FGG source evidence

The review distinguishes gamma-chain work from effects of intact fibrinogen or fibrin. The source-provenance record contains target joins; the review is the decision authority. Publication caches remain unchanged. References are cited without adding quotations.

Assembly, polymerization and platelets

• PMID:8910396, original DOI: Methods and Results on human chains in hamster BHK recipients, including alpha-gamma/beta-gamma intermediates and their subsequent use in fibrinogen assembly. Indexed original Methods lines 461–531 and Results 684–808, 943–1036 and 1112–1227 were read. Human hepatocyte-derived observations are a distinct context. These are structural contributions, not an assembly enzyme or a natural-splice comparison.
• PMID:6451630, original: human fragment-D/gamma peptide experiments, Methods lines 455–614, Results/Figures 3–4 (1511–1730), binding/Figure 6 (2149–2302), sequence Figure 8 and Discussion (3089–3188). The fragment recognizes exposed fibrin termini and inhibits polymerization. The historical gamma373–410 sequence ending AGDV is not evidence for the gamma-prime extension. Figure captions/text were read, not figure pixels.
• PMID:9207064, original: Materials and Methods and the primary-pocket Results resolve GPRP recognition by recombinant human gamma143–411 expressed in Pichia. The gamma a-pocket binds an alpha-knob mimic; it does not cleave fibrinogen. Distal C-terminal heterogeneity/unresolved residues limit tail and isoform conclusions. A separate inhibition statement in the Introduction is explicitly unpublished work, not an experiment demonstrated by this paper.
• PMID:19296670, original DOI: human plasma purification and structure; Methods and polymerization/interface Discussion, printed pages 3878–3884. Gamma D-domain interfaces support structural interpretation, while crystal packing, partial alpha-chain proteolysis and minor fibrinopeptide cleavage prevent treating every contact as an intact-native solution measurement.
• PMID:6281794, original: human plasma/platelet Methods (1101–1627), Results (1738–2251), Tables 1–2 and Figure 1 caption, Discussion (4516–5053). Gamma-chain preparations form multimers and support aggregation of ADP-stimulated human platelets; antibody inhibition supports specificity. This is not autonomous monomer agonism or a gamma-A/gamma-B comparison.

Thrombin exposes fibrin knobs by acting on other chains; factor XIII catalyzes crosslinking. Gamma supplies structural contacts. PMID:16846481 is accessible only as an abstract about alpha-R16C fibrinogen, normal plasminogen activation and delayed lysis. Its gamma-specific activation/fibrinolysis claims remain unresolved. Alpha-C cofactor assays are not reassigned to gamma.

Receptor recognition and contextual responses

• PMID:7822297 explicitly maps gamma117–133/ICAM-1 recognition in its abstract; full Methods were not recovered. PMID:10903502, publisher, independently supports the same region through actual adhesion Methods/Table 1, ERK/survival Results and Discussion. Fibrinogen substrates, gamma peptide and receptor controls support contextual endothelial adhesion and survival. Partial blockade does not establish exclusive ICAM-1 mediation.
• PMID:15739255, publisher: Materials, Results, Figures 6–8 captions and Discussion test human fibrinogen/multichain fragment D in rat vessels. PMID:19193866, original, Methods and Results test human fibrinogen on rat heart endothelial cells, with conditioned-medium ET-1/vWF, WPB staining and ERK phosphorylation. These support complex-mediated responses, not gamma as a hormone, kinase or secretory machine.
• PMID:6777381, original: Results/Figure 3 and Discussion (especially lines 1460–1617) identify fibrin-chain surface labeling and calcium-dependent retention in human platelets. Alpha-granule cargo is described, but no isolated-gamma intragranular image or transmembrane topology is claimed.
• PMID:11001902, original: clinical/minigene Results and Discussion establish aberrant RNA and a predicted severe truncation. The HeLa minigene is not a natural splice product; the predicted truncated protein was not directly demonstrated in the read result.

PMID:10788510 has explicit gamma binding in its abstract, with complete strain/construct Methods unavailable. PMID:26627825 Results/Table 1 explicitly recover FIBG_HUMAN in affinity capture; association does not demonstrate purified gamma affinity. PMID:21546586 leaves the exact gamma band unresolved. No microbial activity is assigned to the host protein.

Proteomics, products and database evidence

PMID:25037231 supplement S2 A/B row 16 contains FGG/P02679/GeneID2266; B repeats A. PMID:28675934 SI002 has 18 measured FGG rows and two separate list-only rows. Full sample labels, mixed accession groups and measurements are retained in the provenance record. Neither matrisome classification nor enrichment alone measures mechanical function. PMID:19056867 Main Table 1 identifies FGG with one unique peptide/one spectrum in urinary membrane fractions; gamma-A precursor naming is not a splice-specific peptide result. Exact target tables remain unresolved for PMID:16502470, 22516433, 23533145, 28327460 and 28344315. The MatrisomeDB finer-compartment record is also unresolved.

The official displayed product is gamma-B/gamma-prime (P02679-1, 453 residues); gamma-A (P02679-2, 437 residues) replaces precursor residues 434–453 with AGDV. All 124 raw GOA contexts use parent P02679. Deterministic projection produced 122 TSV rows and 121 review rows; all raw IDs and joins are preserved. No natural isoform assignment is inferred from fragments or clinical variants.

Four positive PAINT IBDs lie on the exact target path, with an exact 453-residue MSA match and no IRDs in the saved current family table. The older PTHR19143 API tree is explicitly distinguished from current PTHR47221 normals. Self-WITH is valid. All 45 Reactome event routes reach P02679 component records; their canonical coordinates do not resolve splice forms, and event names do not give gamma every event activity. All 14 ARBA target branches and two InterPro matches were checked. Typed GO ancestry supports the two broad hemostatic refinements; fibrin clot formation and platelet aggregation are not ancestors of one another.

Follow-up: generic claims and extended assertions

Five saved QuickGO records carry specimen-specific extensions omitted by the machine seed: rows 3/79 refer to bone marrow (UBERON:0002371), rows 5/78 to mesenchymal stem cell of the bone marrow (CL:0002540), and row 14 to colostrum (UBERON:0001914). Their exact predicates and identifiers are restored in the review with official OLS labels; raw objects and label-source URLs/hashes are retained in the provenance record. Their generic matrix-structural, matrix-location and extracellular claims are independently supported. The complete extended assertions remain UNDECIDED because their original target tables are unverified; this does not dispute the generic function or location.

Rows 97 and 102 have no raw extensions. Independent gamma polymerization experiments and the exact urinary-fraction FGG identification support their generic claims, with explicit curator deference and continued limits on the original alpha-variant/prostatic-secretion experiments. Row 111 retains its clinical/minigene and predicted-protein limitations, while the core clot-formation decision follows independently demonstrated gamma structural participation. Unknown cofactor, heterotypic-adhesion and partner-specific assertions remain unresolved.

For PMID:10788510, the accessible abstract explicitly characterizes the tested partners as microbial surface adhesion proteins and reports direct gamma/C-terminal-peptide recognition. The official GO:0050839 definition supports cell-adhesion-molecule binding as a refinement for the two existing IPI rows. This is binding to the adhesion proteins, not adhesin activity by FGG; full strain/construct and accession-specific mappings remain uninspected. The Efb affinity-capture row retains non-core generic binding under the campaign instruction. The independent ICAM-1 annotations describe signaling-receptor recognition and adhesion-molecule class; the two GO terms are ontology siblings, not redundant ancestor/descendant assertions.