The review contains 28 current original contexts, 35 manually assessed references and one carrier core. The published review passed native validation with one unresolved protein-maturation context. This follow-up adds an explicit proposed carrier function and corrects related citations; affected native checks are complete. The full source inventory and current quote hashes are in GCSH-source-evidence.yaml; the reference assessments distinguish original reading from cache availability.
Every original identifier, date, qualifier, supporting entity and source is retained in the compact context table. There are 28 forecast keys, no collisions, no negated contexts and no annotation extensions. All 28 actual native indices are now joined to their original contexts. The old review has 23 rows, including one authored NEW, and 30 quote associations. Its complete immutable Git object, a public commit URL and every quote pointer are retained in the evidence companion. The original 4,209-byte notes prefix is unchanged; no prior history records were present. Missing historical qualifiers are unknown, not asserted empty values.
The retired authored GO:0031405 row remains fully represented in that archive. It is not an incoming GOA row. No substitute NEW is proposed. The core now names proposed lipoyl-dependent carrier activity under GO:0140104. The existing molecular-carrier term is directly under molecular_function and has a covalent phosphopantetheine-carrier child. The PomBase model assigns it to fission-yeast gcv3; the earlier four-target query was not an exhaustive survey. Yeast GCV3 and bacterial sucB/dlaT IDA lipoic-acid-binding conventions are disclosed in the review alongside the covalent-chemistry distinction. The module H-role and Arabidopsis GDCH references are updated together.
The principal functional original is PMID:36190515(https://pmc.ncbi.nlm.nih.gov/articles/PMC9990993/). Selected Introduction, Results, Discussion, Methods and captions were read; the assessment names the distinct recombinant and cellular experiments and their limits. Full availability does not mean all figures, supplementary files or raw data were inspected.
The 1991 human H-protein study was read from the original scanned main text on pages 351–360. Page 361 was not included in that visual reading. Its cached publication is abstract-only, while the original-main-text access succeeded. The optional scan excerpt is not used as a cache-validated quotation. PMID:2025283 remains abstract-only. The newly available PMID:33890291 original main Methods/Results and Discussion were read from the XML-derived cache. Exome findings were confirmed by Sanger sequencing and segregation; the shortened translation product was predicted, and functional experiments were not performed because further patient samples were unavailable. Figures and supplementary files were not inspected. This is separate from an established natural alternative product.
For MitoCoP, the authentic supplemental Table S1A target is Excel row 403, protein group 4249. Selected experimental methods and the exact target row were read. The accessions are grouped, membrane association is Ambiguous, and the matrix column is inherited annotation. These limits are retained in the annotation reason and reference assessment.
The original HuRI study table and 22 positive IntAct records support the eight current interaction contexts across two studies. Seven HuRI pairs have 21 assay records. These records are not claims of independent physiological validation. The exact target ORF953, partner suffixes and MAGEA11 historical accession/ENSG crosswalk are retained; no modern clone-sequence equivalence is inferred. Partner classes use the separately cited official UniProt records. Selected supplementary methods from PMID:16189514(https://pubmed.ncbi.nlm.nih.gov/16189514/) do not turn the global co-affinity verification rate into target-specific validation.
The actual PTHR11715 book contains the exact 173-residue target at AN32/PTN002482330, below AN2/PTN001711613. Literal tree, alignment, AN/PTN map and IBD joins support all three incoming IBA assertions. No negative record was retained on the target lineage. The original book files and their public URLs remain pinned; private phylogeny retrieval is separate from any normal-cache acquisition.
All seven incoming Reactome events have explicit target chains to human P23434, mature residues 49–173, in mitochondrial matrix. Their catalysts and other participants' compartments remain distinct from GCSH. Actual event and pathway URLs can be resolved from the stable IDs in the evidence table; two pathway bibliography records were also read. The LIPT1/LIAS/LIPT2 chemistry is not attributed to GCSH simply because it is their substrate.
The direct canonical FASTA and official processed-chain feature establish one natural precursor. A processed protein, an engineered mature reagent and an unexplained mutant lower-mass species are different claims; none supplies an invented splice product.
The qualified normal-source plan is 25 paths: 13 publications, nine Reactome records and three existing family files. The actual native inventory has 22 unchanged main objects and three genuine missing-only acquisitions: PMID:2025283, PMID:33890291 and R-HSA-9857492. No source remains absent. Current official protein/partner records and private source-reading inputs have separate provenance.
Five short current excerpts occur once each in the review. Their aggregate totals are 11 words from PMID:36190515, 15 from P23434, five from PMID:1671321 and ten from PMID:3348809. The two UniProt excerpts share one source budget. Historical quote locators are archival, not current supporting-text placements. Every current offset is a UTF-8 byte offset into an identified actual repository source. Native paths, hashes, titles, access statements and offsets are reconciled together; the prior private coordinates remain separately recorded. A genuine EDIT history and private render accompany the published revision's completed checks. The two UniProt quotations now use the exact local file: reference, allowing the normal verbatim, UniProt-source and record-section checks to inspect them. The amendment passed default and strict GCSH validation, direct local-file excerpt/section checks, GDCH validation, the module schema and term checks, three history checks and private renders. GCSH has no annotation advisories; GDCH retains its existing matrix-coverage advisory. The module reports four member-index coverage warnings and one carrier-function coverage warning because the human core uses a proposed term. These coverage warnings do not establish biological absence.