Genuine repository fetching seeded 89 source contexts and three natural products. The review retains every source field and row order. Each context was assessed against primary-source consultations, official interaction and localization records, mouse donor evidence, and the actual PANTHER target path. The source evidence guide records access scope, public retrieval provenance, content hashes and the unchanged normal sources. Provider authentication/quota failures were already established in this campaign, so no repeated provider attempt or provider-branded manual report was created.
The synthesis is one plasma-membrane VEGF-C/D receptor function, combining ligand binding with receptor association and tyrosine kinase activation. Direct human receptor experiments distinguish ligand capture and structural extracellular-domain assays from cellular signaling. Endothelial growth, migration and ERK responses are supported central outputs; carcinoma autocrine responses and early embryonic vascular remodeling remain contextual. Downstream phosphorylation does not establish that every measured protein is a direct FLT4 substrate (PMID:23878260(https://pubmed.ncbi.nlm.nih.gov/23878260/), PMID:9435229(https://pubmed.ncbi.nlm.nih.gov/9435229/), PMID:11532940(https://pubmed.ncbi.nlm.nih.gov/11532940/), PMID:19779139(https://pubmed.ncbi.nlm.nih.gov/19779139/)).
The decisions are 42 ACCEPT, 25 KEEP_AS_NON_CORE, 17 TOO_GENERAL, three MODIFY and two UNDECIDED. Three growth-factor-binding contexts are refined to vascular endothelial growth factor binding using actual ligand evidence and official ontology ancestry. Broad kinase, pathway and selected regulatory terms are TOO_GENERAL where a supported narrower term entails them. Broad cytoplasm is treated consistently with the retained HPA cytosol context; internalized membrane receptor is not called a free cytosolic protein.
Nine supported generic protein-binding contexts remain KEEP_AS_NON_CORE under the explicit campaign instruction. This instruction takes precedence over local requests to remove generic binding or add publication quotations. Associations are scoped to their actual assays: receptor complexes, ligand-domain binding, extracellular screens or chaperone-client co-complexes. They do not create a new adapter, cochaperone or substrate activity. References provide support without new publication quotations. Default validators are unchanged, and their resulting advisories are reported rather than suppressed.
Official name 1/Long is P35916-2, the displayed 1363-aa sequence; name 2/Short is P35916-1, 1298 aa. The accession suffix must not be inferred from the product name. The 830-aa P35916-3 sequence contains altered canonical residues 724–765, lacks 766–1298 and retains the final canonical 65-aa tail. Its mapped membrane-anchor and kinase regions are absent. All three official FASTAs were reconstructed from the UniProt sequence events.
The secreted VEGFR3-765 clone characterized in PMID:18593464(https://pubmed.ncbi.nlm.nih.gov/18593464/) is 765 aa and is not established as identical to current P35916-3. Its ligand binding and secretion do not establish all activities of the 830-aa product. CSF1R/FLT4 and EGFR/FLT4 chimeras, domain deletions, tagged fragments and dominant-negative constructs are experimental tools, not additional natural products. The core therefore describes the characterized membrane receptor and does not transfer kinase signaling to the soluble splice product.
Primary human endothelial ligand-selective experiments are distinguished from Ba/F3 human-receptor-ECD/mouse-EpoR chimeras (PMID:11532940(https://pubmed.ncbi.nlm.nih.gov/11532940/)). EGFR-ECD/FLT4-tail chimeras stimulated with EGF support a tail-dependent JNK branch, with that engineered context explicit (PMID:16076871(https://pubmed.ncbi.nlm.nih.gov/16076871/)). The tumor study separates human SAS-cell signaling and VEGF production from mouse xenograft vascular endpoints (PMID:19779139(https://pubmed.ncbi.nlm.nih.gov/19779139/)).
Independent peer review clarified the growth-assay scope in PMID:11532940: VEGF or VEGF-C was used for the proliferation/PCNA observations. VEGFR3-selective VEGF-C156S was used in separate signaling, survival and migration experiments. The two endothelial-proliferation decisions retain ACCEPT without describing a C156S proliferation assay or claiming that the PCNA experiment isolates FLT4 from VEGFR2.
The two nuclear-location assertions remain UNDECIDED. The authentic PMID:16452200(https://pubmed.ncbi.nlm.nih.gov/16452200/) abstract does not resolve the localization experiments, and bounded full-text retrieval failed. Membrane-receptor architecture is not a reason to declare the nuclear annotation false. Human surface/internalized localization in PMID:20445537(https://pubmed.ncbi.nlm.nih.gov/20445537/) was inspected in selected indexed original Methods/Results/captions; separate mouse experiments, unseen figures and supplements are not silently counted as human evidence.
The receptor-complex annotation citing PMID:23382219(https://pubmed.ncbi.nlm.nih.gov/23382219/) is retained from independently demonstrated human ligand-induced receptor assembly and the on-target IBD. The cited paper's receptor-tail peptide arrays do not demonstrate a native FLT4–SNX trafficking complex, and its detailed cellular validations concern other receptors. The source-specific limitation remains visible in the annotation rationale.
The mixed-species Reactome R-NUL-9604519 event has a human FLT4 promoter and mouse NICD4 input. Its nucleoplasmic entity is DNA; its protein output is membrane FLT4. It supports the recorded protein location, not nuclear receptor localization or a transcriptional role for the receptor.
The actual 2746-node family tree and 2746-sequence alignment reproduce the exact displayed 1363-aa target. All ten incoming IBA origins join positive original IBD records on that target path. Self-donation is legitimate descendant evidence, and donor count is not used as a quality proxy. Node taxonomic ranges are reported as such when an explicit species field is absent.
All 159 existing PTHR24416 PAINT rows match the original IBD GAF. The current API projection contains 135 of them; the remaining 24 valid original rows are preserved unchanged. The API projection is not a replacement cache. No family-node metadata or off-path function produces a new annotation. Nine normal sources were reused byte-for-byte from the approved main commit before genuine fetching; the other 23 sources were absent and unowned at the complete stable PR-ownership check.
The curation history, source-context conservation, normal-source conservation, status calculation, schema/reference/GO checks and private rendering are checked in the accompanying candidate packet. The renderer is a review aid; it does not add a public generated page to this payload. The final status is determined by the ordinary checker, with the campaign binding advisories and unresolved nuclear evidence disclosed.
Priority follow-up questions are endogenous nuclear localization, natural-product-specific function of the 830-aa splice form, and long-versus-short C-terminal signaling differences. None is filled by inventing a new GO assertion or transferring engineered-construct behavior to every natural product.
The ordinary checker computes DRAFT. The full validation reports 13 disclosed advisories: nine generic-binding policy advisories and four source-context action differences. Lymphangiogenesis is central in endothelial/donor evidence but contextual in the tumor-cell paracrine experiment; general proliferation, cellular VEGF response and MAPK regulation retain carcinoma-specific results alongside narrower endothelial/pathway annotations. These distinctions follow actual experimental contexts, not preference for an evidence code. The growth-factor-binding IBA refinement includes structured metadata identifying a sound ancestral transfer and a target-specific term-granularity improvement. The default validator is unchanged.
The initial counts and validation disposition above record the first published candidate. The current decisions are 42 ACCEPT, 23 KEEP_AS_NON_CORE, 17 TOO_GENERAL, five MODIFY and two UNDECIDED. Two generic binding rows with VEGFC as the partner now propose GO:0038085 vascular endothelial growth factor binding, alongside the three existing growth-factor-binding refinements. All 89 original machine contexts and their order, the other 87 reviews, one receptor core and three natural products are unchanged.
For PMID:21130043(https://pubmed.ncbi.nlm.nih.gov/21130043/), two exact positive IntAct records establish solid-phase binding between human receptor extracellular constructs and processed VEGFC. The full primary Methods were not independently recovered, and the curated assay/construct evidence remains identified as such. For PMID:23878260(https://pubmed.ncbi.nlm.nih.gov/23878260/), actual structural and thermodynamic experiments establish receptor-domain binding to engineered mature VEGFC. Neither refinement implies whole-cell affinity, behavior of unprocessed ligand, an exact natural splice-product assignment or a new process annotation.
The campaign permits retaining supported generic binding only where an evidence-backed specific replacement has not been established. These two source contexts meet that specificity condition; the other seven generic-binding decisions remain non-core. The expected default-validator disposition is seven binding advisories plus the same four source-context advisories, with computed status determined by the ordinary checker. The source guide pins the authoritative v4 matrix and the independent specificity consultation. All 32 normal sources and both earlier history records remain unchanged; this follow-up adds no publication quotation or NEW annotation.
The initial review identified additional binding specificity and biological-coverage issues. Seven original generic contexts now have supported replacements: two VEGFC interactions to vascular endothelial growth factor binding, three KDR interactions to signaling receptor binding, and two HSP90AB1 interactions to Hsp90 protein binding. The latter refinements describe measured association, including co-complex assays, without assigning FLT4 agonist/cochaperone activity or purified binary affinity. CLEC14A and CEACAM4 remain the two supported generic NC contexts under the explicit user campaign instruction; neither a narrower activity nor its absence is inferred from a partner name. This instruction also requires reference-only support without new quotations and takes precedence over local skill defaults. Validators are unchanged.
Same-term decisions were reassessed using biological coverage. Lymphangiogenesis is a central function across its retained contexts; the PMID:19779139 tumor-cell perturbation remains distinct from direct host-endothelial perturbation. Broader proliferation and MAPK regulation remain NC because SAS/fibroblast growth and p38 responses are not exhausted by the endothelial-proliferation and ERK terms. Both VEGF-response rows are TOO_GENERAL because their observed receptor-mediated responses are already represented by retained signaling annotations. Three cytoplasm rows become NC: the human internalized receptor pool in PMID:20445537 is not established as identical to the separately observed HPA cytosol pool. No endosomal subtype or nuclear localization is invented; the two nuclear decisions remain UNDECIDED.
The single receptor core now includes five nonredundant process terms: VEGF receptor signaling, lymph vessel development, sprouting angiogenesis, endothelial proliferation and endothelial migration. Official QuickGO is_a/part_of records show no mutual ancestry among these five; lymphangiogenesis is represented in prose rather than redundantly alongside its developmental parent. Human endothelial assays and independently read mouse vascular experiments support receptor participation, with VEGFC156S migration/signaling distinguished from VEGF/VEGFC growth-PCNA assays. The tumor study is not the sole developmental basis, and kinase-dependent lymphatic signaling is not universalized to every blood-vascular response. The current soluble natural product is not assigned membrane-receptor signaling.
All 27 publication entries now have structured reference assessments: 14 scoped VERIFIED and 13 UNVERIFIED for unresolved original-assay details. Their review notes distinguish checked identifiers and actual primary sections from abstract-only, curated-record or indexed-partial access; they do not claim a complete paper, figure-pixel or supplement audit. Public URLs and retrieval receipts remain in the source-evidence guide, whose role is source/construct provenance. No publication quotations were added. New concise summaries cover the changed annotations without mechanically rewriting the remaining reviews.
Current decisions are 43 ACCEPT, 21 KEEP_AS_NON_CORE, 13 TOO_GENERAL, 10 MODIFY and two UNDECIDED. All 89 original machine contexts/order, three products, 32 normal sources and prior histories are conserved. Fifteen reviews differ from the initially published candidate; the other 74 remain unchanged. The remaining lymphangiogenesis rationale was narrowed to human endothelial growth and selective-receptor migration, avoiding an implied selective-C156S proliferation assay. The two remaining generic-binding advisories are expected; final validation and computed status are recorded in the new history and candidate receipt. Earlier journal entries describe their corresponding earlier candidates and remain unchanged.