The authenticated human target is CNGB1 (HGNC:2151, UniProt Q14028). A genuine isolated normal fetch produced 73 assertions, four alternative products and 22 reference identities; every original object and its order is preserved. The final review adds one product-specific molecular-function assertion and no biological process. The original main had no CNGB1 review or history. Complete open-PR and alias checks cleared ownership; the GARP acronym in PR4473 referred to a different complex and was explicitly adjudicated as a homonym.
The actual Falcon deep-research attempt timed out at 90 seconds. The real perplexity-lite fallback returned HTTP401 insufficient quota. Neither attempt produced a usable report, and no manually authored provider-named file was created. This journal and the compact source index document the manual evidence work. Actual primary read ranges, failed retrieval bodies and raw database records remain in the immutable TMP evidence packet.
Human rod preparations resolve three CNGA1 subunits and one CNGB1, with beta contributing a ligand-binding domain and part of the pore. cGMP drives opening; high cAMP partially activates the rod heteromer. The structural beta sequence is NP_001288.3 residues 456–1251, with full-length functional comparisons separately reported PMID:34699778. The older electrophysiology uses human HA-tagged beta with bovine alpha in Xenopus oocytes, not an entirely human receptor PMID:24164424. Actual bovine experiments measure sodium/calcium permeability, rather than relying on calcium block as proof of permeation PMID:10352032.
The human olfactory preparation contains two CNGA2, one CNGA4 and one CNGB1 and exhibits cAMP-dependent activation plus calcium/calmodulin feedback. It also uses residues 456–1251, an engineered 796-aa sequence PMID:41131016. Original rat olfactory cloning defines a distinct 858-aa B1b product and directly tests native-like sodium/potassium conduction and gating PMID:10377344. These assays support one contributed cyclic-nucleotide-gated cation-channel core, while retaining the original ligand-specific rows.
The four exact human products remain RCNC2B/Q14028-1 (1,251 aa), RCNC2A/Q14028-2 (623 aa), GARP2/Q14028-3 (299 aa), and Q14028-4 (1,245 aa). Official FASTAs match independent VAR_SEQ reconstruction. RCNC2A is not automatically the physiological olfactory B1b, and the engineered human construct is not any complete listed product. This is an unresolved transcript-to-physiology mapping question, not a reason to delete a fetched product.
Mouse channel disruption reduces rod responses before progressive degeneration while preserving soluble GARP expression PMID:15634774. Human CNGB1 G993V segregates in the original retinitis-pigmentosa family PMID:11379879. The separate splice-variant study has no CNGB1 family DNA and does not directly assay its predicted splice outcome PMID:15557452. The old PMID:7479749 direct mutation assays concern CNGA1; its reference relationship is flagged while visual function remains supported by independent CNGB1 evidence.
A separate GARP2 core and one NEW GO:0004857 assertion use ISS, human Q14028-3, bovine Q28181-5 and PMID:16407240. Actual native bovine purification, association and enzyme assays support suppression of basal PDE6 activity, with no significant inhibition in the tested transducin- or trypsin-activated conditions. Human activity is inferred; neither equivalent kinetics nor a demonstrated reduction in physiological human dark noise is claimed.
The original human GARP clone has the same 299-aa architecture and terminal splice sequence, with the documented R100H variant relative to current Q14028-3 PMID:7590744. Bovine and human products are not sequence-identical. The official enzyme-inhibitor term fits a direct, noncovalent inhibitory interaction. Its 61 typed descendants contain no suitable PDE6-specific child, and there is no ancestor/descendant overlap with the seeded terms. Human PDE6G is an informative same-enzyme inhibitor comparator; it is not experimental evidence for human GARP2. No new dark-adaptation, noise-control or other process assertion is added.
Human beta C-terminal mapping supports ankyrin binding, with residue/deletion tests and targeting experiments PMID:19299621. GRB14 association remains a positive non-core pair. The exact MINT records map human beta to bovine GRB14, but the original beta co-immunoprecipitation plasmid species and beta-specific two-hybrid experiment are not independently sequence-authenticated. Unlike the alpha construct examined separately, no concrete beta primer contradiction was found PMID:20890309. No generic binding is removed solely for being uninformative.
The mouse olfactory maturation premise is source-specific: adult CNGB1 CaM-site/NCKX4 experiments show altered electrical responses and reduced OMP amount, while the OMP-positive neuron fraction and GAP43 do not show reduced maturation. The original acts_upstream_of qualifier is preserved; the specific developmental allocation is marked over-annotated without denying the positive signaling effects PMID:28053034.
Detection-series terms do not demand that every participating product itself capture photons or odorants. Existing light/odor detection assertions are retained as contextual non-core annotations because CNGB1 performs downstream electrical conversion. Actual rod and olfactory current data support that role; no direct photon or odorant binding is assigned [PMID:15634774; PMID:16980309]. Retinal homeostasis and broad GPCR signaling likewise remain contextual. The rat cultured-neuron bouton/calcium source is positive but abstract-limited and is not generalized to a proven human native site PMID:21809342.
Current donor records were joined literally to original source entities and terms, with species and product identities retained. The historical PAINT tree, alignment and IBD placement were not reconstructed. Target self-donors are legitimate experimental grounding. One-to-many fly mappings and unresolved worm mappings remain explicit; failed namespace requests are not negative biology. Source statuses reflect independent support for each target assertion rather than donor counts.
All seven Reactome event chains were followed through actual complexes or cargo sets to the human Q14028 reference entity. Plasma-membrane channel activity and Golgi-vesicle/ciliary cargo compartments are distinguished; being transported does not make CNGB1 a vesicle-transport catalyst. The authenticated current GO-CAM index contains no Q14028 hit; absence is not treated as justification for NEW processes.
The normal abstracts are preserved even when authentic external full text was read. This applies to PMID:15557452 (indexed author-uploaded primary), PMID:28053034 (actual PMC full HTML), and PMID:20890309 (publisher text and original supplement). The source index supplies actual URLs, hashes and read limits; it carries no primary quotations or circular supporting-text claims. The genuine full PMID:15634774 cache matches published CNGA1 PR4746 exactly. Unused incidental supplemental sources do not become asserted evidence merely because they were fetched.
Only short literal anchors in the YAML quote primary sources. No primary quote is repeated here or copied into the source index. Final normal validation, status, rendering and source applicability are recorded in the sealed handoff after checks finish.
Normal validation and rendering pass. Two intentional advisories remain: positive GRB14 generic binding is retained under the campaign policy, and the same cAMP-channel term has source-specific actions because the rod high-cAMP assay is a secondary agonist context while the other records support olfactory gating. The derived status remains DRAFT under these advisories. The final core uses ciliary membrane alone because the rod outer-segment membrane is its child; source location assertions and specific biological prose remain unchanged.
The complete assessment posted as comment 6107075128 was read, including all nine points. The review job failed after posting the assessment and did not submit a formal review. This follow-up changes four existing decisions from non-core to accepted: the human rod cAMP assay and the three sensory-detection assertions. It preserves every source object, product, prior reference identity and history record; the preceding notes remain an unchanged journal prefix.
Calmodulin and complex binding. PMID:41131016(https://doi.org/10.1038/s41467-025-64436-5), Results/Figures 5–6 and Methods, directly supports calmodulin association with the engineered human CNGB1 456–1251 protein in the olfactory channel. Beta-specific cross-links and deletion tests support the N- and C-terminal sites and inhibitory feedback. The interface map is locally low resolution and its assignment uses prediction-assisted modeling; not every contact was resolved atomically. This performed interaction is now explicit in the channel core with one short primary anchor. The earlier PMID:34699778(https://doi.org/10.1016/j.neuron.2021.10.006) supports rod assembly and gating; its selected full text does not itself report a calmodulin assay, and the later paper revises the earlier beta-helix assignment.
The complex-binding source is now resolved more precisely: the exact current rat donor carries GO:0044877 by IDA from PMID:22786723(https://doi.org/10.1126/scisignal.2003110). Its authentic author-uploaded original Methods identify rat CNGB1b/AF068572 and a 1–858 fluorescent fusion in Xenopus oocytes; Results demonstrate functional subunit combinations and partner-dependent membrane incorporation. The normal cache remains an unchanged abstract; actual external full-text access is recorded separately. This source supports a channel-complex premise, not a mapped calmodulin interface. Official ontology relations do not place GO:0005516 under GO:0044877, so the positive complex assertion is retained rather than replaced laterally. The precise calmodulin term is a child of the existing generic protein-binding term used for GRB14. Under the user's strict NEW ancestor/descendant constraint, no redundant new assertion is manufactured; the representation question is recorded explicitly. The biological interaction itself is affirmed in the core.
Channel and sensory scope. The rod experiment in PMID:34699778 measures cAMP-activated current, despite its low partial-agonist efficacy and closed captured structure. The corresponding assertion is now accepted as part of the gene-level channel capability; physiological rod cGMP and olfactory cAMP contexts remain distinct. GO:0050908 and GO:0050911 describe stimulus-to-signal event series. The channel performs their cyclic-nucleotide-to-current step, so the existing detection assertions now receive ACCEPT and replace broader perception terms in the core synthesis. This adds no new BP and does not claim direct photon or odorant binding.
The integrated rod/olfactory core retains ciliary membrane alone: photoreceptor outer-segment membrane is its official child, so listing both would duplicate location scope. Rod-specific source assertions and prose remain. GARP2's human outer-segment location is now explicitly ortholog-inferred, like its activity. PMID:16407240(https://doi.org/10.1074/jbc.M512138200), Results/Figure 4, actually measures suppression of nonactivated PDE6 catalysis; inhibition is not merely an interpretation of association. Binding is already described within that inhibitor role, so another broad enzyme-binding assertion would add little biological coverage. Basal versus activated enzyme limits and the absence of a direct human assay remain explicit.
The miscited PMID:7479749 is removed only from authored support for the visual assertion. Its machine source field, reference identity and adjudication are preserved. The two cultured-neuron assertions remain non-core because the complete PMID:21809342 abstract positively reports rat CNGB1-containing channels, synapsin colocalization and cGMP-induced calcium entry. Typical sensory expression elsewhere does not negate those experiments; full Results and human in vivo generalization remain outside the inspected scope.
The ClinGen project instruction explicitly implements the user's standing exception: retain supported generic protein binding as non-core when a verified narrower class is unavailable. That task-specific instruction takes precedence over the generic skill recommendation. It is the authority for retaining the positive GRB14 association; merged precedent is not the authority. The beta-construct and interaction-assay limitations remain unchanged.
The source index keeps its JSON format and previous literal records, adding only this follow-up's checkable source locators and dispositions. The added PMID:22786723 cache is genuine normal output. Primary anchors occur only in YAML; the new calmodulin anchor brings PMID:41131016 to 21 lexical words in total, within the 25-word source limit. No source cache or prior history was rewritten.
Final normal validation, status checking and rendering pass. The former cAMP action inconsistency is resolved; one intentional advisory remains for the campaign-authorized positive GRB14 generic-binding assertion. The ordinary derived status remains DRAFT.
The exact published head 140324d46cdd32c43d337d937380203fc563c144 and complete formal/detail feedback were authenticated before this amendment. This appendix supersedes no source observations and preserves the complete prior notes prefix.
CaM is a positive measured interaction. Actual PMID:41131016(https://doi.org/10.1038/s41467-025-64436-5), Results/Figures 5–6 and Methods, assays human CNGB1 residues 456–1251 with CNGA2 and CNGA4. Beta-specific cross-links and differential N- and C-terminal deletion tests support CaM binding and inhibitory feedback. The local interface model remains low resolution and prediction-assisted. The engineered 796-aa product is not silently mapped to a complete named natural human isoform. Its CaM contact is retained in the existing channel core and is not rejected for weaker evidence or lack of a native human assay.
The requested NEW GO:0005516 is nevertheless declined under the explicit current ancestor/descendant rule. The ankyrin row is refined, but the independently supported GRB14 row remains KEEP_AS_NON_CORE at GO:0005515; partner-specific wording does not change that retained ID. Current official is_a ancestry contains GO:0005515. The distinct GARP2 inhibitor assertion has no such ancestry overlap and concerns another product/activity, so comparing evidence strength does not remove the coverage constraint. The CaM-dependent feedback mechanism belongs within the integrated channel role; adding a third core solely to create another term slot would duplicate that role. The source-model question is made explicit. The positive rat channel-complex-binding premise remains unchanged.
UniProt evidence clauses are separated. The archived human SUBUNIT comment contains the rod-complex PMID together with an explicitly by-similarity CaM-modulation clause. Its CaM1 and CaM2 feature entries separately cite bovine Q28181 using ECO:0000250. The actual selected PMID:34699778(https://doi.org/10.1016/j.neuron.2021.10.006) construct/current Methods and Results support rod assembly and gating; its cache has no calmodulin/CaM match. That bounded inspection is not upgraded by a paragraph-level UniProt citation into an independently observed CaM experiment. The later human olfactory source supplies the positive binding evidence. Source caches and their access flags remain unchanged.
Membrane specificity. The broad GO:0016020 row now proposes GO:0060170 ciliary membrane. Actual PMID:19299621(https://doi.org/10.1126/science.1169789) tests human-beta constructs in transgenic Xenopus rods and finds the outer-segment plasma-membrane destination; its host is not human retinal tissue. Original rat ciliary fractionation/immunolabeling PMID:10377344 and mouse olfactory localization PMID:16980309 support the other sensory context. Soluble GARP2 is excluded from this membrane interpretation. The integrated core retains the common ciliary location while explicitly naming both settings. Current official GO:0042622 ancestry includes GO:0060170; existing rod-specific annotations remain and no redundant parent/child core locations or duplicated rod/olfactory activity cores are added.
Remaining optional suggestions. The miscited PMID:7479749 row now states explicitly that ACCEPT adjudicates the supported CNGB1 function while the original citation remains MISCITED and excluded from claim support. No new PDE-inhibitor term is proposed in this finite amendment: GO:0004857 accurately covers the measured direct noncovalent inhibition, with basal-versus-activated and bovine-versus-human limits already explicit. Absence of a PDE6-specific child alone does not establish the need for a new ontology class; the existing human GARP2 experimental question remains. The typed descendant set was rechecked against the earlier authenticated complete definition survey.
No source assertion, product object, qualifier, reference identity, existing NEW annotation or source quotation is removed or added. Original supporting quotations remain confined to YAML; this appendix and the index add no quotations. The campaign authorization for positive generic-binding retention remains applicable and is not replaced with a mechanical REMOVE.