F9

Summary: Activated human FIX is a serine-type endopeptidase, allowing refinement of the broader activity.

Reason: The recovered PMID:20121197 original sections describe human N-terminally truncated FIXa inhibitor complexes and active-site interactions. Complete assay Methods were not recovered; the explicit fluorogenic procedure in the inspected footnote is for uPA and is not assigned to FIXa. Independently read PMID:2592373 directly measures FX activation by human FIX preparations, with a normal FIXa control. Together with the source-supported serine-protease mechanism, GO:0004252 is the appropriate descendant of endopeptidase activity. The coupled Xa reporter is not represented as direct FIXa cleavage of that reporter.

Proposed replacements: serine-type endopeptidase activity

Summary: The measured internal cleavage activating FX supports serine-type endopeptidase activity.

Reason: PMID:2592373 Materials and Factor IX assays use purified human normal/Nagoya FIX, human FVIII and bovine FX. Normal FIX activated by human FXIa is a positive comparator for the mutant activation experiments. Generated Xa is measured with a fluorogenic Xa substrate after the FX-activation reaction. This establishes the internal protein-cleaving function of activated FIX; it does not make FIX responsible for the downstream reporter cleavage. Refine the broad endopeptidase term to GO:0004252, keeping the pathological mutant and its experimental activators distinct from normal FIXa.

Proposed replacements: serine-type endopeptidase activity

Summary: Serine-type endopeptidase activity is well supported for activated human FIX.

Reason: The verified PMID:12444082 abstract describes human FIXa active-site mutations and amidolytic activity, but complete original Methods were not recovered. The experimental annotation is retained using independent executed human FIXa-dependent FX activation in PMID:2592373, without inventing the inaccessible study’s exact expression construct, substrate preparation or kinetic values. This does not imply that unactivated FIX or both natural products have equivalent catalytic activity.

Summary: The human FIXa inhibitor study is consistent with the independently established serine endopeptidase function.

Reason: Selected original Results and structure footnotes of PMID:20121197 concern human truncated FIXa and its catalytic binding pocket. The complete enzymatic protocol remains access-limited, and the inspected uPA assay is not misassigned to FIXa. Human FIXa-mediated FX activation in PMID:2592373 independently supports the imported catalytic annotation. No inhibitor potency, full-length construct equivalence or activity of the unactivated zymogen is inferred.

Summary: Activated human FIX retains its established serine-type endopeptidase assignment.

Reason: PMID:20121198 is verified as the benzothiophene FIXa inhibitor study, with a positive active-site/inhibition abstract and associated human FIX structure record. The complete original biochemical Methods were not recovered. Independent original human FIXa-dependent FX activation in PMID:2592373 supports the activity, so the experimental curator assertion is retained rather than rejected from an access gap. Details of the unread inhibitor assay and equivalence of truncated preparations to both natural products are not asserted.

Summary: The Nagoya study directly measures the FX-activating protease function of human FIXa preparations.

Reason: PMID:2592373 compares normal human FIX activated by FXIa with Nagoya FIX activated under defined chymotrypsin or rat mast-cell chymase conditions. The subsequent human FVIII/bovine FX reaction measures formation of active Xa. This is positive performed-step evidence for the serine endopeptidase function; the different activation routes and species of the FX substrate remain explicit. FIX being cleaved by another enzyme is not itself the evidence for FIX protease activity.

Summary: Retain inherited serine endopeptidase activity.

Reason: The exact 461-residue human leaf PTN002542616 in PTHR24278 v19 descends from the positive GO:0004252 IBD at PTN000669108. There is no path NOT or target-specific loss. Human FIXa directly hydrolyzes the peptide reporter in PMID:34502392; the normal human FIX control also generates an enzyme that activates bovine FX in PMID:2592373. Human self-WITH is valid descendant evidence. These results concern activated enzyme, not activity of every precursor or the untested shorter product.

Summary: Retain the specific serine endopeptidase mapping.

Reason: P00740 actually carries the trypsin-family, coagulation-protease and histidine-site signatures IPR001254/IPR012224/IPR018114, with current GO:0004252 mappings and EC 3.4.21.22. Human FIXa peptide hydrolysis independently supports the chemistry. The captured current ARBA00027237 conditions do not match the current target crossreferences; this version boundary does not invalidate the independently supported components of the combined source.

Summary: Serine endopeptidase activity is an established function of activated FIX.

Reason: The PMID:8632006 abstract addresses the human G12R FIX variant and its activation/function, while the complete original remains unavailable after preserved publisher-access failures. Independent original PMID:2592373 demonstrates human FIXa-dependent FX activation. Retain the valid NAS catalytic assertion with that target-specific corroboration; do not reconstruct the unread G12R Methods or generalize its phenotype to the natural deletion product.

Summary: Human FIX Gla-domain calcium coordination is directly resolved.

Reason: PMID:14722079 Methods and Tables I/III describe a fully gamma-carboxylated synthetic mature human FIX Gla1–47 peptide bound to Fab10C12. The crystal model resolves FIX residues1–45 and six calcium ions (Ca1–5 and Ca7), with explicit protein-ion contacts. This is direct calcium-binding evidence, not merely a calcium requirement in a clotting mixture. It does not establish identical site occupancy in physiological plasma, all intact products or every processing state.

Summary: The resolved human FIX Gla-domain structure directly supports calcium ion binding.

Reason: The actual PMID:14722079 Table III and structure Results resolve calcium contacts in the synthetic mature Gla-domain/Fab preparation. The crystallized peptide is not full-length natural FIX; residues46–47 were omitted from the model. Preserve that boundary while accepting the direct metal-binding function. The table’s comparative bovine prothrombin and prior NMR columns are not counted as additional human FIX crystal observations.

Summary: Calcium binding is independently established for human FIX.

Reason: The complete calcium-site evidence underlying PMID:20121197 was not recovered from its selected inhibitor-study text. Independently read PMID:14722079 directly resolves six calcium ions coordinated by the human synthetic Gla domain. This positive target-specific structure supports the imported function without pretending that calcium in an assay buffer alone proves binding, or assigning the Gla-peptide preparation to every natural product.

Summary: Human FIX has experimentally supported calcium-dependent binding and conformation.

Reason: The recovered PMID:9169594 original Results and captions on p.633 compare normal FIX with propeptide variants across calcium concentrations; p.635 discusses the altered calcium-dependent conformation. Complete Methods remain unavailable. Direct calcium coordination in the independently read human Gla-domain structure of PMID:14722079 establishes the binding function itself. Retain calcium binding without treating every conformational or phospholipid readout as a separate metal-site measurement.

Summary: Retain calcium binding with the tested domain scope.

Reason: Current target GLA/EGF calcium-related entries and ARBA00043315 support the mapping. In PMID:14722079, a chemically synthesized, gamma-carboxylated mature human FIX Gla1–47 peptide was crystallized with 10C12 Fab; residues1–45 and six calcium ions were resolved. This is direct calcium coordination in a defined domain preparation, not a universal native occupancy count or a functional test of both natural products.

Summary: Comparative FVIII-domain pulldowns support protein domain specific binding.

Reason: PMID:26157780 tests recombinant human FVIII A1, A2, A3 and C-domain preparations and identifies preferential FIX association with A2. Methods use E. coli-expressed GST domains with recombinant BeneFIX or albumin-free human plasma, followed by glutathione pulldown and Western detection. GO:0019904 captures this experimentally resolved domain specificity. The plasma preparation is not treated as proof of platelet origin, BeneFIX is not assumed activated, and the pulldown is not upgraded to a tenase activity assay. The imported FVIII-A2 product identifier remains exact.

Proposed replacements: protein domain specific binding

Summary: The resolved FIX glycan-acceptor interaction with XXYLT1 is supported but non-core.

Reason: PMID:26414444 directly resolves a human FIX EGF-repeat glycan acceptor bound to mouse XXYLT1, matching the imported mouse Q3U4G3 partner. The FIX fragment was expressed in E. coli and enzymatically modified; the XXYLT1 ectodomain was expressed in HEK293T and trimmed for crystallography. A stable purified binary complex was also demonstrated before crystallization. Thus the physical interaction is supported. Although the literal enzyme-binding definition encompasses contact with a catalytic protein, replacing generic binding with GO:0019899 would add little functional information here: FIX is the glycan-bearing substrate, while XXYLT1 performs xylose transfer. Under the explicit campaign exception for supported generic binding, retain the original assertion as non-core rather than remove a demonstrated interaction. No glycosyltransferase activity, glycosylation-process participation or regulatory effect on XXYLT1 is assigned to FIX, and this fragment is not equated with either intact natural product.

Summary: The Padua study directly isolates FIX from human blood plasma.

Reason: PMID:19846852 Methods describe affinity and ion-exchange purification of FIX from proband and control plasma. Recombinant human wild-type and R338L FIX are separately assayed in HEK293 conditioned medium. These positive native and recombinant extracellular observations support the circulating extracellular pool. The pathogenic gain-of-function variant and its mature-protein numbering are not treated as a natural splice product or proof of universal product equivalence.

Summary: Human plasma FIX is supported as an extracellular protein.

Reason: The verified PMID:3857619 abstract explicitly reports FIX sequence analyses from individual and pooled human plasma, but the complete original was not recovered: PMC PDF routes returned non-PDF challenge material and the publisher route failed. Independently read PMID:2592373 and PMID:19846852 Methods directly purify FIX from human plasma. Retain the valid extracellular assertion with this independent corroboration, without inventing the unavailable dimorphism study’s purification details.

Summary: The plasma-associated FIX pool is well established.

Reason: The PMID:8295821 abstract explicitly describes the human Strasbourg2 propeptide variant purified from patient plasma; complete original Methods/Results remain unavailable. Independent actual human plasma purification in PMID:2592373 and PMID:19846852 supports extracellular region. The propeptide-retaining disease variant is not equated with a normal processed chain or with the 423-residue natural product, and no general normal-processing conclusion is drawn from it.

Summary: Normal and propeptide-variant FIX occur in the circulating extracellular pool.

Reason: PMID:9169594’s verified abstract and recovered selected original Results concern human circulating normal/variant FIX and its calcium-sensitive conformation. Its complete purification Methods remain inaccessible. Independent human plasma FIX purification in PMID:2592373 directly corroborates extracellular region; retaining the location does not erase the distinct processing and activity of the variants or imply all products are secreted equivalently.

Summary: Retain the inherited extracellular pool.

Reason: The exact target leaf descends from the positive extracellular-region IBD at PTN000669108. Native human FIX was purified from normal plasma in PMID:2592373, independently supporting the extracellular compartment. The inheritance is not rejected because human FIX is itself among the descendant evidence, and it does not assert that all biosynthetic or alternative-product pools share the same activity.

Summary: The original Nagoya study directly purifies FIX from human plasma.

Reason: PMID:2592373 Materials and purification procedures identify normal human and patient plasma as the FIX sources, followed by immunoaffinity purification and functional comparison. This supports the extracellular circulating pool without assigning its observations to the untested 423-residue deletion product. The paper’s pathological variant and experimentally activated preparations remain distinct from the ordinary plasma zymogen.

Summary: Extracellular localization is supported for the circulating FIX protein.

Reason: The cited PMID:8632006 full original was not recovered, so its exact G12R preparation is not reconstructed. Direct human plasma purification in independently read PMID:2592373 and PMID:19846852 establishes the extracellular location. Retain this imported IDA assertion on that positive target evidence, without claiming that every transcript or processed chain has the same secretion or plasma persistence.

Summary: Retain secretion supported by the combined source.

Reason: Target IPR000294/IPR035972 and the official Secreted/SL-0243 mapping support extracellular region; normal human plasma FIX provides independent evidence. Current ARBA00027128 does not match the captured current target conditions, so its historical source identity is retained separately from the independently verified domain/subcellular components.

Summary: The human plasma-proteome compilation contains an explicit FIX target record.

Reason: PMID:14718574 Table II identifies P00740, coagulation factor IX precursor, in the literature, 2DEMS and LCMS2 source columns, but not LCMS1. The recovered Data Sources and Methods describe the human serum/plasma datasets and their sequence-based consolidation. This positive target-specific record supports extracellular region. Selected Methods and table text were recovered rather than the complete main article; the record does not identify a unique natural product, activation state or specific peptide sequence.

Summary: Retain the extracellular FIX substrate and products.

Reason: R-HSA-158333 directly places human P00740 mature FIX140886 and its activation products in extracellular region. FXIa performs the cleavage in this event; FIX is the substrate here. The location is retained without treating processing of FIX as an activity performed by FIX.

Summary: Retain the soluble extracellular FIXa pool.

Reason: R-HSA-5607023 explicitly moves native human FIXa from extracellular region to an anionic membrane. The extracellular heavy227–461 and light47–191 chains are own-compartment P00740 participants, rather than an unrelated reaction output. Soluble extracellular and membrane-associated enzyme pools are compatible.

Summary: Retain the extracellular destination of native FIX.

Reason: R-HSA-9670673 traces native P00740 from pro-FIX9631373 in ER lumen to mature FIX140886 in extracellular region. Native human plasma purification independently supports the destination. FIX is secretory cargo in this model, not the transporter or the enzyme carrying out its own maturation.

Summary: Accept the extracellular location while retaining the activation-resistant variant context.

Reason: R-HSA-9673223 contains the F9 variant set 9673225: eight member entities and one candidate entity, each explicitly derived from P00740 and assigned its own extracellular compartment. The modeled propeptide and activation-site substitutions are distinct from native FIX and the natural P00740-2 deletion product; no native F9 participant is inferred within this event. The annotation being reviewed asserts extracellular location, not failed activation. That location is consistent with independently demonstrated native human plasma FIX and the other accepted extracellular assertions, so it is retained at the same gene-level location scope. The variant-specific failure or delay of FXIa-mediated activation remains a source limitation and is not promoted to native FIX function.

Summary: Retain extracellular FIX during TF:FVIIa activation.

Reason: R-HSA-9929051 places native mature P00740 FIX and the resulting activation peptide/heavy/light chains in extracellular region. TF:FVIIa performs this activation step. The annotation is a valid location assertion and does not make FIX the catalyst of its own activation.

Summary: Retain extracellular release of native FIXa.

Reason: R-HSA-9929054 explicitly releases the native P00740 heavy/light-chain FIXa complex from TF:FVIIa into extracellular region. The own-compartment target route supports the location; it does not assign tissue-factor cofactor activity to FIX.

Summary: Retain extracellular FIX in the kallikrein activation model.

Reason: R-HSA-9936467 directly identifies native extracellular FIX140886 and its extracellular cleavage products. Plasma kallikrein is the modeled catalyst; FIX supplies the substrate. The location is supported by the target model without promoting kallikrein-dependent processing to a core action of FIX or claiming independent review of every cited assay.

Summary: Retain a biosynthetic ER-lumen pool.

Reason: Current ARBA00028691 matches the target by the captured domain/taxon and FunFam conditions. ER localization is independently consistent with native pro-FIX9631373 and gamma-carboxylation models. The ER pool is biosynthetic substrate/cargo, rather than the principal compartment of activated FIXa coagulation catalysis.

Summary: Retain ER-lumen FIX undergoing modification.

Reason: R-HSA-159803 explicitly contains P00740 pro-FIX159765/159866 in ER lumen. GGCX performs gamma-carboxylation; FIX is its substrate. PMID:39880037 independently assays human FIX-derived GGCX substrates and secreted full-length FIX processing, with engineered structural constructs distinguished from native enzyme.

Summary: Retain the ER origin of secretory transport.

Reason: R-HSA-159836 contains carboxylated P00740 pro-FIX159866 in ER lumen as the input to ER-to-Golgi transport. This is a valid transient location of the cargo, not transport activity performed by FIX or a principal site of its protease function.

Summary: Retain native pro-FIX in the secretory pathway.

Reason: The actual native P00740 input9631373 of R-HSA-9670673 is pro-FIX29–461 in ER lumen. The event contrasts this biosynthetic pool with extracellular mature FIX140886. Retain the ER location as non-core without confusing precursor processing with splice-product identity.

Summary: Retain variant retention as a disease context.

Reason: R-HSA-9670888 directly models P00740-derived variants retained in ER lumen or secreted inefficiently. The model explicitly does not represent the underlying splice events; its residue coordinates are reference coordinates and do not establish the actual length of every nonsense product. This disease-retention context is retained without equating it to native catalytic localization or P00740-2.

Summary: Retain the ER compartment while flagging variant nomenclature.

Reason: R-HSA-9673231 contains P00740-derived pro-FIX in ER lumen during defective GGCX modification. Its labels use N43L/N43Q/N43W/N46S, whereas the same entities specify arginine replacements at43/46 and current P00740 has R at both positions. The naming discrepancy is separate from the supported ER location. FIX remains the modified substrate, not the gamma-carboxylase.

Summary: Retain the biosynthetic Golgi-lumen pool.

Reason: The captured current ARBA00028705 matches target domain/taxon and FunFam conditions. Native pro-FIX159726 and mature FIX159822 in the actual Golgi models independently support this transit/maturation compartment, rather than the principal site of coagulation proteolysis.

Summary: Retain Golgi processing of the FIX propeptide.

Reason: R-HSA-159796 places pro-FIX159726, mature FIX159822 and the removed propeptide159735 in Golgi lumen. FURIN performs cleavage of the precursor. The location is retained as biosynthetic context without assigning that cleavage to FIX.

Summary: Retain the Golgi destination of FIX cargo.

Reason: R-HSA-159836 directly places transported native P00740 pro-FIX159726 in Golgi lumen. The target is secretory cargo, so this valid transient location remains distinct from the extracellular and membrane-associated protease unit.

Summary: Specify the external face of the membrane.

Reason: The native P00740 heavy158225 and light140894 chains are members of FIXa5633117 in the catalyst of R-HSA-158164. The actual event describes FIXa proteolysis with FVIIIa cofactor on negatively charged activated-platelet surfaces. This supports peripheral external-face association, not an integral transmembrane FIX protein.

Proposed replacements: external side of plasma membrane

Summary: Specify extracellular surface assembly with FVIIIa.

Reason: R-HSA-158278 directly assembles native FIXa5633117 with FVIIIa at the plasma membrane; the P00740 chains have that own compartment. Its anionic platelet-surface setting supports external-side association. FIXa is the protease and FVIIIa the cofactor, rather than two equivalent enzymes.

Proposed replacements: external side of plasma membrane

Summary: Specify the external membrane-associated FIXa pool.

Reason: R-HSA-5607023 explicitly describes calcium/Gla-dependent binding of native soluble FIXa to anionic membrane surfaces. The membrane heavy158225/light140894 target chains are distinct from the extracellular input chains. The current external-side definition includes peripheral association, matching this topology.

Proposed replacements: external side of plasma membrane

Summary: Refine the Padua-containing surface complex to the external side of the plasma membrane.

Reason: R-HSA-9668253 models FVIIIa with FIX Padua: mutant P00740 heavy-chain entity 9668263 (R384L in displayed precursor numbering) and native light-chain entity 140894 both carry their own plasma-membrane compartment. The event describes an extracellular coagulation complex on activated-platelet surfaces. Together with the independently supported peripheral Gla-dependent membrane association of native FIX in R-HSA-5607023, this supports the more precise external-face location GO:0009897, consistently with the other plasma-membrane annotations. The mutation changes catalytic behavior, not the topology asserted here. Its hyperactivity is not generalized to native FIX or the natural deletion product, and these modeled participants are not an independent intact-cell imaging assay.

Proposed replacements: external side of plasma membrane

Summary: Specify the native FIXa external-surface pool.

Reason: The disease variant in R-HSA-9668365 is FVIIIa, while its FIXa participant5633117 contains native P00740 heavy/light chains. Thus the source still directly supports native FIXa in the extracellular membrane complex. Refine to the external side without transferring the FVIII defect to F9.

Proposed replacements: external side of plasma membrane

Summary: Specify the external surface in PROS1 inhibition.

Reason: R-HSA-9669959 includes native membrane-associated FIXa in the PROS1-regulated intrinsic-tenase complex. The P00740 chains are actual target participants and retain extracellular topology during inhibition. This supports external-side location, not protease-inhibitor activity of FIX itself.

Proposed replacements: external side of plasma membrane

Summary: Refine the dysfunctional FIXa surface complexes to the external side of the plasma membrane.

Reason: R-HSA-9670874 contains native light-chain entity 140894 together with a set of thirteen mutant P00740 heavy-chain entities; all fourteen target entities have their own plasma-membrane compartment. Its summation places FIXa and FVIIIa on negatively charged membrane surfaces, and independently supported native Gla-dependent peripheral association in R-HSA-5607023 identifies the extracellular face. Refine the broad location to GO:0009897 consistently with the other plasma-membrane contexts. The failed FX-activation behavior is specific to these modeled variants; it does not negate native FIXa activity or change the location claim. Neither the variants nor their processed chains are equated with the natural P00740-2 deletion product.

Proposed replacements: external side of plasma membrane

Summary: Specify external-surface FIXa in antibody-assisted assembly.

Reason: R-HSA-9678390 binds emicizumab to native FIXa5633117, whose P00740 chains are membrane-associated. The therapeutic antibody provides a bridge; it does not change FIXa into a cofactor or integral membrane protein. The native target topology supports the external side.

Proposed replacements: external side of plasma membrane

Summary: Specify the native FIXa external-surface catalyst.

Reason: R-HSA-9678434 explicitly identifies native FIXa5633117 as an active unit within the emicizumab-assisted catalyst. FIXa performs FX activation while the antibody supports assembly. Its own membrane compartment and extracellular substrate-facing complex support the external-side refinement.

Proposed replacements: external side of plasma membrane

Summary: The performed proteolytic step is activation of the FX zymogen.

Reason: The imported inhibitor-study context in PMID:20121197 supports protease function, but its complete biochemical Methods remain unavailable. Independent original PMID:2592373 directly measures human FIXa-dependent conversion of bovine FX to active Xa with human FVIII. Current GO:0031638 is a proteolysis descendant defined as processing an inactive enzyme to its active form, matching this performed step. This refinement concerns FIXa acting on FX, not the upstream cleavage that makes FIXa; it does not assert autonomous activation by the FIX zymogen.

Proposed replacements: zymogen activation

Summary: Specify activation of a downstream zymogen.

Reason: IPR001254/IPR018114 correctly support proteolysis, but the more informative demonstrated step is FIXa-mediated activation of FX. PMID:2592373 uses human FIX with bovine FX, human FVIII and phospholipid in a coupled FXa-generation assay; R-HSA-158164 independently models native human intrinsic tenase. This is a step performed by FIXa, not the upstream cleavage of FIX.

Proposed replacements: zymogen activation

Summary: Specify the intrinsic coagulation pathway.

Reason: The positive blood-coagulation IBD at PTN000667592 lies on the exact human target path; mouse F9/MGI:88384 and human self-WITH are legitimate grounding. Current GO:0007597 explicitly includes IXa conversion of X to Xa. The independently inspected human FIXa assay and native intrinsic-tenase model support this finer pathway role, without equating the mouse engineered-human knock-in assay to purified mouse enzyme kinetics.

Proposed replacements: blood coagulation, intrinsic pathway

Summary: The supported coagulation work is the FIXa step of the intrinsic pathway.

Reason: PMID:20121197’s recovered original text places human FIXa inhibitor activity in coagulation, with complete assay details still unavailable. Actual PMID:2592373 directly demonstrates human FIXa-dependent FX activation with human FVIII. The current GO:0007597 definition explicitly includes IXa converting X to Xa before the common pathway. This more-specific term follows the executed catalytic step rather than inferring participation solely from necessity or inhibitor efficacy.

Proposed replacements: blood coagulation, intrinsic pathway

Summary: Human FIX function supports the intrinsic coagulation pathway.

Reason: PMID:2472424 Methods and Table I compare plasma FIX clotting activity and antigen in human hemophilia B families; Results identify Gla-, EGF- and protease-domain substitutions. These are positive functional/clinical observations, not a purified FX-conversion assay. Independent original PMID:2592373 establishes the performed human FIXa→FX activation step, which is explicitly included in the current intrinsic-pathway definition. Refine broad coagulation using that combined evidence without assigning every disease phenotype or natural product the same biochemical effect.

Proposed replacements: blood coagulation, intrinsic pathway

Summary: Specify intrinsic-pathway coagulation.

Reason: Current target IPR012224 and matching ARBA00028854 support blood coagulation. Human FIXa performs the downstream FX-activation step described explicitly by current GO:0007597 and by the native human intrinsic-tenase model. The human FIX/bovine FX experiment is retained with its exact species boundary.

Proposed replacements: blood coagulation, intrinsic pathway

Summary: The human FIXa-dependent FX reaction provides intrinsic-pathway specificity.

Reason: PMID:2592373 measures FX activation by normal human FIXa and experimentally activated Nagoya preparations in the presence of human FVIII, bovine FX and cephalin. The current GO:0007597 definition explicitly includes IXa converting X to Xa. This is direct participation through protease catalysis, not merely an effect of the mutant on clotting or the fact that FIX itself must be activated. Preserve the mutant-specific activation experiments without making them a normal physiological pathway.

Proposed replacements: blood coagulation, intrinsic pathway

Summary: Specify serine endopeptidase activity.

Reason: The source mouse P16294/Ensembl protein is F9, with an actual GO:0008233 IDA assertion to PMID:15178576. The inspected original uses WT-mouse, knockout and engineered human-FIX mouse plasma in human FIX-deficient-plasma clotting assays; it is not an isolated mouse peptide-cleavage measurement. Independent human FIXa peptide hydrolysis and activation of FX establish the narrower serine endopeptidase function without dismissing the positive donor experiment.

Proposed replacements: serine-type endopeptidase activity

Summary: FIX is positively detected in human colon and liver extracellular-matrix-enriched fractions.

Reason: PMID:25037231 Additional file 2 lists F9/P00740 (with additional listed accession Q5FBE1) at row214 in both the complete-MS and matrisome sheets; these are repeated views of the same data. The group has17 total unique peptides, with positive normal-colon patient2/3 and both normal-liver pools as well as tumor/metastasis samples. The original Methods use sequential depletion to enrich insoluble ECM and explicitly allow incomplete fraction separation. Retain this positive specialized location as non-core, without assigning matrix structural activity, exclusive ECM residence or a unique natural/processed product. Both raw anatomy contexts, UBERON:0001155 and UBERON:0002107, remain preserved despite their collapse to one seed row.

Summary: Retain activation of downstream zymogens.

Reason: The actual positive GO:0031638 IBD at PTN000667592 is ancestral to the exact human target, with human FIX and FX descendant evidence. FIXa performs FX activation in the inspected human-FIX assay and native human intrinsic-tenase model. The term is retained for that performed step, not merely because FIX itself is a zymogen substrate.

Summary: Activated FIX performs zymogen activation by converting FX to active Xa.

Reason: PMID:2592373’s coupled assay directly measures generation of Xa from bovine FX by human FIX preparations after defined activation, with a normal human FIXa control and human FVIII. The downstream Xa reporter quantifies the product of that conversion. This fits the current definition of zymogen activation; the separate upstream proteases that cleave FIX are not credited as activity performed by FIX.

Summary: Retain performed zymogen activation.

Reason: Current ARBA00087425 matches the target FunFam combination and predicts zymogen activation. Independent target evidence identifies the performer: activated FIX cleaves FX in the coupled assay, with human FIX/bovine FX explicitly distinguished, and the native human Reactome catalyst is FIXa. This is not an inference from the biosynthetic activation of FIX itself.

Summary: Directly resolved calcium contacts justify refinement of generic metal binding.

Reason: PMID:14722079 resolves six calcium ions coordinated by the synthetic mature human FIX Gla domain in a Fab complex. Calcium ion binding is a current descendant of metal ion binding and specifies the actual metal tested in this source. This does not assert that calcium is the only ion FIX can interact with or infer physiological occupancy from the crystallization mixture; the precise synthetic-domain and structural-model boundaries are retained.

Proposed replacements: calcium ion binding

Summary: FIX is positively identified in the study's EPS-urine exosome-enriched preparations.

Reason: PMID:23533145 Supplemental Table2, row564, lists sole accession P00740/F9 with3 and3 unique peptides in the two PCa-pool analyses and0/0 in the corresponding non-cancer analyses. The original Methods describe EPS-urine pools from12 low-grade PCa and12 biopsy-negative patients, differential ultracentrifugation with a DTT/sucrose wash, duplicate MS and MaxQuant identification at FDR below1%. Retain the positive fraction-associated detection as non-core. The table's light/heavy-chain names do not identify the processed state or natural product, and neither vesicle topology nor a FIX-specific exosomal function is established. Non-detection in the two control analyses does not establish universal absence.