BLNK is a noncatalytic B-cell signaling adaptor. Its tyrosine-phosphorylated docking sites recruit effector SH2 domains, while proline-rich motifs recruit SH3-domain proteins such as CIN85 and CD2AP. These interactions assemble rather than catalyze the receptor-proximal reactions. SYK phosphorylates BLNK, BTK can phosphorylate PLCgamma2, and PLCgamma hydrolyzes membrane phospholipid. BLNK itself performs the structural coordination that makes it a participant in BCR signal transmission, rather than merely a substrate consumed by another protein's chemistry (PMID:9697839; PMID:11606584; PMID:21822214).
The two proposed core functions describe signaling adaptor activity and molecular condensate scaffold activity. They are distinct, already represented molecular functions, not a new disease-derived process collection. Both contribute to the existing BCR pathway. Lipid binding and multivalent CIN85 binding support spatial assembly of signaling granules; human recombinant SLP65 fragments, a defined CIN85 deletion construct and synthetic vesicles were used in reconstitution. Human SLP65 complementation in chicken DT40 cells links the relevant interaction motifs to granules and calcium responses. Endogenous concentration measurements in human B cells are a separate observation. The source expressly identifies in-situ vesicle organization within cellular granules as a remaining question (PMID:32051419; PMID:25140054). No NEW annotation is proposed, and no gene-specific GO-CAM index entry was found by the exact BLNK/Q8WV28 search.
All 45 source objects, their qualifiers, partner identifiers, evidence codes and original references are preserved. All three UniProt alternative products are preserved, and the absent products slot remains absent. Historical long/short SLP-65 forms (PMID:16912232) and pp70/pp68 species (PMID:9341187) have not been automatically matched to current product accessions. The proposal has 26 ACCEPT, 10 MODIFY, 4 KEEP_AS_NON_CORE and 5 UNDECIDED decisions. Generic protein binding is refined when an experimentally identified interface supports a better term. The SCIMP association remains non-core because the selected experiments demonstrate a cooperative complex rather than an independently isolated binary interaction. The unavailable BTK source and unadjudicated AR/KIT supplemental measurements remain unresolved. Term breadth by itself is not treated as proof that an interaction is false.
The transmembrane-receptor-tyrosine-kinase adaptor term is generalized to signaling adaptor activity for the original BCR source. GO:0005068 refers to bringing together a receptor that is itself a tyrosine kinase with other molecules; BCR-associated cytosolic SYK does not meet that receptor identity. The RTK-specific IBA process row is handled separately. The actual local PTHR14098 PAINT export places GO:0007169 at PTN001019935, with experimental mouse Clnk (MGI:1351468) as a descendant. The MGI record establishes that identity. A single experimental descendant can support an ancestral assertion, and a target among its own descendants is normal grounding. No target-specific loss or erroneous node placement is asserted here. The ancestral experimental chain was not fully resolved against the receptor-class restriction, so this IBA remains UNDECIDED rather than being called over-propagated. The broad intracellular-signaling, cytoplasmic, adaptor and BCR IBAs are supported by their actual PAINT rows and the direct BLNK evidence.
The source partner Q5ZMQ7 is chicken SH3KBP1/CIN85, independently mapped in NCBI Gene 418608. F1N9Y5 is chicken SYK, supported by the Reactome entity and KEGG gene. These species-specific partner identities remain unchanged. Direct UniProt endpoint attempts were inaccessible or returned a JavaScript fallback; the identity statements rely on the successful named primary database pages, not those failed attempts. Other partner classes used here are identified in the original GOA/UniProt record and read source descriptions. The author-assigned replacement definitions were checked in AmiGO: SH3 domain binding, protein tyrosine kinase binding, phospholipase binding, and the existing adaptor/condensate terms. The retrieved ontology pages report their snapshot dates; no exhaustive ontology audit is claimed.
The four Reactome summaries were read completely. Their reaction titles do not assign the chemistry of SYK, PLCgamma or BTK to every component of a BLNK complex. The BTK-inhibitor summary does not itself mention BLNK; the location assertion is retained with independent signalosome-localization corroboration, not as a BLNK drug-target claim. Broad humoral immunity, gene-expression regulation and B-cell differentiation remain useful non-core biological context. The original inflammatory-response TAS claim remains unresolved from its available abstract, without asserting misattribution.
All eleven PMID entries were inspected. Ten contain complete abstract narratives; PMID:16969585 is citation-only. The normal cache has full extracted text for PMID:21822214, PMID:24728074 and PMID:32051419. Selected load-bearing Results, Methods, Discussion and figure-legend text from these were read, with exact paragraph/line ranges recorded in each reference_review and the owner reading receipt. The individual paragraphs were read without clipping after initial broad displays of duplicate aggregate XML content clipped. No whole-paper, figure-image, supplement or raw-data audit is claimed.
PMID:21822214 reading includes lines 92,94,100,102,108,114,116,122,128,134,154,160: interaction kinetics, motif mutations, direct binding, localization, complementation, and construct/cell-system Methods. PMID:32051419 reading includes lines 101,105,107,113,117,119,125,135,152,156,158,166,170,184: phase boundaries, small-vesicle preference, mutant rescue, cryo-ET text, multivalency, stated limitations and recombinant/cellular constructs. PMID:24728074 reading includes lines 108,116,120,126,190,192,198,200,266,270,272,274,276,278,280: assay design, receptor-specific results, and interpretation. Its cached body contains no BLNK-name match; the exact BLNK pairs in supplementary data remain unreviewed. A direct PMC attempt encountered a browser check. The PMID:16969585 publisher attempt was inaccessible. Neither limitation is evidence of an erroneous experimental annotation.
The four existing different PMID caches were preserved. In particular, PMID:21930792 remains canonically abstract-only. Additional full text obtained by the same normal Seed52 fetch was read in the staged variant, characters 32500–40500. The relevant primary article describes pervanadate-treated human Ramos SCIMP–SLP65–GRB2 association, and separate endogenous MHC-II-stimulated murine K46 co-immunoprecipitation. It does not establish isolated binary binding or human K46 experiments. No supporting_text quotes are taken from that staged-only full text. The canonical abstract remains the installed source for automated quotation validation.
Seven previously absent normal publication files were imported by the authorized copier and verified in tmp/seed52-auxiliary-import-receipt.json (actual835154) and postverification (actual3b1eee), after ROOT independent peer90ee5c. Four differing PMID caches, four equal Reactome caches, and both existing family files were preserved. All final supporting_text snippets in this proposal are exact substrings after whitespace normalization of canonical caches; aggregate quotes are at most 25 words per source, counted across rows and cores. This is the assistant quotation boundary, not a repository validation rule.
ROOT made one normal just deep-research-falcon human BLNK --fallback perplexity-lite invocation. The primary and fallback both failed before research because the offline deep-research-client[cyberian]==0.2.7rc1 dependency was unavailable (actualca9689, exit1). No provider-named research files were produced. No provider retry or new source-fetch command was made by this proposal owner. This file records manual literature assessment and is not presented as a successful provider report.
The proposal is temporary and awaits independent ROOT biological review. Canonical review, notes, HTML and history were not changed by this preparation. Reference identifiers/titles, raw GOA, UniProt and cached sources are protected. Application, normal gene-path validation, rendering and append-only history belong to the subsequent authorized canonical step.
Focused temporary validation completed with schema, reference/title/supporting-text and ontology checks (startb2888f, terminal830522, exit0). The single advisory is the deliberately retained generic SCIMP interaction; its treatment follows the supplied ActionEnum and the explicit cooperative-complex evidence. A first pass had also requested structured propagation metadata; the final proposal records the inspected node and Clnk identity as UNRESOLVED without classifying a failure that was not established. That small addition was revalidated. The earlier proposal, helper and validation output remain under before-propagation-detail. Because the review is outside the normal gene directory, validation used --no-goa; a separate exact parser check (actual60be7e) matched all 45 raw GOA tuples, all 45 complete source objects and all three extracted UniProt products. The absent products slot and every protected cache remained unchanged.
ROOT independently reviewed all 45 annotation decisions and both core functions against the primary-source assessment, selected Results and Methods, and official GO definitions (6205a3). The exact proposal was applied (5aff4b), then passed normal gene-path validation (d9b925), HTML rendering (b9744f) and history validation (8d55c8). The one advisory concerns the explicitly justified non-core SCIMP complex association; the supplied ActionEnum supports retaining a valid contextual association. The rendered page embeds the reviewed YAML. All 45 source objects and three alternative products are preserved, with no NEW annotations. Five unresolved assertions remain explicit. Source caches are unchanged. No whole-paper, figure-image, supplement or repository-wide audit is claimed, and no successful provider report is claimed.
This entry supersedes the earlier reason for leaving the SCIMP interaction generic. BLNK recognizes a phosphorylated residue on SCIMP through its own SH2 domain; cooperative GRB2 recruitment does not negate that mapped interface. The existing IPI annotation is refined to phosphotyrosine residue binding, GO:0001784. The reciprocal SH2-domain-binding term describes a different binding direction.
Evidence: the unchanged UniProt record specifies the BLNK SH2–SCIMP Tyr131 interaction, citing PMID:21930792(https://pubmed.ncbi.nlm.nih.gov/21930792/). The complete cached abstract and selected primary Results on human Ramos-cell mutagenesis were read again. Endogenous MHC-II confirmation remains a murine K46 experiment. Neither purified-pair measurements nor a complete figure/supplement audit is claimed. The exact short abstract quotation is recorded once in the YAML.
The corrected totals are 26 ACCEPT, 11 MODIFY, 3 KEEP_AS_NON_CORE and 5 UNDECIDED. Both core functions and all 45 original source objects remain unchanged. No NEW annotation or product is added. The unresolved inflammatory, BTK, AR/KIT and RTK assertions retain their evidence limits. Lipid binding is already retained experimentally and described in condensate assembly; this focused change adds no new core activity or redundant process annotation. Existing journal entries are preserved as dated history.
Independent science review approved the SCIMP phosphotyrosine-binding refinement. The exact proposal was applied; focused canonical validation passed without advisories, rendering passed and the new history record validated. All raw source caches and previous history remain unchanged.
All 45 baseline annotations were assessed, including functions outside BCR signaling. Original terms, evidence, qualifiers, source partners and flags are unchanged. The panel acquisition performed real Falcon/fallback attempts and publication caching; provider attempts failed and no provider prose was fabricated. Prior substantive human BLNK source work was rechecked and retained where sound.
The adaptor, lipid-vesicle and condensate mechanisms remain supported by the original BLNK identification, 11606584 purified assembly, 21822214 partner-domain work, 25140054 vesicle binding and full 32051419 phase separation. Human constructs in chicken DT40 are still human proteins. The recovered full mouse donor source 22949502 supports contextual positive gene-expression regulation through inducible BASH–ER signaling and LAPTM5 induction. The kinase-independent organizational role is distinct from phosphorylation or phospholipase catalysis. Original source-specific uncertainties (16969585 and AR/KIT pair-level 24728074 evidence) remain explicit.
The RTK-pathway IBA at PTN001019935 remains unresolved: BCR/SYK signaling is not receptor-tyrosine-kinase signaling, but that distinction does not itself disprove a broader ancestral adaptor function. The single Clnk evidence source is not weak by donor count. No new annotations were proposed.
Normal recovery obtained full PMID:18369315 and the author manuscript for PMID:11606584. Earlier abstract-only access descriptions for those sources are superseded. The recovered texts support Syk–SLP65 feedback and purified BLNK/PLCgamma2 assembly; mixed-species construct boundaries remain in force.