The accepted intake preserves genuine source acquisitions and the unsuccessful provider attempt. The following manual-source notes record the biological assessment; no provider-authored report is fabricated.
2026-10-25: Reviewed the 33 assigned source contexts, leaving all 30 direct-original contexts PENDING in this private candidate. The manual result is 24 ACCEPT, four MODIFY and five KEEP_AS_NON_CORE. The four refinements specify native external membrane association, performed zymogen activation, intrinsic coagulation and serine endopeptidase activity. The direct owner confirmed that its imported catalytic rows already use GO:0004252, its intrinsic-pathway row uses GO:0007597, and its protein-processing refinement uses GO:0031638; no unexplained shared-term conflict remains.
The five non-core contexts retain positive or inferred evidence: calcium-binding EGF structure, inherited mouse-Hgfac rough-ER biosynthesis, and three disease-F12 processing/localization assertions. They do not reject extracellular location or treat a mutant partner as a mutant F12 molecule.
The actual PTHR24264 v19 target sequence/path and all six path IBD rows were verified. This differs from the earlier F11 family-access problem and must not inherit that limitation. The Reactome graph is complete at the stated participant scope (173 fetched objects plus six inline terminal chemical leaves); failed rate-limited captures remain preserved. Two actual residue objects verify T328R/K without implying that residue 328 lies inside every processed C-terminal fragment.
Selected human PMID:18725990 and PMID:34967109 passages were independently read, with enzyme-versus-downstream-reporter and processing-form distinctions. The two mouse donor originals PMID:15351846 and PMID:12923239 were recovered and read at the documented operative text/caption scope. No new pixels, publication quotations or provider-branded research were authored. The source guide supplies public section/source locators; the private read ledger retains exact source bytes, line selections and hashes.
The gene source objects, reference identities/findings, displayed-product structure, status and description remain unchanged in the lane candidate. Reference assessments and three genuine added references are delivered as explicit proposals. Original 63 normal pairs and 232 runtime/source preimages remain protected; the normal union includes two database additions and the byte-exact direct-lane addition, without overwrite or restoration. No canonical, Git, history or public action was taken.
The key chemical distinction is between F12 being activated and activated FXII performing a cleavage. Normal human plasma studies and explicit human FXIIa reactions support prekallikrein and FXI activation. The broad protein-processing context can therefore be refined to zymogen activation. The independent FXII-only silica/polyphosphate assay supports surface-dependent functional autoactivation without added prekallikrein; its S-2302 readout is calibrated to FXIIa. This does not replace the separate limitation that the misfolded-aggregate experiment chiefly measures downstream kallikrein. See PMID:18725990(https://www.jci.org/articles/view/35424), PMID:6793628(https://www.jci.org/articles/view/110325) and PMID:34967109(https://doi.org/10.1111/jth.15631).
The aggregate-response study separates nonfibrillar activation of the kallikrein system from surface-dependent coagulation. Recognition of misfolded proteins remains an inference from that functional contrast; aggregate clearance is a proposed model. Fibrinolytic effects in the older reconstructed cascade are mediated chiefly through downstream proteases, not strong direct plasminogen activation by the Hageman factor fragments. The source guide records the selected-page access boundary for PMID:89876(https://doi.org/10.1182/blood.v54.4.850.850).
Inhibitor binding is not inhibitor activity of F12. Hamadarin, triafestins and HRG have distinct original assay scopes, and HRG preferentially binds the tested α-FXIIa rather than the zymogen or β form. APP/LAMP2 two-hybrid pairs are positive source-specific associations without established physiological processing or new partner-derived functions. The ayaconin source remains unresolved because its operative controls and target provenance were not recovered. Matrix- and urine-vesicle-enriched fraction detections are retained as contextual locations, with carryover and protein-group limits intact.
The 615-residue displayed precursor, activated chains and historical experimental fragments remain distinct. No splice-product array, new publication quotation or new annotation is introduced. The 33 database contexts are reserved for their separate review. One genuinely fetched normal reference, PMID:34967109, is added; its abstract-only cache is preserved while the accessible external original supplies the actual Methods and Results. All existing normal publication and machine-source bytes remain unchanged. The complete 13-source direct review and its independent supporting reference are documented in the source guide.
The two accepted lanes cover all 63 imported contexts: 37 ACCEPT, five MODIFY, 19 KEEP_AS_NON_CORE and two UNDECIDED. All source objects and all 35 original reference identities/findings remain exact. Three genuinely acquired references are appended, giving 38 references. Shared PMID:18725990 and PMID:34967109 retain both labeled reading scopes and consistent original-availability assessments. The absent alternative-products field, seed description and INITIALIZED status remain unchanged pending core synthesis. No NEW annotation or publication quotation was added. Native F12 executes FXI/prekallikrein activation; processing of the F12 substrate by another enzyme is a distinct role. Disease-F12 location contexts remain explicitly non-core. Rough-ER and coagulation IBD assertions have mouse-only donors; general self-WITH guidance in the lane prose does not assert that human F12 is a donor for those two rows.
One catalytic unit represents the human FXIIa serine protease acting on FXI and prekallikrein, with intrinsic coagulation, zymogen activation and protein autoprocessing as distinct process facets. Current official typed closures establish that these three process terms do not stand in an ancestor/descendant relationship. No separate binding or calcium core is introduced. Rough-ER biosynthesis, inhibitor associations, pathological variant processing and detection in extracellular fractions retain their accepted contextual reviews.
The synthesis independently read the saved complete-original text of PMID:34967109 at the human-reagent, isolated autoactivation, FXI/PK activation and operative Results sections. CTI is added before downstream FXIa/kallikrein readouts; the no-PK autoactivation assay is a separate experiment. The activity endpoint does not determine the initial single-chain versus preprocessed species or newly map a cleavage site. The current GO:0016540 definition has no explicit intramolecular restriction; its use here represents the accepted FXII-on-FXII processing/amplification role, not a claim that a molecule cuts its own chain in cis. PMID:18725990 Methods and operative Results distinguish nonfibrillar aggregate-driven kallikrein generation from the surface conditions that also support FXI activation; its protective-clearance proposal remains a hypothesis. No new pixels or supplements were read.
The extracellular/contact-surface setting is supported by the accepted human source-model routes as well as the enzyme assays. Reactome R-HSA-9909048 explicitly assigns an F12 active unit in a receptor-associated complex and discusses catalytically competent single-chain FXII; this model is not erased by focusing the core on the demonstrated FXIIa reactions. R-HSA-9936430 instead uses extracellular alpha-FXIIa against a membrane-associated FXI substrate complex. These source-specific roles do not imply that every bound FXII molecule has the same activation state or that all endothelial contact activation requires F12.
The 615-residue displayed precursor and alpha/beta processing chains remain distinct from external transcript models; the alternative-products field remains absent. No activity, localization or processing equivalence is assigned to shorter models. The two paired experiments address pathway selection and the initiating species in FXII-only surface activation, using chain-resolved measurements alongside protease-specific endpoints.
FXIIa performs the conversion of prekallikrein to kallikrein in the plasma kallikrein–kinin cascade; kallikrein then cleaves kininogen. This existing accepted role is now explicit in the same catalytic core as FXI activation and surface-dependent autoactivation. GO:0002353 is not an ancestor or descendant of the three other core process terms under current is_a/part_of closure. Human assays distinguish FXII-only activation from downstream kallikrein reporters (PMID:34967109(https://doi.org/10.1111/jth.15631), Methods 2.1–2.7; PMID:18725990(https://www.jci.org/articles/view/35424), Methods and Results). The FXII-activation term is not separately added to the core: it is part_of GO:0002353, and the performed self-processing facet is already explicit.
The rough-ER assertion requires a different distinction. Mouse Hgfac immunogold labeling establishes residence, while the antibodies cannot determine activity state and the authors interpret the intracellular pool as precursor (PMID:12923239(https://journals.sagepub.com/doi/10.1177/002215540305100904), Methods, Results/Figure 4 and Discussion). Its transfer to human F12 does not establish the literal is_active_in relation. That assertion is marked over-annotated without denying conserved residence, claiming experimental inactivity or disputing the verified PAINT ancestry. T328K/R precursor and fragment locations remain non-core disease-processing contexts, with explicit consistency exceptions distinguishing them from native F12 location assertions (R-HSA-9653249, R-HSA-9655046).
Positive proteomic localization observations are retained. The insoluble tissue ECM-enriched fraction contains multiple F12 peptides after sequential extraction and compartment-marker controls (PMID:25037231(https://pmc.ncbi.nlm.nih.gov/articles/PMC4223627/), Methods and Additional file 2, sheets A/B, F12 row). Washed EPS-urine vesicle-enriched preparations likewise contain multiple F12 peptides in both study pools (PMID:23533145(https://pmc.ncbi.nlm.nih.gov/articles/PMC3773505/), Methods, Supplementary Information and Supplemental Table 2). Fraction overlap and carryover limit topology and mechanism claims, but are not demonstrated explanations for these particular target detections. Neither annotation gives F12 a structural matrix or vesicle-biogenesis function.
The retained plasminogen-regulation context also distinguishes a regulator from the enzyme acting directly on the final substrate. Activated Hageman factor fragments activate upstream prekallikrein/FXI in the inspected experiments; the fragments’ direct plasminogen-activator contribution is comparatively small (PMID:89876(https://doi.org/10.1182/blood.v54.4.850.850), selected original pages 851, 853, 855, 857 and 859). This supports contextual positive regulation without making FXII the principal direct plasminogen activator.
The standing campaign instruction explicitly retains supported generic binding as non-core when no evidence-backed specific molecular function is established. It overrides the generic-binding REMOVE default for the eight supported interaction contexts here; the unresolved ayaconin context remains unresolved. Inhibitor binding does not give FXII the inhibitor’s activity, and yeast two-hybrid partners do not establish secretion or a specialized physiological role. The campaign also prohibits adding new publication quotations in this review. Reference-only support and public original locators are retained. Complete-original availability is assessed separately from the immutable normal-cache full-text flag and from the narrower experiments actually read; no access flag is changed solely because the cache is abstract-only. Earlier dated paragraphs above remain the historical journal of previous checkpoints.
The three variant-location reviews retain their specific disease-processing explanations and non-core actions. Their structured consistency_exception fields have been removed because those fields apply to the separate action-consistency rule table, which does not define these ACCEPT/non-core or MODIFY/non-core pairs as contradictions. The older same-term action check still reports the extracellular-region and plasma-membrane differences and does not consult that field. Those two warnings remain documented alongside the campaign-supported generic-binding warnings; no warning suppression or change to the underlying biology is claimed. The earlier journal paragraph describing explicit consistency exceptions records the preceding checkpoint. Sources, reference assessments, actions and core functions are unchanged.