All 127 machine-supplied assertions and four alternative products are preserved. Root reviewed the independent all-row annotation consultation and subsequent core synthesis. Decisions are 74 ACCEPT, 43 KEEP_AS_NON_CORE, eight UNDECIDED and two MARK_AS_OVER_ANNOTATED. One sodium/potassium pump core is retained and no NEW annotation is proposed. The normal auxiliary sources were imported unchanged before final authorship; existing alternate caches were preserved.
ATP1A1 performs the ATP-driven sodium-export/potassium-import step. Human alpha/beta oocyte assays and the human alpha1/beta1/FXYD2 structures support the catalytic core, in contrast to merely being necessary for a downstream outcome [PMID:10636900; PMID:35803952]. The source assertions retain their original evidence codes, qualifiers, partners and species boundaries. PAINT target-self evidence is legitimate grounding, not circularity. No uninspected tree placement or donor mechanism is invented.
The two electrical-coupling assertions are too specific. GO:0086064 describes current between adjacent cytoplasms through intercellular protein channels. PMID:19683723 measures mouse beta1 deletion, secondary alpha1 loss, pump activity and calcium/contractility phenotypes; it does not establish that step. This conclusion follows the full Methods/Results/Discussion reading and the official term definition, rather than a title-based claim that the paper omitted alpha1. Existing contraction, relaxation and repolarization roles remain non-core physiology.
Generic partner binding is retained as non-core when a correct target association is supported independently. This follows the user-supplied ActionEnum rather than declaring correct interactions false for lack of specificity. The porcine lens LCK/phosphorylation study leaves its human IPI association unresolved. The exosome, raft, postsynaptic and sperm-specific gaps remain explicit; no alpha4 localization or motility result is transferred to alpha1. No ligase, calcium-transporter, exocytic-machinery or additional signaling core is inferred from association or substrate status.
The existing proton NAS assertion is non-core: the inspected experiments use Xenopus alpha1/beta3, and proton entry is not obligatory for Na/K exchange PMID:24688018. The existing phosphatase transfer is also non-core; an original mouse-alpha1 fragment assay supports p-nitrophenylphosphate hydrolysis, not a demonstrated human physiological phosphatase core. See the primary-source link below. Official GO and ChEBI classify the measured ouabain interaction within steroid-hormone binding, so no broader replacement is needed for PMID:14742675.
The production GO-CAM645d887900001323 already distinguishes the ATP1A1 pump from the ZNRF2 ligase acting on it. PMID:22797923 and its publisher note PMID:35267011 were examined with the independent original-paper assessment. The correction concerns duplicated phosphorylation-control blots, unavailable original LICOR and oversaturated alternate controls. It is not a retraction and does not establish that all original panels were independently verified.
Falcon and the configured Perplexity fallback could not start: after relocating the tool directory into the permitted temporary area, the pinned client was unavailable in the offline cache. No provider report was produced or fabricated. The primary-source research below and the root reading notes are manual research.
The following scopes distinguish each normal cache from original material read elsewhere. Whole-paper, figure-pixel or supplement inspection is claimed only when explicitly stated.
All nine human alpha/beta combinations were expressed in Xenopus oocytes. Complete normal abstract and selected original author-hosted Introduction/Methods were read, including pulse-chase, assembly immunoprecipitation and voltage-clamp methods. Human proteins in a frog host are direct human-protein evidence. Complete Results, supplementary data and figure pixels were not read.
Complete normal abstract explicitly includes human alpha1 with beta-subunit assembly and maturation experiments, alongside nongastric H/K-ATPase. The title does not exclude ATP1A1. Full Methods and construct-specific beta2 pair panels were not independently read.
Complete normal abstract reports alpha1 apical and lateral localization in human prostate. Renal basolateral localization does not impose a universal polarity rule. Full localization images were not inspected.
Complete normal abstract describes ouabain/cardiac-glycoside affinity and kinetics for human alpha isoforms in membranes. This supports inhibitor-response context; it is not a direct demonstration of every downstream endocrine signaling process.
Complete normal abstract read; human B-cell exosome proteomics. The exact ATP1A1 target identification and purification/enrichment evidence were not inspected.
Complete normal abstract read: human alpha1/alpha2 and chimeric/mutant constructs in Xenopus oocytes, assayed for ouabain association/dissociation. Official ChEBI defines ouabain as a steroid hormone, so the binding class is appropriate. Whole Methods and figure pixels were not read.
Complete normal abstract plus selected author-uploaded original Methods/Results and captions read. Porcine lens membranes were exposed to commercial Src-family kinases. Alpha1 immunoprecipitation followed by phosphotyrosine Western measures substrate phosphorylation, not LCK-alpha1 co-immunoprecipitation. Kinase donor species was unresolved; a human Jurkat expression control does not establish the human target pair. Lck did not significantly change pump activity. Root independently read complete Methods/Results and confirmed this boundary.
Complete normal abstract distinguishes TCTP-pump association from SNX6 antagonism; SNX6 is not directly shown binding the pump domains. TPT1 association is also independently curated in the immutable ATP1A1 UniProt entry. Full construct-specific pair panels were not read.
Complete normal abstract is a human tagged-protein localization screen. Exact ATP1A1 screen image was not inspected. Plasma-membrane localization is independently established directly, so its existing HDA assertion can be retained with curator deference.
Complete normal abstract includes human alpha1 and alpha4 in sperm and recombinant alpha1 as comparator, despite the alpha4 title. It does not resolve the exact alpha1 flagellar localization experiment. No alpha4 motility result was transferred to alpha1.
Complete normal abstract describes melanosome proteomics. The immutable ATP1A1 UniProt record explicitly attributes target detection in stages I-IV to this PMID; the original target mass-spectrometry table was not independently read.
Complete normal abstract describes NDRG1 interactome work. The exact target table was not inspected; the immutable ATP1A1 UniProt interaction section independently records NDRG1. This corroborates association but does not establish a purified binary interface.
Complete normal abstract explicitly includes human alpha1 and alpha2 paired with porcine His-tagged beta1 in Pichia, purified as active complexes. Human FXYD1 association is distinct from alpha2-specific instability. The title is not target misattribution.
Complete normal abstract and selected cached Methods/Results paragraphs read, notably characters 44817-45745 describing reciprocal alpha1/Naked2 co-IP in transfected HEK293 cells with CD147 negative control. MDCK vesicle cargo discovery is separate from human-cell co-IP. Complete construct sequence provenance and figure pixels were not checked; association does not make ATP1A1 exocytic machinery.
Complete normal abstract directly measures endogenous alpha1 and beta-subunit surface residence in HeLa cells. Alpha and beta degradation differ. Full figures/Methods were not inspected.
Complete normal abstract and selected body paragraph on parotid exosome proteins read. The exact ATP1A1 table row or peptide evidence was not inspected despite available HTML body text.
Complete normal abstract distinguishes rabbit ventricular myocytes, alpha1/beta1 constructs in Xenopus and pig-kidney enzyme. The expressed construct species is not stated in the abstract. Beta1 glutathionylation must not become alpha1 modification. Core pump assertions are retained through curator deference plus independent direct human assays; direct human repolarization was not established in this reading.
Complete normal abstract plus selected original cached Results (39269-43700) and Discussion (70904-74100) independently read. Mouse ventricular beta1 deletion decreases alpha1 protein and alters contractility; root additionally read complete Methods/Results/Discussion. These are excitation-contraction and membrane-pump readouts, not transfer of current between adjacent cytoplasms through intercellular channels. No figure-pixel or full cited-alpha1-paper audit is claimed.
Complete normal abstract directly describes human alpha/beta isoforms expressed in yeast and localization in isolated human atrial cardiomyocytes. Alpha1/beta1 occurs in T tubules. Whole original Methods and figure pixels were not inspected.
Complete normal abstract is Dahl-rat alpha1 allele/rubidium-influx work. It lists a 1991 Science erratum whose separate identity/content remains unresolved. Broad pump function is independently human-supported; this review does not endorse the specific historical variant mechanism or infer retraction.
Complete normal abstract read: B-cell exosome proteome and MHC-II-associated subset. Exact ATP1A1 target table, peptides and enrichment were not inspected.
Complete normal abstract and selected Introduction read. Purified recombinant human alpha1/beta1 complexes and separately rat alpha1 expressed in HeLa were studied. FXYD stabilization must not be relabeled alpha1 chaperone activity. Beta1 is an established complex partner.
Complete normal abstract read: TCTP-dependent Src release/signaling in human breast epithelial cells, with pump association described. Immutable UniProt also corroborates TPT1 binding. No new intrinsic kinase activity or broad oncogenic process is assigned to ATP1A1.
Complete normal abstract and selected original cached target Results (13100-15450) read: human POST affinity purification from HEK293, ATP1A1 NP_000692 identification, endogenous co-immunoprecipitation in human cells and POST-dependent STIM1 association. Functional calcium-pump readouts concern PMCA, not ATP1A1 calcium transport.
Complete normal abstract read and bounded independent AMER original-paper assessment incorporated: human ATP1A1 full-length/fragments, co-IP and ubiquitination substrate relationship. Root GO-CAM and complete local model independently read. Correction PMID35267011 concerns phosphorylation-control blots; normal correction metadata plus primary-notice scope are preserved, not treated as retraction or proof that every panel is unaffected.
Complete normal review abstract read: FXYD1 coordinates Na/K-ATPase and NCX regulation during cardiac stress. ATP1A1 supplies an ion gradient used for calcium handling. This is tissue-level physiological context, not an independent pump core or direct intercellular channel assay.
Complete normal abstract independently read; root complete Methods and selected Results/Discussion assessment incorporated. Xenopus alpha1/beta3 constructs in oocytes show ouabain-sensitive proton inward current, including near-physiological external ions. It is a back-step, not an obligatory H stoichiometry. No human construct or established physiological human proton-pump role is claimed.
Complete normal abstract read: human CSF extracellular-vesicle proteomics. Exact ATP1A1 identification/enrichment was not inspected; assay availability remains unresolved at target level.
Complete normal abstract and selected cached Results (6500-11000) read. Human kidney alpha1 co-IP with beta1/AE1 is clear; purified direct binding and reported affinity concern AE1-beta1. Neither that affinity nor a direct alpha1-AE1 interface is transferred to ATP1A1.
Complete normal abstract and selected target Results (35402-37164) read. GFP alpha1 and alpha3 sequences are Xenopus; alpha4 is human. Alpha1 inner-segment localization is ortholog evidence, not human protein in a frog. Original ISS provenance is preserved.
Complete normal abstract read; exact ATP1A1-beta1 crosslink cell was not inspected in the available HTML body. Independent human alpha1/beta1 assembly and structural evidence corroborate the pair; the histone-focused title is not evidence against it.
Complete normal abstract explicitly reports RNF183 association with alpha1 and beta1, beta1 ubiquitination and transport of both subunits from plasma membrane to lysosomes. Exact construct species and figure panels were not independently resolved. Lysosomal location is degradation context, not a lysosomal sodium-pump core.
Complete normal abstract read: BioPlex affinity-purification networks in human 293T/HCT116. Exact ATP1A1-beta1 table was not inspected. The established structural pair can be retained with source-specific curator deference.
Complete normal abstract read: endogenous human tagging/localization and affinity proteomics. Exact beta1/NDRG1 pair cells were not inspected. Independent structural beta1 and curated NDRG1 evidence support retaining these associations without claiming target-table verification.
Complete normal abstract and selected original Results/Introduction text read, including human alpha1/beta1/FXYD2 three-state structures. Root separately read construct/purification Methods: human HEK293F expression, affinity-tagged alpha1/beta1 plus FXYD2. This is purified-complex structure, not a newly claimed reconstituted flux measurement.
Complete available normal abstract read; publisher/PubMed truncates it at 250 words. Human soleus fractionation and immunocytochemistry directly localize alpha1 mainly to surface membrane. No whole-paper or unseen intracellular alpha1 compartment claim.
Resolve the exact target observations for the eight undecided rows. Human pump assays should separate normal coupled transport from variant-induced leaks and downstream calcium/magnesium phenotypes. The disease-oriented preseed reading distinguishes human-protein experiments from rat constructs modeling human variants; it does not support a single mechanism for every ATP1A1 allele. Validation and rendering results will be appended after they actually run.
A later agent inadvertently repeated the provider command after the earlier closure. That additional attempt also produced no report and returned an API-key diagnostic; it is preserved as an additional failed attempt, not described as the first. The accompanying normal fetch-gene-pmids call completed without new files and preserved all 41 protected files. No further provider retry is planned.
Focused validation passed with 14 advisories for supported generic-binding rows retained as non-core under the user action definitions. There were no schema, term or reference errors in that command. Rendering passed. The review remains DRAFT, and no new repository-wide validation pass is claimed.
The biological model remains one ATP-driven sodium-export/potassium-import pump. All 127 seeded source objects, four alternative products and the original core are preserved. The 128 archived GOA rows include one pair differing only in date; 127 distinct seeded assertions are reviewed. The 34 catch-all reasons now explain the actual ion, reaction or process and each source type. Remaining summaries distinguish activity, transport and cellular location instead of conflating them.
Ouabain ligand classification is confirmed directly: CHEBI:472805 explicitly has an is_a relationship to steroid hormone CHEBI:26764 as well as cardenolide glycoside. The non-core binding assertion from human alpha1/alpha2 constructs therefore remains supported PMID:14742675. This classification does not establish a cellular signaling response.
The indexed original PMID:11546672 publisher text was inspected for membrane preparation, immunoblotting, equilibrium binding, association/dissociation kinetics and relevant Results/Discussion. These are membrane assays. A cell-state endpoint was not verified, so both GO:0071383 and GO:1903416 are proposed UNDECIDED. This supersedes the earlier binding-based retention rationale for these two rows without asserting that ATP1A1 cannot mediate ouabain responses. Complete cached and official identity/abstract evidence remains available; original figure pixels, supplements and every full-paper section were not read, and the normal cache stays abstract-only.
Existing broad annotations are retained when correct at their source scope; the precise child annotations already present carry the narrower function. No duplicate child replacements are manufactured merely to increase the MODIFY count. The general ZNRF-associated complex is retained without guessing a more specific GO complex. Supported generic associations remain non-core under the user's ActionEnum, and the unresolved LCK pair remains UNDECIDED. No binding-policy REMOVE, NEW annotation or second core is introduced.
This prospective response changes only the two response-row actions (74 ACCEPT, 41 KEEP_AS_NON_CORE, 10 UNDECIDED and two MARK_AS_OVER_ANNOTATED). Earlier validation and publication statements above describe the initial version. Canonical application, validation, history and publication of this proposal have not occurred.
Root applied the reviewed proposal after checking all changed reasons, reference fields and 81 summary edits. Focused validation passed on 2026-09-29 with the same 14 generic-binding advisories; source objects, four alternative products and the single pump core are unchanged. The two response claims remain UNDECIDED at the inspected evidence scope. No new repository-wide validation pass is claimed.
Rows 110 and 118 are retained as non-core following a separate primary-paper check.
PMID11546672 remains a membrane ligand-binding source; its inspected experiments do
not themselves establish an intact-cell response. The additional evidence is the
original PMID10636900: human alpha1/beta1 expressed in Xenopus oocytes, pump-current
Methods, voltage-dependence Results, and Figure 6 caption explicitly compare currents
before and after 100 micromolar ouabain and subtract endogenous currents. This is a
functional change in cellular activity, which fits the existing steroid-stimulus and
glycoside-response terms. It is not a claim of glycoside metabolism, general endocrine
signaling, or a new core function. The source GOA reference/evidence fields are unchanged;
PMID10636900 is additional corroboration, not a reassignment of what PMID11546672 assayed.
Original paper: https://www.researchgate.net/publication/12677380_Transport_and_Pharmacological_Properties_of_Nine_Different_Human_NaK-ATPase_Isozymes
Read scope: author-uploaded original Methods (especially indexed lines 1699–1722),
matched Results (3010–3090), and Figure 6 caption (4488–4544). Complete-paper, supplementary
data, and figure-pixel inspection are not claimed. The normal PMID10636900 cache remains
abstract-only; these external Methods are not presented as literal cache quotations.
Definitions checked: https://amigo.geneontology.org/amigo/term/GO:0071383 and
https://amigo.geneontology.org/amigo/term/GO:1903416. Both allow a functional change in
activity as the response endpoint; neither requires a transcriptional response.
All 127 source assertions, four alternative products, one core, other review decisions,
existing reference findings, and availability flags are preserved.
Focused validation after this follow-up passed with 14 existing generic-binding
advisories; HTML rendering and the new history record validation passed. The
scaffolded history filename retains its generated actor token; its actor metadata
correctly records Codex. No new repository-wide validation pass is claimed.
Seven generic binding rows now map to GO:0046982 protein heterodimerization activity: the beta2 interaction (PMID:11027149) and six beta1 interactions. The official molecular-function definition describes association with a nonidentical protein to form a heterodimer. This is a specific assembly function, not a substitution of molecular function by cellular-complex membership and not a new annotation added alongside the original.
Root reread the complete normal PMID:11027149 abstract, which explicitly includes human alpha1/beta combinations in Xenopus, alongside the nongastric constructs. For beta1, PMID:35803952(https://pubmed.ncbi.nlm.nih.gov/35803952/) provides direct human alpha1/beta1/FXYD2 structural context. The cached first structural-determination Results paragraph and expression/purification Methods were reread: tagged human subunits are coexpressed in HEK293F, affinity-purified and separated chromatographically before structural analysis. No figure pixels or new stoichiometry analysis was performed. The beta-pair refinement concerns the alpha/beta association inside that accessory-containing complex. The uninspected BioPlex, OpenCell and crosslink target entries remain explicitly uninspected, with their original IPI provenance retained and the independent structure cited separately.
The ChEBI ouabain outgoing relationships, rechecked on 2026-09-29, explicitly contain CHEBI:472805 is_a CHEBI:26764 (steroid hormone) and separately CHEBI:472805 is_a CHEBI:38092 (cardenolide glycoside). These are ontology assertions, not evidence for endogenous endocrine physiology. The existing non-core response rationale now names the exact axiom and retains the human-protein/Xenopus-host boundary. The PMID:10636900 reference assessment now also records the previously inspected selected Results/Figure 6 caption. Its finding uses a short cached abstract anchor for the isoform-dependent ouabain effect; the abstract-only cache is unchanged.
Seven other generic associations retain KEEP_AS_NON_CORE and the LCK pair retains UNDECIDED under the evidence judgments already recorded. This is Codex's curation decision under the supplied action definitions, not a claimed maintainer sign-off or an assertion that the automated binding policy has changed. The remaining policy disagreement is explicit; no unsupported REMOVE is introduced to obtain approval. Earlier notes describing binding warnings as pre-existing refer to the previous version of this review, not to an exemption for a previously merged repository record. The source assertions, four products and single pump core remain unchanged.
Focused validation after these refinements passed with seven generic-binding advisories and no blocking error. The seven supported associations deliberately retain KEEP_AS_NON_CORE under the supplied action definitions; the unresolved LCK pair remains UNDECIDED. Rendering and the matching generated EDIT history validation passed. Status remains DRAFT. No new repository-wide validation success, current-head approval or merge is asserted.
The remaining seven generic-binding decisions implement the user's explicit instruction for this ClinGen task: retain a supported, biologically correct interaction as KEEP_AS_NON_CORE when no defensible, more specific replacement has been established, rather than remove it solely for lacking functional information. This is a deliberate task-specific departure from the annotation-reviewer skill's informational-exclusion recommendation. That general recommendation can remove an uninformative annotation without declaring the interaction false; the distinction is acknowledged here. Earlier references to the supplied action definitions describe the criterion applied in this task, not an exemption that Codex granted itself, a changed repository-wide policy, or an observed maintainer sign-off. No external approval comment is claimed.
The seven retained rows remain TPT1/TCTP twice, NDRG1 twice, NKD2, POST/SLC35G1 (Q2M3R5), and RNF183. Their existing source-specific reasons and access limits govern what is retained. In particular, the NDRG1 target-table entries were not independently inspected; curated-source deference and the separate UniProt corroboration must not be described as primary-table verification. Retention does not assign a partner's activity to ATP1A1 or establish direct binary contact for every association. The unresolved LCK pair remains UNDECIDED; this task instruction does not turn missing or unadjudicated evidence into positive support.
The seven evidence-backed alpha/beta refinements to GO:0046982 and all other annotation decisions remain unchanged. All 127 source and review objects, four alternative products, the single sodium/potassium-pump core, literal evidence excerpts, and cached sources are preserved. This addendum clarifies the authority and exclusion criterion for the existing decisions; it adds no primary-source reading or biological claim. Prior validation results remain historical observations, and this clarification does not claim reviewer approval or completion of the PR.