by ROOT.
The prior decision to omit a separate transition-zone assertion is superseded by the target-resolved human microscopy in PMID:28530676(https://pmc.ncbi.nlm.nih.gov/articles/PMC5687889/), Results describing Figure 4j–k and the immunofluorescence/3D-SIM Methods. Endogenous DZIP1L colocalizes with TCTN1 in ciliated human fibroblasts; RPE-1 superresolution microscopy places the signals in close proximity. One IDA cellular-component assertion and the location in the existing core now capture this evidence. Centriole and basal-body observations are retained. These locations are not ancestor/descendant duplicates in the current typed ontology, and imaging proximity does not establish a purified gate complex or exclusive residence. The human family findings also support concise autosomal-recessive polycystic kidney disease 5 context.
The governing ClinGen curation instruction explicitly retains supported, biologically correct generic binding as KEEP_AS_NON_CORE when no evidence-backed specific replacement is established. It directs MODIFY for justified specific terms and UNDECIDED for unadjudicable evidence. The seven positive CBY2, PRPF40A, LRRN4, TEPSIN and PNMA5 rows retain their existing non-core dispositions under this task-specific instruction; expected generic-binding advisories are not evidence that those interactions are false. This is a project instruction, not a new repository-wide policy.
The six challenged CEP76, CBY1 and AMOTL2 refinements retain their existing actions. CEP76 is independently documented as a centrosomal component with CCP110/CEP97 associations; CBY1 has human basal-body recruitment evidence and a separately demonstrated mouse appendage-component role. Cytoskeletal protein binding includes such components and does not require the partner itself to form a filament. Human AMOTL2 signaling interactions support its scaffold class independently of its separately inferred MAGI1/CDH5 role. Exact pair methods, partner fragments and mutation-feature ownership remain explicit. A protein-class binding term does not transfer the partner's molecular activity to DZIP1L.
The follow-up source guide gives direct primary and official-record links and a compact partner table. It separates the incomplete PMID:29892012 representation from the positive curated pair and its independent corroboration, without rewriting the machine cache flag or treating citation correctness as synonymous with full-text access. Existing source-evidence, input archives, parser and results are retained. No new quotation is introduced; primary-reference links and source-specific paraphrases supply support. Earlier notes remain historical context, including genuine provider and access failures.
The existing transition-zone assertion now carries one short exact excerpt from PMID:28530676 at its supporting reference. It anchors the endogenous human fibroblast observation; the accompanying reason retains the separate RPE-1 3D-SIM evidence and its approximately 120-nm lateral resolution. This does not establish molecular contact, exclusive residence or membership in a defined gate complex. Earlier statements that no quotation was introduced describe the earlier versions of these notes.
Official QuickGO responses checked on 2026-10-18 establish the following literal child-to-parent edges: GO:0005814 centriole part_of GO:0005815 microtubule organizing center; GO:0005815 part_of GO:0015630 microtubule cytoskeleton; GO:0015630 is_a GO:0005856 cytoskeleton. GO:0036064 ciliary basal body is_a GO:0005815. These relations are exposed by the parent records' children in the official term query.
The corresponding official is_a/part_of ancestor query does not place either replacement or cytoskeleton beneath GO:0005737 cytoplasm. Thus the suggested final cytoskeleton-to-cytoplasm edge is not present in this graph. The existing cytoplasm-review disclaimer is retained: target-specific microscopy supports a biological refinement to measured centriole and basal-body compartments, without asserting a formal descendant relationship or an exhaustive inventory of DZIP1L molecules. The uncertain HPA cytosol observation remains uncertain.