The reviewed source set contains 19 GOA annotations and one inherited proposed molecular-function annotation. All 19 source objects are preserved after the standard GOA parser backfills their qualifiers and supporting entities. There are no alternative-product records in the normal UniProt source or existing review. Raw UniProt, GOA, provider research and cached publications are unchanged.
The proposed review retains 12 annotations as ACCEPT, refines two generic molecular functions, retains two as non-core, and leaves three UNDECIDED. The existing NEW proposal is narrowed to protein-transporting ATPase activity (GO:0015450); it is not accompanied by an additional ancestor annotation or a new process annotation. Nucleotide binding is refined to ATP binding and hydrolase activity to ATP hydrolysis. The specific existing ATP-hydrolysis annotation is retained alongside the coupled activity.
The conserved-function ISS proposal therefore does not assert direct human transport kinetics, a fixed ATP-per-cargo ratio, or a uniquely demonstrated human heptameric mechanism. The human identity and disease evidence and the ortholog mechanistic evidence have complementary roles.
The consulted official AmiGO snapshot was loaded on 2026-08-06. GO:0015450 is a child of protein transmembrane transporter activity (GO:0008320) and ATPase-coupled transmembrane transporter activity. It captures the coupled protein movement. The more narrowly named mitochondrial child GO:0008566 describes import through the inner-membrane translocase, which does not match this client-export mechanism; GO:0008564 describes secretory-pathway export. Neither is chosen merely from its name. The GO:0016887 comment supports representing hydrolysis together with an ATP-dependent overall activity. GO:0015450, GO:0016887.
The original IBA node placement was not independently reconstructed. Retention rests on agreement with established conserved function and no identified target-specific loss; the human target in the complex III IBA WITH/FROM is not treated as circular. No historical mapping failure mechanism is invented for the unresolved mature-complex annotation.
Later structural papers PMID:38821922 and PMID:40410623 remain research leads, not read evidence or formal reference additions. Short exact anchors are used only from the normal cached sources. The assistant's conservative per-source quotation budget is separate from the repository's mechanical reference checks.
Read-only ownership clearance found no current competing changes in the observed main/head/base scope. That snapshot does not replace fresh scope verification before publication. Canonical application, focused validation, rendering and an append-only history record remain separate steps after independent review.
The preceding proposal-stage pending statements are superseded by the applied review. Independent science review passed, and focused just validate human BCS1L passed with two advisories. At this stage, the supported generic protein-binding assertion was still retained as non-core and emitted a generic-binding advisory; the 2026-10-05 cleanup below supersedes that action under the repository's codified protein-binding policy. The unchanged Falcon report is preserved without being cited as primary experimental evidence. Rendering and the scaffolded Codex/gpt-6 EDIT history validation passed. The rendered page embeds the reviewed YAML. These are focused checks, not a claim that repository-wide validation was run. All nine protected raw, research and source files remain unchanged. Publication remains a separate step subject to fresh ownership and exact-scope checks.
This appendix supersedes the earlier unresolved mature-complex assessment. The proposal retains all 19 source assertions and the one inherited NEW molecular-function proposal, with no product changes. It changes the complex III component TAS decision from UNDECIDED to REMOVE: BCS1L supplies a separate assembly motor, whereas GO:0045275 denotes the electron-transfer complex. The conserved motor architecture in PMID:37821516(https://pubmed.ncbi.nlm.nih.gov/37821516/), human UniProt function and Reactome composition are considered together. Reactome does not explicitly call BCS1L an assembly factor, and omission from its list alone would be insufficient. The historical basis and full original TAS paper remain unresolved; this is a biological assessment, not a reconstruction of curator intent. Proposed totals are 12 ACCEPT, 2 MODIFY, 2 KEEP_AS_NON_CORE, 2 UNDECIDED, 1 REMOVE and 1 inherited NEW.
The transport choice remains GO:0015450. The definition of GO:0008566 and its EC:7.4.2.3 source describe TIM-mediated import. This is a definition-based distinction from folded Rieske export, not an inference from the word mitochondrial. GO:0008564 likewise describes secretory-pathway protein export. The official consulted AmiGO snapshot (loaded 2026-08-06) labels GO:0008320 protein transmembrane transporter activity.
The cached FlyBase model 6918f23700002487 was inspected at activity gomodel:6918f23700002487/6918f23700002635. It does assign GO:0008566 to FB:FBgn0032195, a source entity in the human IBA. The molecular_function association has no evidence block of its own; enabled_by carries ECO:0000255/PMID:34274978. The yeast UniProtKB:P32839 ISS (ECO:0000250, GO_REF:0000024) is attached to occurs_in GO:0005743, not to the MF association. This distinction corrects the feedback's evidence attribution without denying the curator's term choice. An ontology question now asks how the term definition and this usage should be reconciled. The model's cited paper was not newly read and is not added as direct human evidence.
The publisher Tamai article provides the traceable Fig. 7C and Fig. 8A-B text behind the amended localization and complex III reasons. The latter now describes respiratory-complex effects instead of implying an isolated Rieske-loading measurement. Complex I/IV actions remain UNDECIDED because participation is unresolved in the inspected experiments, not because the full text was inaccessible. At this follow-up stage, the supported LETM1 association still remained non-core; the 2026-10-05 cleanup below supersedes that policy choice.
Full-text availability and reading scope are distinguished for PMID:34800366. Its full XML cache was already present; the prior reading was abstract-only. This follow-up read its complete abstract and the Results on subtractive and spatial proteomics, but not the BCS1L supplemental measurement. For PMID:37821516, the follow-up read the complete abstract and selected architecture, translocation-context, HS-AFM and gate-mechanism paragraphs and the expression/reconstitution Methods. The mouse clone and Pichia host remain explicit in evidence notes. Neither full paper, original figures nor supplemental data are claimed as newly audited. The standalone description and core prose now describe biology; kinetic and species-specific evidence limits remain in the annotation reasoning and these notes. Existing short cached quotations are unchanged.
This is a TMP proposal pending independent science review and canonical application. Focused proposal validation is recorded separately; no new canonical files, source caches, history records or remote changes were made.
Independent science review passed for this focused follow-up (2026-10-01). The exact approved YAML and appended notes were applied after matching the canonical preimages. All 19 original source objects, the inherited transport annotation, raw source files, normal reference caches and prior history remain unchanged. The only action change rejects mature complex III membership on combined evidence for a separate assembly motor; original TAS provenance remains unresolved. Focused canonical validation, rendering and a new history record follow this application.
The follow-up passed focused canonical validation, HTML rendering and new history validation (2026-10-01). The rendered page embeds the exact reviewed YAML. At this follow-up stage, two canonical advisories remained: the supported generic LETM1 interaction was retained as non-core under the supplied action definitions, and the unchanged Falcon report was not used as annotation evidence. The earlier TMP check had only the binding advisory because it lacked the adjacent provider report. No full-repository validation is claimed. Normal reference caches, raw GOA/UniProt bytes and prior history remain unchanged.
This cleanup supersedes the earlier statements that the LETM1 GO:0005515 row was retained as non-core under the supplied ActionEnum. The ActionEnum permits KEEP_AS_NON_CORE in general, but the repository protein-binding policy treats touched GO:0005515 rows as compliant only when they use MODIFY, REMOVE, UNDECIDED or PENDING. The LETM1 co-immunoprecipitation remains supported but does not establish a more informative BCS1L molecular function, so the row is now REMOVE; that action drops the uninformative generic annotation and does not assert that the interaction is false.
The proposed tallies after this cleanup are 12 ACCEPT, 2 MODIFY, 1 KEEP_AS_NON_CORE, 2 UNDECIDED, 2 REMOVE and 1 inherited NEW. The duplicate PMID:37821516 additional_reference_ids entry was also removed from the complex III component row; that publication remains present in supported_by.
This entry supersedes the 2026-10-05 removal of the supported LETM1 interaction. The authority is the explicit user instruction in the CLINGEN_MENDELIAN project at immutable main revision f7dc8b60: retain a supported, biologically correct GO:0005515 (protein binding) annotation as KEEP_AS_NON_CORE when no evidence-backed, more specific replacement has been established. This supersedes the earlier ActionEnum-only justification and subsequent blanket removal policy for this project.
The complete normal abstract and selected PMID:18628306 publisher Results and Figure 7A-B legends were read for the exact interaction. Endogenous LETM1 co-precipitates with human BCS1L-3FLAG in HeLa cells, but not the AIF-3FLAG control; the BCS1L R155P mutant loses the association. The surrounding Figure 7D Results distinguish LETM1 complex formation from stable association with most BCS1L on native gels. This supports non-core retention of the source IPI association, without assigning purified binary affinity, stable complex membership, LETM1 translocation or a narrower BCS1L molecular function. The human BCS1L construct provenance was checked in the selected cloning Methods. Original figure pixels, complete Methods and supplements were not newly audited. The abstract-only normal cache remains unchanged; selected publisher access does not constitute a complete-paper review.
Only the LETM1 review summary/action/reason and this notes appendix change. All 19 source assertions, the inherited NEW transport annotation, other decisions, reference metadata, products-as-seeded and core function remain unchanged. The mature complex III component REMOVE and complex I/IV UNDECIDED judgments are separate biological assessments. Totals are 12 ACCEPT, 2 MODIFY, 2 KEEP_AS_NON_CORE, 2 UNDECIDED, 1 REMOVE and 1 inherited NEW. Every existing scientific quotation is unchanged; PMID:18628306 totals 16 words across its three occurrences and all per-source totals remain at most 25. No new scientific quote or cache edit is included. Campaign completion remains pending the fresh follow-up PR's review, CI and merge.