The current GO:0009055 definition includes redox transformations during enzymatic reactions; AprA performs the FAD-dependent reduction of APS. QuickGO GO:0009055, checked 2026-09-20, therefore supports ACCEPT rather than restricting electron transfer to AprB/Fe-S carriers.
All seven rows were re-reviewed. Retain the succinate-dehydrogenase rejection because APS-specific reaction and domain evidence contradict that activity, not because every SDH/FRD must use covalent FAD. Existing OpenScientist and Falcon adjudications were read and incorporated; no duplicate run requested. Falcon proposes GO:0033748, which is hydrogenase (acceptor) activity, not APS reductase. This identifier is not adopted. Preserve the existing GO:0009973. The plasma-membrane refinement and sulfate-respiration refinement remain reasonable, but direct AprAB interaction with membrane QmoABC means soluble does not itself prove complete absence of peripheral membrane association; a second opinion on GO:0005886 would benefit from direct membrane-fractionation evidence.
Coordinator cross-check: GO:0005886 is now UNDECIDED pending the focused localization report. The evidence does not justify a confident wrong-location claim from soluble topology alone.
OpenScientist job 81dccdeb-fe21-441f-ac3c-3c52ef5efa6e recovered direct membrane-preparation and reciprocal Qmo/Apr purification evidence in the exact Hildenborough strain from PMID:23842468(https://pubmed.ncbi.nlm.nih.gov/23842468/). The fetched cache and publisher abstract explicitly describe these experiments; the cache remains abstract-only. This is evidence for a membrane-associated pool, beyond merely being a soluble electron-transfer partner.
GO:0005886 includes associated proteins; GO:0031234 reaches it through part_of/is_a relations. Live QuickGO definitions and ancestry are preserved in projects/TREEGRAFTER/rereview-2026-09-20/apra-membrane-term-check.json. The report accepts a peripheral pool while rejecting the broad plasma-membrane term; that distinction is unsupported by GO. Restore the original row to ACCEPT and add the membrane pool to the biological description/core locations. Do not convert its original IEA into experimental evidence and do not assert exclusive localization. The report claims a sequence-analysis CSV but delivers only PDF/HTML; its numerical results are not independently adopted. The succinate-dehydrogenase rejection remains supported by catalytic chemistry, without claiming the exact grafting error has been reconstructed.
The live GO:0009055 definition reads: "A molecular function representing the directed movement of electrons from one molecular entity to another, typically mediated by electron carriers or acceptors, resulting in the transfer of energy and/or the reduction-oxidation (redox) transformation of chemical species. This activity is fundamental to various biological processes, including cellular respiration and photosynthesis, as well as numerous enzymatic reactions involved in metabolic pathways."
This expressly includes enzymatic redox reactions. AprA supplies FAD-to-APS electron transfer downstream of the AprB relay, so restricting the term to carrier subunits would add a limitation absent from the definition. The exact response is now included in apra-membrane-term-check.json, shipped with this gene batch rather than only the later tracking PR. The broad oxidoreductase parent also remains a core catalytic description.